Publications
2026
2025
2024
2023
2022
2021
2020
2019
2018
2017
2016
2015
2014
2012
Conference
UACP: Uncertainty-Aware Channel Prediction in MIMO Networks
Mohammad Abdi, Arman Elyasi, Nathaniel D. Bastian, Francesca Meneghello, Francesco Restuccia
Proceedings of the Twenty-seventh International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc 2026)
The key issue with existing compression strategies for multiple-input multiple-output (MIMO) channel state information (CSI) feedback is that they rely on strong assumptions about channel statistics. Ultimately, this may lead to suboptimal CSI compression and, in turn, inaccurate MIMO precoding in case of model uncertainty and/or model misspecification. In this paper, we propose UACP, the first uncertainty-aware CSI prediction and compression strategy for MIMO systems. UACP is rooted in imprecise probability theory and aims at dynamically selecting the lowest number of subcarriers for channel reporting to guarantee a low CSI prediction error at the transmitter for accurate precoding. UACP uses evidential regression to obtain an estimate of the CSI prediction error in addition to the prediction itself. The key intuition is that such an error is linked with epistemic and aleatoric uncertainty of the prediction process. These two components are effectively disentangled by UACP and used as a proxy to decide the compression level, i.e., the number of subcarriers to be fed back. The results show that UACP reduces feedback overhead by 92.7% and improves throughput by 21.3% relative to Wi-Fi (Ng=4), while incurring only a 0.33% increase in error vector magnitude (EVM). UACP falls back to full feedback and improves bit error rate (BER) by 6.3% under distribution shift.
@inproceedings{abdi2026uacp,
title={{UACP: Uncertainty-Aware Channel Prediction in MIMO Networks}},
author={Abdi, Mohammad and Elyasi, Arman and Bastian, Nathaniel D. and Meneghello, Francesca and Restuccia, Francesco},
booktitle={Proceedings of the Twenty-seventh International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing},
year={2026}
}
2026
Conference
Semantic Multiplexing
Mohammad Abdi, Francesca Meneghello, and Francesco Restuccia
Proceedings of the Twenty-seventh International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc 2026)
Mobile devices require the parallel execution of several processing tasks at the edge to reduce their computing load. However, existing communication systems only support parallel transmissions at the bit level. In this paper, we introduce the new concept of Semantic Multiplexing, which shifts stream multiplexing from bits to tasks. By merging multiple task inputs into a single semantic representation, Semantic Multiplexing can multiplex more tasks than the number of physical streams by adding a (semantic) degree of freedom to the communication system. Semantic Multiplexing has been prototyped in an experimental testbed with a Jetson Orin Nano and millimeter-wave software-defined radios. Both image classification (vision) and sentiment analysis tasks (language) have been considered to evaluate Semantic Multiplexing against state-of-the-art baselines. Our experiments indicate that Semantic Multiplexing reduces latency, energy consumption, and communication load by up to 8×, 25×, and 54×, respectively, compared to existing baselines while maintaining comparable performance. For example, image classification accuracy drops only by less than 4% when increasing from 2 to 8 the number of tasks multiplexed over a 4×4 channel. The complete software codebase and the collected datasets are publicly shared for reproducibility. A video demonstration of Semantic Multiplexing is also available.
@inproceedings{abdi2026sm,
title={{Semantic Multiplexing}},
author={Abdi, Mohammad and Meneghello, Francesca and Restuccia, Francesco},
booktitle={Proceedings of the Twenty-seventh International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing},
year={2026}
} 2026
Conference
Heterogeneous Sheaf Neural Networks
Luke Braithwaite, Alessio Borgi, Gabriele Onorato, Kristjan Tarantelli, Francesco Restuccia, Fabrizio Silvestri and Pietro Liò
NeurIPS 2026 Workshop on Symmetry and Geometry in Neural Representations
Proceedings of Machine Learning Research
Heterogeneous graphs, whose nodes and edges can belong to different types and feature spaces, arise in many real-world domains, including biology, recommendation, social networks, and computer systems. Existing heterogeneous graph neural networks typically handle this heterogeneity at the architectural level through relation-specific modules, meta-path machinery or type-aware attention, which often leads to increasingly specialised, parameter-heavy designs. In this work, we propose HetSheaf, a framework for learning heterogeneous graphs through cellular sheaves. Instead of encoding heterogeneity solely in the architecture, HetSheaf represents it directly in the underlying data structure by assigning type-aware local feature spaces and learning restriction maps conditioned on node features, node types, and edge types. To support graph-level prediction, we further introduce SheafPool, a universal stalk-space readout that aggregates node representations while being invariant to local changes of basis, thereby making graph classification with sheaf networks well-defined and achieving an F1 Score up to 42 percentage points higher than mean pooling. Across a diverse suite of benchmarks (node classification, link prediction and graph classification), HetSheaf consistently achieves up to 2 percentage points higher performance (up to 94.93% Macro F1 Score on node classification and up to 99.64% on link prediction) on the Heterogeneous Graph Benchmark (HGB) framework against homogeneous (GCN, GAT, GIN, GraphSAGE), heterogeneous (R-GCN, HAN, HGT) and type-agnostic sheaf baselines, while reducing the number of parameters by up to 10×.
@inproceedings{braithwaite2026hetero,
title={Heterogeneous Sheaf Neural Networks},
author={Luke Braithwaite and Alessio Borgi and Gabriele Onorato and Kristjan Tarantelli and Francesco Restuccia and Fabrizio Silvestri and Pietro Li\`{o}},
year={2026},
booktitle={NeurIPS 2026 Workshop on Symmetry and Geometry in Neural Representations},
editor = {Louisa Cornelis, Francisco Acosta, Alison Pouplin, David Klindt, Fatih Dinc, Nina Miolane},
series = {Proceedings of Machine Learning Research},
publisher = {PMLR},
} 2026
Conference
Semantic Heterogeneous-Agent Reinforcement Learning Systems
Mohammad Abdi, Francesca Meneghello, Nathaniel D. Bastian, and Francesco Restuccia
Proceedings of the Sixth International Conference on Dynamic Data Driven Applications Systems (DDDAS 2026)
Heterogeneous-agent reinforcement learning (HARL) enables cooperative agents with different observations and action roles to learn specialized policies. However, state-of-the-art deep neural network (DNN)-based policies are difficult to execute continuously on resource-constrained mobile agents. Edge-assisted policy execution mitigates this computation bottleneck. Yet, existing systems either offload raw observations or transfer intermediate latents through conventional network protocols, making communication a dominant source of delay. We present Semantic HARL (S-HARL), a dynamic data-driven framework that instantiates the dynamic data driven applications systems (DDDAS) model–instrumentation feedback loop for cooperative mobile agents: the edge-assisted policy dynamically incorporates live channel measurements to adapt its communication interface at execution time, while the actions it produces govern agent behavior and shape future observations. Each agent processes its observation with a local policy DNN, produces a stochastic semantic latent, and maps it through a network-conditioned precoder into a policy waveform. The edge decodes the received signals, aggregates the recovered latents with a shared global policy module, and produces actions through agent-specific heads. Instead of requiring bit-exact feature delivery, S-HARL learns representations that preserve the task semantics needed for cooperative control under current network conditions. We train S-HARL end-to-end with a unified heterogeneous-agent proximal policy optimization (HAPPO) objective and a multi-agent information-bottleneck regularizer that controls the information density of each transmitted policy waveform. We prototype S-HARL using Raspberry Pi agents, Universal Software Radio Peripheral (USRP) software-defined radios (SDRs), and an edge server. In Google Research Football (GRF), S-HARL reduces the latency of policy execution by 2.7× and the total agent-side energy by 70.7%, while matching the score rate of full-stack distributed policy execution within 0.6 percentage points.
@inproceedings{abdi2026semantic,
title={Semantic Heterogeneous-Agent Reinforcement Learning Systems},
author={Abdi, Mohammad and Meneghello, Francesca and Bastian, Nathaniel D and Restuccia, Francesco},
booktitle={Proceedings of the International Conference on Dynamic Data Driven Applications Systems},
year={2026},
organization={Springer}
} 2026
Conference
PERU: Policy Editing for Reinforcement Unlearning
Manuel Fabiano, Francesco Longo, Giovanni Merlino and Francesco Restuccia
Proceedings of Network and Distributed System Security Symposium (NDSS)
Reinforcement learning agents deployed in production systems may later be required to remove the influence of specific training trajectories: to comply with privacy regulations such as GDPR, to roll back the effect of poisoned data discovered after deployment, or to align the agent with updated operational rules. Unlike supervised models, where unlearning a sample affects only its own prediction, an agent that has unlearned a trajectory continues to operate in the environment, and the behavior it adopts on the affected states is itself a security-relevant property. Existing reinforcement unlearning approaches leave that behavior undefined. We propose Policy Editing for Reinforcement Unlearning (PERU), a framework for deployed agents that treats unlearning as behavior replacement: the data owner specifies what the agent should do on the affected states, while an auditor verifies divergence from the recorded behavior. PERU operates directly on the policy network through a two-term objective that steers it toward the specified behavior on the forget states while preserving its original output elsewhere. We evaluate PERU on Atari and MuJoCo across seven online and offline RL algorithms, under an audit protocol adapted for policy-based unlearning, and against retraining and fine-tuning baselines. PERU raises the divergence from the recorded actions on forget trajectories to 99.0% while leaving it at 2.2% on retain trajectories, and repairs an agent compromised by a published poisoning attack using corrective actions for 0.83% of the poisoned transitions.
@misc{fabiano2027peru,
title={{PERU: Policy Editing for Reinforcement Unlearning}},
author={Manuel Fabiano and Francesco Longo and Giovanni Merlino and Francesco Restuccia},
journal={Proceedings of Network and Distributed System Security Symposium (NDSS)},
year={2027}
} 2026
Conference
When Uncertainty Is the Target: Adversarial Attacks on Uncertainty-Aware Predictors
A Q M Sazzad Sayyed, Michele Caprio, Nathaniel D. Bastian, Francesco Restuccia
Proceedings of the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS)
As machine learning becomes increasingly deployed in high-stakes scenarios, studying adversarial vulnerability of uncertainty-aware models has become a compelling necessity. In this paper, we focus on adversarial attacks that manipulate predictive uncertainty rather than model predictions. We focus on predictors that decompose uncertainty into aleatoric uncertainty (AU) and epistemic uncertainty (EU) and formalize uncertainty vulnerability as the maximum increase in AU achievable under a stealth constraint where the EU is kept below a threshold. This framing yields three complementary findings. First, we theoretically show that knowledge of the baseline EU can enlarge the attacker’s feasible stealth region and increase the maximal achievable increase in AU. Second, a first-order analysis reveals that vulnerability is governed by the component of the AU gradient orthogonal to EU, enabling stealthy manipulation. Third, contrary to standard margin-based intuition, vulnerability peaks at moderate margins. Experiments across MNIST, CIFAR-10, and CIFAR-100 with standard architectures show that the proposed attack achieves non-trivial aleatoric damage while maintaining zero epistemic-stealth violation (e.g., mean AU increase of 0.0286, 0.0853, and 0.1568, respectively), whereas unconstrained uncertainty attacks violate the stealth constraint on over 90% of samples. Experiments indicate that the largest damage typically occurs in an intermediate uncertainty regime rather than at the extremes. Furthermore, consistent with our theoretical analysis, feasible attackability collapses under increasing epistemic uncertainty (e.g., mean attackability decreases from 0.0363 to 0). Overall, these results intimately connect uncertainty decomposition to adversarial geometry.
@misc{sazzad2026attk_unc,
title={When Uncertainty Is the Target:Adversarial Attacks on Uncertainty-Aware Predictors},
author={A Q M Sazzad Sayyed and Michele Caprio and Nathaniel D. Bastian and Francesco Restuccia},
journal={Proceedings of the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS)},
year={2026}
} 2026
Conference
PLACE: Patch-Level Agnostic Concept Extraction
Gabriele Onorato, Fabrizio Silvestri, Nathaniel D. Bastian, Francesco Restuccia
Proceedings of the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS)
While Vision Foundation Models (VFMs) are increasingly used in several computer vision tasks, their internal representations remain substantially opaque. This mostly stems from polysemanticity, i.e., individual neurons encode mixtures of unrelated visual patterns. In stark contrast, human reasoning is inherently monosemantic, i.e., it naturally relies on distinct, isolated features to process information. Therefore, monosemanticity needs to be implemented within the model's activation space to obtain representations that are human-understandable concepts. Existing methods mostly operate on convolutional networks, so when applied to VFMs their "crop-and-resize" approach distorts the model's representations by disrupting global self-attention mechanisms and discarding spatial geometry. Furthermore, prior work requires downstream classification labels and is based on KL-divergence, which requires to propagate the gradients back from the classifier head. Ultimately incurring in excessive concept extraction time, and making it hardly applicable to extract task-agnostic concepts. To overcome these issues, we propose PLACE, a fully unsupervised concept extraction framework that operates directly on the native patch-token activations of a VFM. By introducing a geometric Gram-matrix alignment loss, PLACE mechanistically controls downstream KL divergence, thus enforcing faithfulness -- i.e., the notion that extracted concepts, although obtained in an unsupervised fashion, are relevant to downstream classification tasks. Evaluations on image classification tasks (ImageNet and COCO) on DINOv2, ViT-MAE and ViT-B/16 show that PLACE (i) using Gram-alignment it extracts concepts about faster compared to KL-supervised extraction and faster than Sparse Autoencoders (SAEs). Moreover, PLACE yields highly sparse concepts that (ii) achieve up to 40 percentage points sparser compared to the state of the art (i.e., max 0.933 Gini sparsity score), and (iii) are up to 92.3% similar to those obtained under classifier supervision (i.e., 0.923 cosine similarity).
@misc{onorato2026place,
title={{PLACE}: Patch-Level Agnostic Concept Extraction},
author={Gabriele Onorato and Fabrizio Silvestri and Nathaniel D. Bastian and Francesco Restuccia},
journal={Proceedings of the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS)}
year={2026}
} 2026
Conference
A Metacognitive Perspective on Anomaly Detection with Vision Foundation Models
Matteo Galiazzo, Dario Loi, Gabriele Onorato, Nathaniel D. Bastian, Francesco Restuccia
Proceedings of the AAAI Fall Symposium on Artificial Metacognition (AAAI METACOG)
Anomaly Detection (AD) in images identifies defective regions by scoring against a model of normality learned from defect- free data. Current state of the art methods encode images via a frozen Vision Foundation Model (VFM) backbone and score them by their Nearest-Neighbor (NN) distance to a memory bank of normal samples. However, due to the parallel evolution of VFM encoders and scoring mechanism, the need to disentangle their individual contributions has emerged, in order to guarantee the deployment of the best performing model in specific industrial production environments. In our work, we analyze AD with a metacognitive perspective, we freeze the backbone and swap the scorer, replacing NN distance with a Mahalanobis score over a Gaussian distribution. By doing that, we can evaluate two different metacognitive functions of the scoring mechanisms: monitoring how far an observation is from the learned model of normality via a threshold-free metric, and deciding when the difference justifies an action according to a threshold. We show that these two metacognitive functions can dissociate. We further notice that replacing a NN memory bank with a simpler Mahalanobis model improves pixel-level monitoring (+8.00 AU-PRO0.05 on MVTec AD 2) while degrading the segmentation performance after the thresholding step (−13.46 F1 ). Lastly, we show that a threshold derived by Extreme Value Theory (EVT) recovers that loss without using anomalous data: the F1 difference between the two scorers narrows from 13.46 to 0.61 points on MVTec AD 2, with consistent gains on MVTec AD and VisA.
@inproceedings{galiazzo2026metacognitive,
title = {A Metacognitive Perspective on Anomaly Detection with Vision Foundation Models},
author = {Galiazzo, Matteo and Onorato, Gabriele and Loi, Dario and Bastian, Nathaniel D. and Restuccia, Francesco},
booktitle = {Proceedings of the AAAI Fall Symposium on Artificial Metacognition (METACOG)},
series = {AAAI Fall Symposium Series},
year = {2026},
month = {November},
address = {Arlington, VA, USA},
note = {To appear},
} 2026
Conference
A Metacognitive Framework for Test-Time Adaptation: Taxonomy and Research Challenges
Dario Loi, Gabriele Onorato, Antonella Coviello, Shahriar Rifat, Nathaniel D. Bastian, Francesco Restuccia
Proceedings of the AAAI Fall Symposium on Artificial Metacognition (AAAI METACOG)
Top 3 Paper Award
In Test-Time Adaptation (TTA), when out-of-distribution sam- ples interleave unpredictably with corrupted in-distribution samples, adaptation without corrective measures is prone to significant performance decrease. Existing protocols rely on unreliable factors (e.g., predictive entropy) which are unable to distinguish corrupted in-distribution data from semantically novel open-set samples. To address this issue, we formal- ize TTA through a two-level metacognitive framework that decouples task execution from control, facilitating a unified taxonomy that categorizes existing TTA protocols across three core axes, i.e., which knowledge is monitored, which part of the models are targeted for adaptation, and how adaptation occurs. We provide evidence that robust corrective measures in TTA require monitoring grounded in structured internal representations to reliably estimate model competence. Lastly, we frame monitoring as a resource-allocation problem, exam- ine neuromorphic substrates for bounded self-regulation, and outline key open challenges.
@inproceedings{loi2026metacognitive,
title = {A Metacognitive Framework for Test-Time Adaptation: Taxonomy and Research Challenges},
author = {Loi, Dario and Onorato, Gabriele and Coviello, Antonella and Rifat, Shahriar and Bastian, Nathaniel D. and Restuccia, Francesco},
booktitle = {Proceedings of the AAAI Fall Symposium on Artificial Metacognition (METACOG)},
series = {AAAI Fall Symposium Series},
year = {2026},
month = {November},
address = {Arlington, VA, USA},
note = {To appear},
} 2026
Conference
Semantic Aerial Open Radio Access Networks
Andrew Ashdown, Francesca Meneghello, Francesco Restuccia
Proceedings of IEEE MILCOM 2026
he disaggregation and interoperability enabled by Open Radio Access Networks (RANs) is redefining the possible for NextG cellular standards like 5G and 6G. While this revolution started with traditional ground-based RANs, there are now efforts to deploy fully functional Open RANs on Unmanned Aerial Vehicle (UAV) swarms to provide aerial network coverage in remote locations where installing terrestrial RANs is not an option or is overly costly. Despite the promise of such endeavors, a key deployment challenge that remains to be addressed is the optimization of the wireless fronthaul connection between UAV-based Open Radio Units (O-RUs) and UAV-based Open Distributed Units (O-DUs). Specifically, meeting the Hybrid Automatic Repeat Request (HARQ) timing constraints may become infeasible when the number of tasks requested by users increases. In this paper, we tackle this issue from a semantic perspective, accounting for the specific users’ task requirements. HARQ-imposed timing constraints are semantically alleviated by modifying both MAC-Layer HARQ and Radio Link Control (RLC)-mode retransmission functionalities. Our solution is a dynamic task-aware UAV-swarm-based Open RAN architecture entailing four O-DU robustness routes with differing RLC-mode and HARQ retransmission capabilities. Tasks are routed through one of these four O-DU routes based on semantic task requirements, UAV compute power, and wireless channel conditions. Our experimental results show up to a 51.0% reduction in HARQ failure rate with respect to a UAV-swarm-based Open RAN baseline with traditional HARQ and RLC retransmission functionalities.
@inproceedings{ashdown2026semantic,
title={Semantic Aerial Open Radio Access Networks},
author={Ashdown, Andrew and Meneghello, Francesca and Restuccia, Francesco},
booktitle={Proc. of IEEE Military Communications Conference (MILCOM)},
year={2025},
organization={IEEE}
} 2026
Conference
Neuromorphic Radio Fingerprinting
Roberta Bucchignani, Francesca Meneghello, Francesco Restuccia
Proceedings of IEEE MILCOM 2026
Real-time device authentication in multi-user multiple-input multiple-output (MU-MIMO) networks demands continuous, cryptography-free approaches deployable at the network edge. Radio fingerprinting (RFP) addresses this by leveraging tiny device-specific hardware imperfections that affect radio signal generation and processing. Traditional artificial neural networks (ANNs) are extensively used to extract such features from the radio signal collected at a monitoring device or from related metrics such as the channel state information (CSI). However, their dense, multiply-accumulate-heavy processing makes real-time edge deployment prohibitive. In this paper, we tackle this challenge by moving from dense to sparse processing. To our knowledge, we introduce NeuRFP, the first neuromorphic framework for device fingerprinting from compressed MU-MIMO CSI feedback matrices. NeuRFP encodes these matrices into binary spikes and processes them through spiking neural networks (SNNs), replacing costly multiply-accumulate operations with sparse additions. We evaluate the performance–complexity tradeoff of NeuRFP and compare it with a state-of-the-art ANN-based baseline using a publicly available 10-device RFP dataset. Our SNN-based architecture matches the baseline accuracy while reducing estimated energy consumption by more than 98%.
@inproceedings{bucchignani2026neuromorphic,
title={Neuromorphic Radio Fingerprinting},
author={Bucchignani, Roberta and Meneghello, Francesca and Restuccia, Francesco},
booktitle={Proc. of IEEE Military Communications Conference (MILCOM)},
year={2025},
organization={IEEE}
} 2026
Journal
RareCount: Interpretable Out-of-Distribution Detection in Sparse Concept Space
Antonella Coviello, Gabriele Onorato, Nathaniel D. Bastian and Francesco Restuccia
Preprint
Vision Foundation Models (VFMs) are increasingly deployed in critical applications where confident predictions on Outof- Distribution (OOD) inputs can pose safety risks. Existing OOD detectors can flag unreliable inputs but do not explain what drives their decisions, thus limiting adoption in highstakes scenarios. We introduce RareCount, an interpretable, label- and head-free OOD detector that uses Sparse autoencoders (SAEs) to extract concepts. RareCount identifies OOD samples through unusually high activation of rare concepts relative to similar samples. Operating in concept space enables score decomposition and spatial grounding of the concepts responsible for detection. On OpenOOD, RareCount is competitive with post-hoc detectors, reduces the mean far-OOD FPR95 of the only other SAE-based detector by up to 14.8 points, and generalizes across DINOv2, ViT-MAE, DINOv3, and different sparsity levels. Local conditioning improves AUROC over global rarity by 26.06 points on near-OOD and 34.75 on far-OOD, while agreement, occlusion, and patch-atypicality analyses validate concept grounding, further supported by a controlled object-level demo in mixed In-Distribution (ID)/OOD scenes.
@misc{coviello2026rarecount,
title={Implicit Sheaf Operators for Effective and Efficient Cellular Sheaf Learning},
author={Antonella Coviello and Gabriele Onorato and Nathaniel D. Bastian and Francesco Restuccia},
year={2026},
note={Preprint}
} 2026
Journal
Credal Rate-Distortion Theory
Nicolò Bonacorsi, Michele Caprio, Francesca Meneghello and F. Restuccia
under review at IEEE Transactions on Information Theory (TIT)
We develop a rate–distortion theory for lossy compression when each source symbol is associated with a credal set of plausible downstream laws. Existing universal and minimax formulations account for uncertainty in the source or distortion model but typically retain a point-valued probabilistic description for each input. Credal-set metrics, on the other hand, compare set-valued uncertainty models without specifying how several such sets should be represented when their inputs share a codeword. Under support-only semantics, we assign each codeword the convex hull of its constituent credal sets. We introduce a directed free-energy replacement distortion and prove that the distortion of a compression cell equals its worst pairwise constituent distortion. Consequently, one-shot codebook size and minimum output entropy are characterized by the chromatic number and chromatic entropy of an incompatibility graph. For product credal families, coordinatewise observation yields disjunctive graph powers, with a fixed-length exponent determined by the fractional chromatic number and an expected-prefix rate given by Körner graph entropy, while bounded-window observation yields non-tensorizing block graphs and a regularized expected-prefix rate. Under unrestricted observation and every distortion tolerance below one, the asymptotic costs reduce to those of transmitting the credal equality classes. We characterize the exact graph-saturation window and bound its size. A three-symbol construction separates the coordinatewise, two-coordinate, and unrestricted expected-prefix rates as 0, 2/3, and log2 3 bits per symbol. Experiments on the Street View House Numbers dataset illustrate the difference between credal-preserving compression and point-summary clustering.
@article{bonacorsi2026credal,
author = {Nicol\`o Bonacorsi and Michele Caprio and Francesca Meneghello and Francesco Restuccia},
title = {{Credal Rate-Distortion Theory}},
year = 2026
}
2026
Conference
EFFEKT: Efficient Federated Knowledge Transfer to Foundation Models
Matteo Calligiuri, Francesco Barbato, Pietro Zanuttigh and Francesco Restuccia
Transactions on Machine Learning Research (TMLR)
Recent data protection laws have accelerated the adoption of Federated Learning (FL) for privacy-preserving decentralized training. Nevertheless, increasing model sizes impose substantial computational demands on client devices, limiting FL applicability in resourceconstrained settings. We introduce a novel multi-domain federated learning framework in which lightweight client-side proxy models collaborate with a server-side Foundation Model (FM) to learn new concepts without sharing private data. Our approach, EFFEKT, enables efficient server-side training of domain-specific LoRA adapters while preserving feature-space alignment between the FM and proxy extractors via novel bi-directional cross-distillation strategies. Experiments on multiple real-world datasets and deployments on low-power edge devices demonstrate improvements over state-of-the-art baselines in most considered domains while maintaining lightweight computation at the client side.
@article{caligiuri2026effekt,
title={{EFFEKT: Efficient Federated Knowledge Transfer to Foundation Models}},
author={Caligiuri, Matteo and Barbato, Francesco and Zanuttigh, Pietro and Restuccia, Francesco},
journal={Transactions on Machine Learning Research},
year={2026}
} 2026
Conference
FIDEI: Hyperparameter-Free Out-of-Distribution Detection
AQM Sazzad Sayyed, Nathaniel D. Bastian and Francesco Restuccia
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
We propose a hyperparameter-free Out-of-Distribution (OOD) detection framework named FIDEI, which combines geometric and output-space cues to achieve superior OOD detection performance. FIDEI constructs a classifier-aligned hyperspherical feature space, computes a relative class-alignment score conditional to background alignment, and combines this geometric signal with logit-level confidence in a calibrated way that matches the scale of the two. FIDEI operates post hoc, does not require retraining of the backbone Deep Neural Network (DNN) nor access to auxiliary OOD data for hyperparameter tuning – thus avoiding catastrophic failure upon wrong choice of the hyperparameter. We provide a geometric interpretation showing that the resulting score approximates a cosine-margin criterion conditioned on background deviation, supporting the robustness of FIDEI across near- and far-OOD settings. Extensive experiments on CIFAR and ImageNet benchmarks demonstrate that FIDEI achieves consistently strong and balanced performance across datasets and architectures, with average relative improvement of 10% in terms of False Positive Rate (FPR) on ImageNet200 evaluations, while being more than 9× faster than methods comparable in performance in high-throughput settings. Experiments on edge devices show that FIDEI introduces less than 3% additional latency beyond backbone inference.
@inproceedings{sayyed2026fidei,
title={{FIDEI: Hyperparameter-Free Out-of-Distribution Detection}},
author={Sayyed, AQM Sazzad and Bastian, Nathaniel D. and Restuccia, Francesco},
booktitle={Proceedings of Winter Conference on Applications of Computer Vision (WACV)},
year={2026}
} 2026
Conference
Simplex Regularization Improves Out-of-Distribution Detection on Graphs
AQM Sazzad Sayyed, Nathaniel D. Bastian and Francesco Restuccia
Preprint
Node-level graph Out-of-Distribution (OOD) detection is challenging because message passing smooths node scores across neighborhoods, making In-Distribution (ID) and OOD samples appear similar. Existing propagation-based detectors improve detection by regularizing energy scores but overlook classifier geometry. As such, we introduce CollapseSafe (CSafe), a simplex-Equiangular Tight Frame (ETF) regularizer that aligns the logit space with uniformly separated prototypes of learnable norm while retaining the standard energy-propagation pipeline. This isolates the effect of classifier geometry from changes to the detector. The objective concentrates class outputs around separated prototypes, reduces class-dependent energy variation, and improves propagated ID/OOD separability. Across 20 dataset-shift pairs on eight node-level graph OOD benchmarks, CSafe outperforms the state-of-the-art NodeSafe on 17 pairs. On Cora, Citeseer, and Pubmed, it reduces the False Positive Rate (FPR) by 4.82% with a Graph Convolutional Network (GCN), showing that representation geometry is a complementary lever to information propagation for graph OOD detection.
@inproceedings{sayyed2026simplex,
title={{Simplex Regularization Improves Out-of-Distribution Detection on Graphs}},
author={Sayyed, AQM Sazzad and Bastian, Nathaniel D. and Restuccia, Francesco},
year={2026}
} 2026
Conference
On the Fragility of Out-of-Distribution Detectors Under Model Updates
AQM Sazzad Sayyed, Nathaniel D. Bastian and Francesco Restuccia
Preprint
While post-hoc Out-of-Distribution (OOD) detectors are typically evaluated assuming a static classifier, deployed models are often updated through techniques such as Machine Unlearning (MU) or Test-time Adaptation (TTA). Existing work has largely neglected this aspect; as such, we study the resulting OOD fragility, defined as the change in detection performance induced by a model update. Our study considers 16 update methods, 11 detectors, and 2 image benchmarks. Our key finding is that model updates can substantially degrade OOD detection by up to 23.82% in terms of Area Under the Receiver Operating Characteristic Curve (AUROC), while preserving In-Distribution (ID) utility. Specifically, more aggressive model updates—increasing the forget-set size for MU or the corruption severity for TTA—cause larger degradation. The degree of performance deterioration varies across OOD datasets and detector families, and feature-based detectors ($\Delta$AUROC = $-0.0338$ for RMDS) are generally more robust than confidence-based methods ($\Delta$AUROC = $-0.1429$ for GEN). These results expose a hidden reliability risk in dynamic deployment pipelines and motivate post-update reassessment and internal-signal-based prediction of OOD performance without access to OOD data.
@inproceedings{sayyed2026encore,
title={{On the Fragility of Out-of-Distribution Detectors Under Model Updates}},
author={Sayyed, AQM Sazzad and Bastian, Nathaniel D. and Restuccia, Francesco},
year={2026}
} 2026
Conference
Implicit Sheaf Operators for Effective and Efficient Cellular Sheaf Learning
Gabriele Onorato, Dario Loi, Cristian Apostol, Nathaniel D. Bastian, Francesco Restuccia
Preprint
While Sheaf Neural Networks (SNNs) enrich n-node m-edge diffusion of f -dimensional node features with d-dimensional node and edge stalks, they require computing a sparse nd × nd sheaf Laplacian, whose computation is prohibitive on edge devices. This paper addresses this issue with Implicit Sheaf Operators (ISOs), a type of sheaf topological operators that leverage the topological structure of cellular sheaves to decompose the global diffusion process into local operator- vector products. We mathematically prove that ISO eliminates the O(md2 log(md2)) sparse-coalescing term and reduces the space complexity by O(min(d, f )) with respect to traditional Neural Sheaf Diffusion (NSD). Our experiments indicate that ISO-based operators achieve latency reduction of up to 4.87× and save up to 3.92× peak GPU memory. On real graph classification data, ISO uses up to 1.24× less energy and 3.08× less on-device memory as measured on a Jetson Orin Nano. Finally, to bring ISO to practice, we introduce Sheaf Convolutional Networks (SCNs), an ISO-based SNN attaining up to 3.93% higher accuracy compared to NSD-based SNNs on node classification benchmarks.
@misc{onorato2026operators,
title={Implicit Sheaf Operators for Effective and Efficient Cellular Sheaf Learning},
author={Gabriele Onorato and Dario Loi and Cristian Apostol and Nathaniel D. Bastian and Francesco Restuccia},
year={2026},
note={Preprint}
} 2026
Journal
M3-CFR: A Domain-Adaptive Bi-Static mmWave MIMO CFR Dataset for Micro-Gesture Recognition
Khandaker Foysal Haque, Francesca Meneghello, Francesco Restuccia
IEEE Sensors Letters
In this letter, we introduce M3-channel frequency response (CFR), a bistatic millimeter-wave (mmWave) multiple-input multiple-output (MIMO) CFR dataset for fine-grained microgesture recognition under domain shifts. As wireless systems move toward integrated sensing and communication (ISAC), sensing will increasingly be realized within communication infrastructures rather than via standalone sensing hardware. Yet most existing mmWave sensing datasets are radar-centric and sensing-specific, which misaligns with communication-driven ISAC deployment. In contrast, M3-CFR provides communication-native channel frequency response (CFR) from an 8×8 fully digital mmWave MIMO ISAC testbed at 58 GHz with 1 GHz bandwidth. It contains ten microgestures from three subjects across three indoor environments (3000 synchronized CFR instances). The dataset consists of raw and domain-adaptively processed CFR. The latter includes path-loss compensation, frequency normalization, and an entropy-guided subchannel filtering pipeline, enabling cross-environment and cross-subject sensing benchmarking.
@article{haque2026m3,
title={{M3-CFR: A Domain-Adaptive Bi-Static mmWave MIMO CFR Dataset for Micro-Gesture Recognition}},
author={Haque, Khandaker Foysal and Meneghello, Francesca and Restuccia, Francesco},
journal={IEEE Sensors Letters},
year={2026},
publisher={IEEE}
} 2026
Journal
Datasets for Human Activity Recognition with Integrated Sensing and Communications in Wi-Fi
Khandaker Foysal Haque, Francesca Meneghello, Francesco Restuccia
IEEE Communications Magazine
Wi-Fi sensing has gained significant attention for its non-intrusive nature and the widespread availability of Wi-Fi-enabled devices that can be leveraged for sensing. Given the complexity of wireless propagation, sensing applications are increasingly relying on machine learning tools, which allow effectively extracting meaningful features from the radio measurements. However, developing robust, domain-adaptive algorithms that perform well across diverse environments remains challenging, requiring extensive datasets with rich diversity in terms of environments, subjects, and configurations. In this view, here we present and release two comprehensive datasets for human activity recognition (HAR) through Wi-Fi consisting of the channel state information (CSI) and beamforming feedback information (BFI) collected through commercial IEEE 802.11ac devices in both line of sight (LoS) and non-line of sight (NLoS) scenarios in different days. The first dataset entails traces related to 20 activities with six subjects across six different setups. The second dataset focuses on simultaneous multi-subject activity classification, with three individuals performing distinct activities concurrently. Our datasets are the first that also provide corresponding BFI measurements - a newly investigated and appealing primitive for Wi-Fi sensing given its ease of collection procedure. Overall, the datasets contain about 240 GB of CSI data and 230 GB for BFI.
@article{haque2026datasets,
title={{Datasets for Human Activity Recognition with Integrated Sensing and Communications in Wi-Fi}},
author={Haque, Khandaker Foysal and Meneghello, Francesca and Restuccia, Francesco},
journal={IEEE Communications Magazine},
year={2026},
publisher={IEEE}
} 2026
Conference
Multi-TAB: Multi-View Inference at the Edge with Resource-Aware Split Computing
Tanzil Bin Hassan, Kevin S. Chan, Fikadu Dagefu, Jonathan Ashdown, Flavio Esposito, Francesco Restuccia
Proceedings of IEEE WOWMOM 2026 (Best Paper Award)
Multi-view collaborative inference through deep neural networks (DNNs) aggregates views from mobile devices, thus achieving holistic situational awareness in applications such as surveillance and autonomous driving. Since on-device execution incurs significant resource consumption, offloading DNN computation to edge servers can alleviate latency demands. To this end, split computing partitions DNNs across devices and transmits intermediate compressed features to decrease the networking load. Nevertheless, unreliable wireless links between mobile devices and edge servers can increase end-to-end latency during offloading. Accordingly, we propose a novel multi-view inference framework named Multi-TAB to optimize the trade-off between DNN performance and networking resource consumption. In contrast to prior work, Multi-TAB supports split computing while adaptively compressing the intermediate latent representations at runtime based on current capacity of the wireless channel. We demonstrate the effectiveness of our framework on two challenging pedestrian density detection tasks on the WILDTRACK and MULTIVIEWX datasets. We prototyped Multi-TAB on the Colosseum network emulator and evaluated its latency and overhead data volume on Jetson Nano and Raspberry Pi 4. Experimental results show that Multi-TAB can reduce end-to-end inference latency by 33.7% compared to state-of-the-art approaches while maintaining accuracy loss within 2.4%.
@inproceedings{hassan2026multitab,
author = {Tanzil Bin Hassan and Kevin S. Chan and Fikadu Dagefu and Jonathan Ashdown and Flavio Esposito and Francesco Restuccia},
title = {{Multi-TAB}: Multi-View Inference at the Edge with Resource-Aware Split Computing},
booktitle = {Proceedings of the IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)},
year = {2026},
note = {To appear. Best Paper Award}
} 2026
Journal
FB-AII: Feedback-Based Adaptive Intentional Interference in NextG Open Radio Access Networks
Andrew Ashdown, Francesco Restuccia
Elsevier Computer Networks
The rise of Open RAN is reshaping the cellular landscape, enabling unprecedented NextG applications. Unfortunately, Open RAN networks are riddled with vulnerabilities that can be leveraged to degrade UE performance. Interference that disrupts a designated target UE are among the most harmful due to their subtlety and they have been shown to completely eliminate a target UE’s uplink data throughput. Despite the success of existing attacks, some important problems remain; (1) the interferer cannot practically determine the success of its attack since it does not have access to the target UE’s throughput and (2) completely eliminating the target UE’s uplink throughput (as does the current literature) may be unnecessary and is likely to result in the interferer’s detection. In this work we formulate a subtle interference technique that utilizes feedback from the target UE (or set of target UEs) to estimate target UE throughput and tune the level of interference dynamically. This allows the interferer to degrade the target UE throughput to a desired level. Furthermore, we expand the attack surface from a single target UE to a set of target UEs, and present two unique methods of generating the interference signal, making our approach more complex than existing work. We implement a prototype of our approach using OpenAirInterface and validate its performance using Northeastern University’s Colosseum testbed as well as simulations and find that we can degrade the throughputs of a set of target UEs 39.1% closer on average to the desired target throughputs than a baseline. As such, our interference approach highlights the need for increased security measures in Open RAN networks.
@article{ashdown2026fb,
title={{FB-AII: Feedback-Based Adaptive Intentional Interference in NextG Open Radio Access Networks}},
author={Ashdown, Andrew and Restuccia, Francesco},
journal={Computer Networks},
pages={112334},
year={2026},
publisher={Elsevier}
} 2026
Conference
SpikeCSI: CSI Feedback Compression for MIMOWireless Systems using Spiking Neural Networks
Eduardo David Lotto, Eleonora Cicciarella, Francesco Restuccia, Francesca Meneghello
Proceedings of the 24th International Symposium on Modeling and Optimization in Mobile, Ad Hoc, and Wireless Networks (WiOpt)
Multi-input, multi-output (MIMO) systems are key for next generation wireless networks to increase capacity without widening the operational bandwidth. The higher the number of antennas, the higher the capacity of the wireless system. However, this requires increasing the amount of control data – the channel state information (CSI) – that the receiver should feed back to the transmitter to properly precode data streams. In this paper, we propose SpikeCSI, a new approach for CSI compression and reconstruction that relies on spiking neural networks (SNNs) to reduce the computational complexity of the procedure. By operating through spikes rather than real numbers, SpikeCSI allows reducing the energy consumption required for MIMO channel sounding. We implement SpikeCSI and compare its performance with state-of-the-art approaches using synthetic data obtained through ray tracing simulations and real data collected through a software-defined radio (SDR)-based mmWave testbed. The results show that SpikeCSI approaches and, in some configurations, even outperform traditional methods while reducing neuron activity in the SNN compared to standard artificial neural networks (ANNs) which rely on dense processing.
2026
Conference
Credal Rate-Distortion Theory
N. Bonacorsi, M.Caprio, F. Meneghello, and F. Restuccia
Proc. of IEEE Information Theory Workshop (ITW 2026)
Traditional rate–distortion theory compresses inputs described by a single probability distribution. In this paper, we propose a new compression framework for robust inference where the uncertainty attached to an input is captured by considering a set of plausible likelihoods for the downstream task, i.e., a credal set. Existing distances between credal sets compare uncertainty models, but they do not specify the distortion induced when several inputs are merged and the decoder receives only one aggregate credal set for the whole cell. To fill this gap, we introduce a directed free-energy replacement distortion that measures the worst-case downstream cost of compression. We show that zero directed distortion occurs exactly when the replacement does not enlarge the set of plausible likelihoods. We show that the one-shot problem is exactly graph coloring, i.e., two inputs can be merged precisely when their credal likelihoods are compatible under the distortion constraint, and the minimum number of codewords is the chromatic number of the associated incompatibility graph. For product credal families and coordinate tests, block admissibility is coordinatewise and the block incompatibility graph is the OR power of the one- shot graph. The corresponding asymptotic local prefix rate is Körner graph entropy. Thus, once the credal tolerance is fixed, the compression rule is certified by the incompatibility graph, without reducing each credal set to a point prediction. On predictive credal sets built from the SVHN image-classification dataset, exact credal coloring compresses 60 inputs into 14 codewords (with a 4.286 compression ratio) while a barycenter baseline with the same number of codewords merges inputs that are credally incompatible 18.6% of the time.
@article{bonacorsi2026credal,
author = {Nicol\`o Bonacorsi and Michele Caprio and Francesca Meneghello and Francesco Restuccia},
journal = {IEEE Information Theory Workshop (ITW)},
title = {{Credal Rate-Distortion Theory}},
year = 2026
} 2026
Conference
BeamID: Domain-Adaptive Radio Fingerprinting with MIMO Beamforming Feedback
K. F. Haque, F. Meneghello, and F. Restuccia
Proc. of IEEE International Conference on Network Softwarization (NetSoft 2026)
Client identification at the network control plane has gained increasing attention for supporting monitoring and securing workflows in wireless deployment. Interestingly, device- specific hardware characteristics in the network interface card (NIC) introduce systematic impairments in channel estima- tion that percolate into multiple-input, multiple-output (MIMO) beamforming feedback. Recent work has shown that beamform- ing feedback matrices reconstructed from passively captured beamforming feedback information (BFI) frames can be exploited for accurate radio fingerprinting (RFP). However, existing fin- gerprinting approaches lead to severe performance degradation under domain shifts caused by changes in the propagation channel or deployment configuration. In this paper, we present BeamID, a domain-adaptive RFP framework that enables reliable cross-environment identification from beamforming feedback. BeamID learns radio-discriminative embeddings in a source do- main using supervised contrastive learning and performs few-shot adaptation directly in embedding space to align target-domain representations without full retraining. This design explicitly addresses environment-induced distribution shifts while preserv- ing inter-radio separability. We evaluate BeamID using real Wi-Fi measurements collected across nine distinct locations with fifteen different NICs, demonstrating that few-shot adaptation substantially restores cross-domain RFP accuracy. With as few as five labeled samples per client in a new unseen environment, BeamID achieves up to ∼99% identification accuracy while requiring only ∼2.3 seconds of adaptation time.
@article{haque2026bandweave,
author = {Khandaker Foysal Haque and Francesca Meneghello and Francesco Restuccia},
journal = {Proceedings of IEEE International Conference on Network Softwarization (NetSoft 2026)},
title = {{BeamID: Domain-Adaptive Radio Fingerprinting with MIMO Beamforming Feedback}},
year = 2026
} 2026
Journal
Metacognitive Artificial Intelligence in Vision Foundation Models: Research Challenges
Shahriar Rifat, A Q M Sazzad Sayyed, Nathaniel D. Bastian, Francesco Restuccia
IEEE Intelligent Systems
The adoption of Vision Foundation Models (VFMs) in high stakes scenarios has spurred the demand for task specific, high performance models. However, the lack of explainability of VFMs makes it challenging to ensure safety, reliability, and resilience across tasks when facing data distributions different from those seen during training. Recently, approaches based on metacognition, the human ability to regulate cognitive processes, have emerged as a promising way to understand these large models. This paper surveys the interdisciplinary connection between metacognition and state of the art VFMs, and further examines its relationship with knowledge distillation (KD), a widely used technique in VFMs. The paper concludes by outlining possible avenues for future research on the topic.
@ARTICLE{rifat2026,
author={Rifat, Shahriar and Sayyed, A. Q. M. Sazzad and Bastian, Nathaniel D. and Restuccia, Francesco},
journal={IEEE Intelligent Systems},
title={Metacognitive Artificial Intelligence in Vision Foundation Models: Research Challenges},
year={2026},
pages={1-6}
}
2026
Conference
Decision-Driven Credal Information Fusion
AQM Sazzad Sayyed, Michele Caprio, Nathaniel D. Bastian and Francesco Restuccia
2026 29th International Conference on Information Fusion (FUSION)
Modern safety-critical systems rely on multiple sensors and must fuse outputs from multiple processing streams under uncertainty. Credal sets offer a natural way to represent this uncertainty and provide a geometric structure for uncertainty-aware multi-source fusion. While prior work often treats the fusion rule as a heuristic design choice, we show that the optimal fusion rule is theoretically dictated by the selected aleatoric uncertainty (AU) functional and its order-theoretic behavior. We study two canonical credal fusion operators, i.e., conjunctive fusion, defined as the intersection of credal sets, and disjunctive fusion, defined as the core induced by the pointwise minimum of their lower probabilities. We prove that the choice between them is deterministic, i.e., if the chosen AU functional is inclusion-monotone (AU increases as the credal set grows), then conjunctive fusion minimizes uncertainty; if it is inclusion-antitone (AU decreases as the credal set grows), then disjunctive fusion is optimal. Thus, operator selection is not empirical or ad hoc, but follows directly from the monotonicity properties of the downstream objective. For finite outcome spaces, we further provide a polyhedral characterization of the disjunctive fusion core and a dual feasibility criterion ensuring coherence. We validate these predictions in a multi-sensor image classification setting under distribution shift on CIFAR10, CIFAR100, and SVHN. Using ResNet architectures, we find that AU-guided fusion improves calibration and stabilizes performance under shift relative to fixed fusion strategies. On CIFAR10, predictive accuracy improves by up to 24%, and on SVHN by up to 16%. Calibration error is reduced by up to 68.64% on CIFAR10 and 20% on SVHN.
@inproceedings{sayyed2026credalfusion,
title={{Decision-Driven Credal Information Fusion}},
author={Sayyed, AQM Sazzad and Caprio, Michele and Bastian, Nathaniel D. and Restuccia, Francesco},
booktitle={2026 29th International Conference on Information Fusion (FUSION)}
year={2026}
} 2026
Conference
DARTS: Distance-Aware Robust Training for Selective Classification
AQM Sazzad Sayyed, Nathaniel D. Bastian and Francesco Restuccia
Proceedings of the IEEE/CVF Computer Vision and Pattern Recognition Conference (Findings Track)
Neural networks are vulnerable to overconfidence. Selective classification addresses this by abstaining models from uncertain predictions. Existing selective classification approaches do not consider the geometry of the feature space; as such, they require an auxiliary classification head or rely on computation-heavy methods. To address this, we propose DARTS, a new distance-to-boundary-aware training approach that significantly improves performance by explicitly shaping feature-space geometry. DARTS provides a closed-form expression for the distance of an input to its nearest decision boundary to encourage correctly classified samples to lie farther from the boundary, while constraining misclassified ones to remain closer. Being geometry based, DARTS is architecture-agnostic and improves confidence separation and robustness without additional parameters or inference overhead. Experimental results across the CIFAR, Imagenet, CIFAR-C, and Imagenet-C benchmarks with ResNet, ConvNext, and ViT architectures show that DARTS improves Area Under Risk Coverage Curve (AURC) by upto 58% compared to state-of-the-art approaches, with consistent gains of over 20% on both CIFAR100 and Imagenet. We will release our code for reproducibility.
@inproceedings{sayyed2026darts,
title={{DARTS: Distance-Aware Robust Training for Selective Classification}},
author={Sayyed, AQM Sazzad and Bastian, Nathaniel D. and Restuccia, Francesco},
booktitle={Proceedings of the Computer Vision and Pattern Recognition Conference (Finding Track)}
year={2026}
} 2026
Journal
Wi-Fi: 25 Years and Counting
Giovanni Geraci, Francesca Meneghello, Francesc Wilhelmi, David López-Pérez, Iñaki Val, Lorenzo Galati Giordano, Carlos Cordeiro, Monisha Ghosh, Edward Knightly, Boris Bellalta
Proceedings of the IEEE
Today, Wi-Fi is over 25 years old. Yet, despite sharing the same branding name, today’s Wi-Fi boasts entirely new capabilities that were not even on the roadmap 25 years ago. This article aims to provide a holistic and comprehensive technical and historical tutorial on Wi-Fi, beginning with Institute of Electrical and Electronics Engineers (IEEE) 802.11b (Wi-Fi 1) and looking forward to IEEE 802.11bn (Wi-Fi 8). This is the first tutorial article to span these eight generations. Rather than a generation-by-generation exposition, we describe the key mechanisms that have advanced Wi-Fi. We begin by discussing spectrum allocation and coexistence, and detailing the IEEE 802.11 standardization cycle. Second, we provide an overview of the physical layer (PHY) and describe key elements that have enabled data rates to increase by over 1000× . Third, we describe how Wi-Fi medium access control (MAC) has been enhanced from the original distributed coordination function (DCF) to now include capabilities spanning from frame aggregation to wideband spectrum access. Fourth, we describe how Wi-Fi 5 first broke the one-user-at-a-time paradigm and introduced multi-user (MU) access. Fifth, given the increasing use of mobile, battery-powered devices, we describe Wi-Fi’s energy-saving mechanisms over the generations. Sixth, we discuss how Wi-Fi was enhanced to seamlessly aggregate spectrum across 2.4-, 5-, and 6-GHz bands to improve throughput, reliability, and latency. Finally, we describe how Wi-Fi enables nearby access points (APs) to coordinate in order to improve performance and efficiency. In the Appendix, we further discuss Wi-Fi developments beyond 802.11bn, including integrated millimeter-wave (IMMW) operations, sensing, security and privacy extensions, and the adoption of artificial intelligence (AI)/machine learning (ML).
@article{geraci2026wi,
title={{Wi-Fi: 25 Years and Counting}},
author={Geraci, Giovanni and Meneghello, Francesca and Wilhelmi, Francesc and L{\'o}pez-P{\'e}rez, David and Val, Inaki and Giordano, Lorenzo Galati and Cordeiro, Carlos and Ghosh, Monisha and Knightly, Edward and Bellalta, Boris},
journal={Proceedings of the IEEE},
year={2026},
publisher={IEEE}
} 2026
Conference
Crafting Adversarial Attacks to MU-MIMO OFDMA Transmissions in Wi-Fi Networks
Linda Traverso, Francesco Gringoli, Francesco Restuccia, Francesca Meneghello
Proceedings of the IEEE 23rd Consumer Communications & Networking Conference (CCNC)
The exponential growth of Wi-Fi-enabled devices and the changing nature of online activities highlight the urgent need to find innovative solutions to solve the ever-growing spectrum crunch. To address this challenge, the IEEE 802.11ax standard breaks away from traditional approaches leveraged in previous standards by jointly enabling orthogonal frequency-division multiple-access (OFDMA) and multi-user MIMO (MU-MIMO). While IEEE standardization groups are working to increase the network capacity (IEEE 802.11be) and its reliability (IEEE 802.11bn), there are still serious security issues in Wi-Fi. Some recent work has revealed that MU-MIMO transmissions can be thwarted by a malicious user that interferes with the procedure followed to set up simultaneous transmissions to multiple stations (STAs). In this work, we show that a similar attack is also effective in OFDMA MU-MIMO transmissions. Specifically, a malicious user in the network can alter the precoding procedure by transmitting adversarial feedback during the channel sounding phase, thus increasing the bit error rate (BER) experienced by legitimate STAs to up to 0.5 depending on the portion of feedback that is poisoned. We shared the code to implement our attack for reproducibility purposes and to ease its integration into digital twin frameworks of Wi-Fi networks to study and evaluate effective countermeasures.
@inproceedings{traverso2026crafting,
title={{Crafting Adversarial Attacks to MU-MIMO OFDMA Transmissions in Wi-Fi Networks}},
author={Traverso, Linda and Gringoli, Francesco and Restuccia, Francesco and Meneghello, Francesca},
booktitle={Proceedings of the IEEE 23rd Consumer Communications \& Networking Conference (CCNC)},
year={2026},
doi={10.1109/CCNC65079.2026.11366382},
organization={IEEE}
} 2026
Conference
ENCORE: A Neural Collapse Perspective on Out-of-Distribution Detection in Deep Neural Networks
AQM Sazzad Sayyed, Nathaniel D. Bastian and Francesco Restuccia
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
Out-of-Distribution (OOD) detection is of paramount importance in guaranteeing safe and reliable deployment of a Deep Neural Network (DNN) model in real-world settings. However, most OOD detection approaches still lack motivation rooted in established properties of the DNNs. This disconnect between the proposed approach and theoretical underpinning to measurable DNN properties makes these approaches unreliable. To bridge this gap, we take a different perspective to using energy scoring for OOD detection. Specifically, we look at energy score through the lens of the properties of neural collapse and observe that simple feature scaling can improve the separation between In-Distribution (ID) and OOD inputs. Based on this observation, we propose ENCORE, which scales features of each input adaptively and uses them to obtain modified logits based on insights from theory of neural collapse. We show that ENCORE outperforms state-of-the-art approaches – for example, by 1.37% on CIFAR10 and by 1.07% on Imagenet benchmarks.
@inproceedings{sayyed2026encore,
title={{ENCORE: A Neural Collapse Perspective on Out-of-Distribution Detection in Deep Neural Networks}},
author={Sayyed, AQM Sazzad and Bastian, Nathaniel D. and Restuccia, Francesco},
booktitle={Proceedings of Winter Conference on Applications of Computer Vision (WACV)},
year={2026}
} 2026
Conference
CLUE: Bringing Machine Unlearning to Mobile Devices
AQM Sazzad Sayyed, Nathaniel D. Bastian, Michael De Lucia, Ananthram Swami and Francesco Restuccia
Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
Class-level machine unlearning has been proposed to address security and privacy issues of deep neural networks (DNNs). However, existing approaches either exhibit low performance or have excessive computation/storage requirements. This makes them inapplicable in mobile computing scenarios, where computation and memory are severely constrained yet unlearning has to be performed frequently and effectively. This limitation is mainly due to the usage of a retain dataset, i.e., a sub-dataset containing the knowledge that the DNN should maintain after the unlearning. In this paper, we propose CLUE, an unlearning algorithm that does not require a retain dataset. Our key idea is to treat inputs coming from the forget class as out-of-distribution data and to use knowledge distillation to impose this constraint on the updated DNN. We have experimentally evaluated CLUE on Resnet-20, ViT-Base, and ViT-Large DNNs trained on CIFAR10, CIFAR100, and VGGFace2 datasets. We have also implemented CLUE on Raspberry PI and compared the power consumption and latency of CLUE with respect to several existing baselines. We show that CLUE improves power consumption by 68% and latency by 90% while improving the unlearning performance by up to 4.74%.
@inproceedings{sayyed2026clue,
title={{CLUE: Bringing Machine Unlearning to Mobile Devices}},
author={Sayyed, AQM Sazzad and Bastian, Nathaniel D. and De_Lucia, Michael and Swami, Ananthram and Restuccia, Francesco},
booktitle={Proceedings of Winter Conference on Applications of Computer Vision (WACV)}
year={2026}
} 2026
Conference
BandWeave: Enhanced Channel Estimation in MIMO Networks with Multi-Band Fusion
K. F. Haque, F. Meneghello, J. Ashdown and F. Restuccia
Proc. of IEEE International Conference on Computer Communications (IEEE INFOCOM 2026) — Best Paper Award
High-throughput wireless communication relies on high-resolution channel estimates to accurately compensate for the distortions imposed by the underlying wireless multipath propagation. However, due to the frequency-dependent super-position of delayed propagation paths, certain components may undergo destructive interference at specific frequencies, thus obscuring their contribution. This leads to a partial reconstruction of the channel which ultimately degrades communication performance. As such, we propose BANDW EAVE, a new general-purpose multi-band channel frequency response (CFR) fusion approach that significantly improves channel estimation accuracy by leveraging complementary multipath information across multiple frequency bands. BANDW EAVE enables adaptive, robust and efficient CFR fusion through a three-phase progressive learning algorithm. We experimentally evaluated BANDW EAVE with an IEEE 802.11ac multi-user (MU)-multiple-input multiple-output (MIMO) testbed and a 60 GHz single-user (SU)-MIMO system in three propagation environments with different fading conditions. Experimental results show that BANDW EAVE channel estimation outperforms single-band estimation by achieving more than 16% of throughput improvement and up to 10.2% bit error rate (BER) reduction. BANDW EAVE achieves a gain of more than 7.5× in communication performance compared to state of the art (SOTA) band-merging approaches. Measurements on Jetson Nano, Jetson Orin Nano and Raspberry Pi 5 show that BANDW EAVE reduces inference time and energy consumption by up to 20× and 18×, respectively, compared to SOTA methods. A video demonstration of BANDW EAVE is available.
@article{haque2026bandweave,
author = {Khandaker Foysal Haque and Francesca Meneghello and Jonathan Ashdown and Francesco Restuccia},
journal = {Proceedings of IEEE International Conference on Computer Communications (INFOCOM 2026)},
title = {{BandWeave: Enhanced Channel Estimation inMIMO Networks with Multi-Band Fusion}},
year = 2026
} 2026
Journal
Si-FI: Learning the Beamforming Feedback for Simultaneous Multi-Subject Sensing
K. F. Haque, M. Zhang, F. Meneghello, F. Restuccia
Computer Networks
There has been significant progress in Wi-Fi sensing applications pertaining to home surveillance, remote healthcare, and home entertainment among others. However, most of the work leverages manual extraction of Channel State Information (CSI) from Wi-Fi network interface card (NIC) and targets single-subject sensing. In this work, we devise a simultaneous multi-subject sensing strategy that can adapt to different environments and people being monitored. Si-FI leverages standard-compliant beamforming feedback information (BFI) as a proxy of CSI to characterize the propagation environment. Unlike CSI, BFI (i) can be captured without any firmware modifications and (ii) captures the multiple channels between the access point and the stations without any direct access to the sensing devices. Thus, conversely, from existing work, the edge server in Si-FI records the BFI of the channels between Access Point (AP) and all the stations (STAs) (sensing devices) with a single capture, reducing the channel occupation, transmission and system latency dramatically. To achieve generalization over unseen environments and people, we develop a few-shot learning algorithm named Si-FI FREL to operate with beamforming feedback angles (BFAs) (compressed BFI). We validate Si-FI through an extensive data collection campaign in 3 different environments and 3 subjects performing 20 different activities simultaneously. We demonstrate that Si-FI achieves classification accuracy of up to 99%, while Si-FI FREL improves the accuracy up to 27% when compared to the state-of-the-art domain adaptation algorithm. Si-FI reduces the system latency by 50% and channel occupation by 110 KB per sample for each sensing device compared to the state-of-the-art simultaneous multi-subject sensing work.
@article{haque2025si,
title={Si-FI: Learning the Beamforming Feedback for Simultaneous Multi-Subject Sensing},
author={Haque, Khandaker Foysal and Zhang, Milin and Meneghello, Francesca and Restuccia, Francesco},
journal={Computer Networks},
pages={111907},
year={2025},
publisher={Elsevier}
} 2025
Conference
Semantic Multiplexing
Mohammad Abdi, Francesca Meneghello, Francesco Restuccia
Technical Report
Mobile devices increasingly require the parallel execution of several computing tasks offloaded at the wireless edge. Existing communication systems only support parallel transmissions at the bit level, which fundamentally limits the number of tasks that can be concurrently processed. To address this bottleneck, this paper introduces the new concept of Semantic Multiplexing. Our approach shifts stream multiplexing from bits to tasks by merging multiple task-related compressed representations into a single semantic representation. As such, Semantic Multiplexing can multiplex more tasks than the number of physical channels without adding antennas or widening bandwidth by extending the effective degrees of freedom at the semantic layer, without contradicting Shannon capacity rules. We have prototyped Semantic Multiplexing on an experimental testbed with Jetson Orin Nano and millimeter-wave software-defined radios and tested its performance on image classification and sentiment analysis while comparing to several existing baselines in semantic communications. Our experiments demonstrate that Semantic Multiplexing allows jointly processing multiple tasks at the semantic level while maintaining sufficient task accuracy. For example, image classification accuracy drops by less than 4% when increasing from 2 to 8 the number of tasks multiplexed over a 4×4 channel. Semantic Multiplexing reduces latency, energy consumption, and communication load respectively by up to 8×, 25×, and 54× compared to the baselines while keeping comparable performance. We pledge to publicly share the complete software codebase and the collected datasets for reproducibility.
@article{abdi2025semantic,
title={Semantic Multiplexing},
author={Abdi, Mohammad and Meneghello, Francesca and Restuccia, Francesco},
journal={arXiv preprint arXiv:2511.13779},
year={2025}
} 2025
Conference
SDE-HARL: Scalable Distributed Policy Execution for Heterogeneous-Agent Reinforcement Learning
Toan D. Gian and Mohammad Abdi and Nathaniel D. Bastian and Francesco Restuccia
The Annual AAAI Conference on Artificial Intelligence (AAAI), Singapore, January 2026
Heterogeneous-agent reinforcement learning (HARL) enables agents to execute cooperative tasks by adopting agent-specific policies. Most of existing HARL methods use individual policy neural networks to ensure monotonic improvement, which leads to substantial computational overhead. The proposed SDE-HARL overcomes this limitation by decomposing each agent’s policy neural network into a lightweight local neural network and a global neural network executed at an edge server. Each local neural network generates and sends a compressed latent representation to the edge server, which aggregates the representations and produces agent-specific inferences. As such, SDE-HARL allows to significantly save computing and networking resources while preserving agent-specific behavior. A key feature of SDE-HARL is grouping agents with similar roles via a role-aware mechanism and share partial parameters in their global networks, while an identity-aware mechanism is introduced to promote behavioral diversity among agents within the same group. We prototyped SDE-HARL on an experimental testbed composed of a Jetson Nano and Raspberry PI to measure latency and network resource consumption. We evaluated SDE-HARL’s performance on several benchmark datasets, including Google Research Football and StarCraft II. Experimental results show that SDE-HARL reaches up to 90% win rate while reducing latency, energy consumption, and networking overhead respectively by 2×, 2.5×, and 5× compared to existing work. Source code is available here: https://tinyurl.com/5cvmj259.
@inproceedings{gian2026sdeharl,
title={{SDE-HARL: Scalable Distributed Policy Execution for Heterogeneous-Agent Reinforcement Learning}},
author={Toan D. Gian and Mohammad Abdi and Nathaniel D. Bastian and Francesco Restuccia},
booktitle={Proceedings of the Annual AAAI Conference on Artificial Intelligence (AAAI)},
year={2026},
} 2025
Conference
μ-VF: Enabling Virtualization of Embedded FPGAs
Vincenzo Alessio Bucaria, Francesco Longo, Giovanni Merlino and Francesco Restuccia
Proc. of ACM International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS)
Despite growing interest in virtualization of Field-Programmable Gate Arrays (FPGAs), existing approaches predominantly target datacenter-class FPGAs, which heavily rely on external (powerful) servers for hypervisor execution and resource management. This significantly limits their suitability for edge environments where autonomy, energy efficiency, and direct low-latency access to physical Input/Output (I/O) are critical. To address this goal, this paper introduces μ-VF, a lightweight virtualization framework specifically designed to enable robust multi-tenancy on embedded FPGAs operating autonomously at the network edge. μ-VF embeds all virtualization logic entirely onboard the FPGA unit, eliminating the need for any off-chip infrastructure and thus significantly reducing overall system power consumption. Each tenant operates within a secure and isolated container on the on-chip Processing System (PS), coupled with exclusive access to a dedicated Programmable Logic (PL) region. Additionally, μ-VF fully virtualizes external General-Purpose Input/Output (GPIO) directly within the PL fabric, thus enabling independent, concurrent and latency-sensitive access to shared peripherals. We have implemented a prototype of μ-VF with a Zynq UltraScale+ ZCU102 board with PL operating at 100 MHz. Experimental results demonstrate that the hardware virtualization layer utilizes less than 10% of the FPGA’s logic resources, with 85% available for tenant applications compared to 50% in prior work. Moreover, μ-VF adds 2.93% to Memory-Mapped I/O (MMIO) access latency compared to native execution for single-tenant operation, increasing to 6.5% with four concurrent tenants. Memory throughput measurements show 1.8% overhead for write operations and negligible impact on read operations, with aggregate throughput 20.58% higher than previous frameworks. Hardware-based GPIO remapping completes in 20 nanoseconds.
@article{bucaria2025muvf,
author = {Bucaria, Vincenzo Alessio and Longo, Francesco and Merlino, Giovanni and Restuccia, Francesco},
title = {$\mu${-VF: Enabling Virtualization of Embedded FPGAs}},
year = {2025},
issue_date = {December 2025},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
volume = {9},
number = {3},
url = {https://doi.org/10.1145/3771581},
doi = {10.1145/3771581},
journal = {Proc. ACM Meas. Anal. Comput. Syst.},
month = dec,
articleno = {66},
numpages = {26},
} 2025
Conference
FlexRFML: Dynamic Neural Networks on FPGAs for Next-Generation Radio Spectrum Perception
Francesco Pessia, Sazzad Sayyed and Francesco Restuccia
Proc. of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (ACM MobiHoc)
Enabling spectrum perception with deep neural networks (DNNs) directly connected to the radio front-end is of fundamental importance to realize next-generation spectrum-aware wireless systems. As such, low-latency DNN inference in reconfigurable hardware such as Field Programmable Gate Arrays (FPGAs) is a necessary precursor to enable spectrum perception in real-world wireless systems. The key issue with existing work is that it considers DNNs that have fixed weights and architecture. On the other hand, it hasbeen shown that dynamically changing the structure and weights of the DNN at runtime can lead to improved efficiency and adaptability. This work fills the current research gap by proposing FlexRFML, the first framework to integrate dynamic DNNs in the RF-front spectrum perception loop. FlexRFML includes High-Level Synthesis (HLS)-based design as well as customized circuits to achieve dynamic hardware reconfiguration and accelerate the DNN. We have prototyped FlexRFML on a Xilinx system-on-chip (SoC) ZCU102 by considering both modulation recognition and radio fingerprinting classification problems where the DNNs are dynamically adapted based on a preliminary classification of the input. Experimental results show that FlexRFML can decrease the inference latency by up to 35.6% with respect to static DNN inference with negligible additional hardware overhead. We pledge to release the FlexRFML hardware and software code.
@inproceedings{pessia2025flexrfml,
title={{FlexRFML: Dynamic Neural Networks on FPGAs for Next-Generation Radio Spectrum Perception}},
author={Francesco Pessia and Sazzad Sayyed and Francesco Restuccia},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (ACM MobiHoc)},
address={Houston, Texas},
month={October},
pages={XXX--XXX},
year={2025},
doi={XXX}
} 2025
Conference
AdvO-RAN: Adversarial Deep Reinforcement Learning in AI-Driven Open Radio Access Networks
T. Hassan, F. Meneghello and F. Restuccia
Proc. of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (ACM MobiHoc)
While artificial intelligence (AI) is improving the performance of O-RAN, it will also expose the network to adversarial machine learning (AML) attacks. For this reason, in this paper, we are the first to investigate AML in the context of deep reinforcement learning (DRL)-based O-RAN xApps. What separates AML in O-RAN from traditional settings is the need to design and analyze adversarial attacks based on RAN-specific Key Performance Measures (KPMs) such as transmitted bit rate, downlink buffer occupancy, transmitted packets, etc. As such, we propose the AdvO-RAN framework, which includes (i) a new adversarial perturbation generator using preference-based reinforcement learning (PbRL) to learn the pertur- bation that most violate the user service level agreements (SLA) and (ii) a robust training module for enhancing DRL agent resilience to the attacks in (i). We experimentally evaluate AdvO-RAN on the Colosseum network emulator. Experimental results show that AdvO-RAN can enhance xApp performance by reducing SLA violations from 44% to 27% on average and reducing by 46% the latency under the most challenging attack scenario for Ultra-Reliable Low-Latency Communications (URLLC) traffic. AdvO-RAN can improve up to 75% of throughput in the victim Enhanced Mobile Broadband (eMBB) slice users during a constant bit-rate traffic scenario.
@inproceedings{hassan2025advoran,
title={AdvO-RAN: Adversarial Deep Reinforcement Learning in AI-Driven Open Radio Access Networks},
author={Hassan, T. and Meneghello, F. and Restuccia, F.},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (ACM MobiHoc)},
address={Houston, Texas},
month={oct},
pages={XXX--XXX},
year={2025},
doi={XXX},
note={Acceptance rate: 23\%}
} 2025
Conference
SHRINK: Reducing MIMO Feedback Overhead in Wi-Fi with Dynamic Data-Driven Channel Sounding
KM Rumman, F. Meneghello, K. F. Haque, F. Gringoli and F. Restuccia
Proc. of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (ACM MobiHoc)
The performance of multiple-input, multiple-output (MIMO) systems highly depends on the precision of channel estimates provided by the mobile users. However, the current Wi-Fi standard requires an update interval of 10 ms, irrespective of the channel dynamics. This imposes a substantial overhead for the MIMO channel estimation. Recent work mainly targets different compression strategies, potentially compromising precoding accuracy and, in turn, the network performance. In stark opposition, we propose SHRINK, a framework to dynamically adapt the feedback transmission rate to the propagation environments and performance requirements. SHRINK determines whether the users should send back their channel estimates by predicting network performance through a data-driven analysis of prior and current channel estimates. We have experimentally evaluated SHRINK using off-the-shelf Wi-Fi devices in multiple environments, including an anechoic chamber, and benchmarked its performance against several state-of-the-art approaches. Experimental results show that SHRINK reduces airtime and data overhead by 81% on average compared to the IEEE 802.11 standard without impacting the precoding performance. Moreover, SHRINK outperforms state-of-the-art approaches by an average gain of 33.6% in airtime and data overhead reduction, corresponding to an increase in throughput of 24.5%.
@inproceedings{rumman2025shrink,
title={SHRINK: Reducing MIMO Feedback Overhead in Wi-Fi with Dynamic Data-Driven Channel Sounding},
author={Rumman, K. M. and Meneghello, Francesca and Haque, Khandaker Foysal and Gringoli, Francesco and Restuccia, Francesco},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (ACM MobiHoc)},
address={Houston, Texas},
month={October},
pages={XXX--XXX},
year={2025},
doi={XXX},
note={Acceptance rate: 23\%}
}
2025
Conference
Breaking the Barrier to Downlink Packet Capture in OFDMA Wi-Fi Networks
S. Fontana, F. Meneghello, J. Link, F. Restuccia, M. Hollick, F. Gringoli,
Proc. of ACM Workshop on Wireless Network Testbeds, Experimental evaluation & Characterization (ACM WINTECH)
Orthogonal frequency-division multiple access (OFDMA) is considered as one of the main enablers for scaling Wi-Fi throughput. Although the technology was formally introduced with the publication of the IEEE 802.11ax amendment in May 2021, the first commercially available products had already appeared in 2018, and since then millions of access points and client devices have been sold. Still, researchers lack a practical way to observe the OFDMA-based multi-user frame exchanges in the wild. We close this gap by designing and implementing the first open-source complete toolkit that captures every downlink frame of an 802.11ax OFDMA network using only commodity hardware augmented with lightweight firmware patches. An extensive experimental campaign across diverse access points quantifies capture and timing accuracy, error rates, and reveals scheduler behaviors that are opaque to conventional sniffers. By exposing both the capabilities and the limits of COTS-based monitoring, our work provides a foundation for future research on next-generation Wi-Fi MAC protocols and performance tuning.
@inproceedings{fontana2025breaking,
title={Breaking the Barrier to Downlink Packet Capture in OFDMA Wi-Fi Networks},
author={Fontana, S. and Meneghello, F. and Link, J. and Restuccia, F. and Hollick, M. and Gringoli, F.},
booktitle={Proceedings of the 19th ACM Workshop on Wireless Network Testbeds, Experimental evaluation \& Characterization},
pages={XXX--XXX},
year={2025},
note={To appear; Hong Kong, China, November 4--8, 2025; DOI: TBD}
}
2025
Conference
FEARL: AI-Assisted Energy-Aware Real-Time Receiver Adaptation to Dynamic Environments
Mohammad Abdi, Diptashree Das, and Minghan Liu, Marvin Onabajo and Francesco Restuccia
IEEE International Conference on Communications (ICC)
@inproceedings{abdi2025fearl,
title={FEARL: AI-Assisted Energy-Aware Real-Time Receiver Adaptation to Dynamic Environments},
author={Abdi, Mohammad and Das, Diptashree and Liu, Minghan and Onabajo, Marvin and Restuccia, Francesco},
year={2025},
organization={IEEE International Conference on Communications (ICC)}
} 2025
Journal
6G Horizons: Toward Intelligent and Sustainable Connectivity
Anwer Al-Dulaimi, Sumei Sun, Francesco Restuccia, Tommy Svensson, Hao Guo
IEEE Vehicular Technology Magazine
@article{al20256g,
title={6G Horizons: Toward Intelligent and Sustainable Connectivity [From the Guest Editors]},
author={Al-Dulaimi, Anwer and Sun, Sumei and Restuccia, Francesco and Svensson, Tommy and Guo, Hao},
journal={IEEE Vehicular Technology Magazine},
volume={20},
number={1},
pages={40--42},
year={2025},
publisher={IEEE}
} 2025
Conference
Radio Fingerprinting of Wi-Fi Devices Through MIMO Compressed Channel Feedback
Francesca Meneghello, Khandaker Foysal Haque and Francesco Restuccia
Proc. of IEEE Conference on Computer Communications Workshops (INFOCOM Workshops)
In this paper, we present DeepCSIv2, a data-driven radio fingerprinting (RFP) algorithm to characterize Wi-Fi devices acting as stations (STAs) at the physical layer. Our approach relies on STA-specific hardware impairments extracted from the multiple-input, multiple-output (MIMO) channel state information (CSI) feedback transmitted unencrypted by the STAs to Wi-Fi access points (APs) to establish MIMO transmissions. Recent work showed that such feedback can be effectively used to fingerprint devices acting as APs. In this work, we demonstrate that the same control information can be leveraged to obtain relevant features describing the STAs. Our intuition is that even tiny STA-specific hardware characteristics introduce detectable impairments in the channel estimated by the STA, and percolate in the CSI feedback – consisting of the compressed and quantized channel estimate. DeepCSIv2 is based on a neural network architecture that automatically extracts the STA’s radio fingerprint from the feedback captured over the air and identifies the device. We evaluated DeepCSIv2 through an extensive data collection campaign using 18 commercial IEEE 802.11ax network interface cards (NICs) of the same type from the same vendor. We considered different experiment configurations, changing the propagation environment and the operational bandwidth. The results show that DeepCSIv2 reaches an accuracy higher than 96% in identifying the 18 NICs in our dataset. We will share the dataset and RFP code with the community for reproducibility.
@INPROCEEDINGS{11152893,
author={Meneghello, Francesca and Haque, Khandaker Foysal and Restuccia, Francesco},
booktitle={IEEE INFOCOM 2025 - IEEE Conference on Computer Communications Workshops (INFOCOM WKSHPS)},
title={Radio Fingerprinting of Wi-Fi Devices Through MIMO Compressed Channel Feedback},
year={2025},
volume={},
number={},
pages={1-6},
keywords={Neural networks;Channel estimation;Fingerprint recognition;Physical layer;Hardware;Reproducibility of results;MIMO;Object recognition;Network interfaces;Wireless fidelity;Radio fingerprinting;Wi-Fi;MIMO;compressed beamforming feedback;channel sounding},
doi={10.1109/INFOCOMWKSHPS65812.2025.11152893}}
2025
Conference
Digitally Tunable CMOS Mixer Design for Adaptive RF Front-Ends
Diptashree Das, Minghan Liu, Mohammad Abdi, Francesco Restuccia, Marvin Onabajo
IEEE International Symposium on Circuits and Systems (ISCAS) 2025
Reconfigurability and self-optimization have become essential during radio frequency integrated circuit (RFIC) design to support the growing number of devices and fast changes in the surrounding wireless spectrum. This has created the need to develop new design approaches for RFICs based on end-to-end wireless system-level performance metrics during operation in dynamically changing communication environments. This paper introduces a CMOS mixer with a wide range of digitally tunable bias current for machine learning (ML) based adaptation. The mixer is designed to become part of a self-adaptive receiver (RX) architecture that is capable of optimizing its performance in accordance to wireless channel conditions by evaluating systemlevel parameters. The proposed mixer topology has digitally tunable bias current and a programmable helper current (PHC) circuit to maintain the voltage headroom during operation with a wide tuning range. It is capable of dynamically minimizing power consumption based on performance and wireless network requirements. Post-layout simulation results show that the power consumption can be reduced up to 8x depending on momentary performance needs, while maintaining an IIP3 greater than -2.4 dBm for the entire range of operation.
@inproceedings{das2025digitally,
title={Digitally Tunable CMOS Mixer Design for Adaptive RF Front-Ends},
author={Das, Diptashree and Liu, Minghan and Abdi, Mohammad and Restuccia, Francesco and Onabajo, Marvin},
booktitle={2025 IEEE International Symposium on Circuits and Systems (ISCAS)},
pages={1--5},
year={2025},
organization={IEEE}
} 2025
Journal
A Low-Power Differential Temperature Sensor with Chopped Cascode Transistors and Switched-Capacitor Integration
Junyi Yang, Thomas Gourousis, Mengting Yan, Ruyi Ding ,Ankit Mittal, Milin Zhang, Francesco Restuccia, Aatmesh Shrivastava, Yunsi Fei, and Marvin Onabajo
Electronics
Embedded differential temperature sensors can be utilized to monitor the power consumption of circuits, taking advantage of the inherent on-chip electrothermal coupling. Potential applications range from hardware security to linearity, gain/bandwidth calibration, defect-oriented testing, and compensation for circuit aging effects. This paper introduces the use of on-chip differential temperature sensors as part of a wireless Internet of Things system. A new low-power differential temperature sensor circuit with chopped cascode transistors and switched-capacitor integration is described. This design approach leverages chopper stabilization in combination with a switched-capacitor integrator that acts as a low-pass filter such that the circuit provides offset and low-frequency noise mitigation. Simulation results of the proposed differential temperature sensor in a 65 nm complementary metal-oxide-semiconductor (CMOS) process show a sensitivity of 33.18V/∘C within a linear range of ±36.5m∘C and an integrated output noise of 0.862mVrms (from 1 to 441.7 Hz) with an overall power consumption of 0.187mW . Considering a figure of merit that involves sensitivity, linear range, noise, and power, the new temperature sensor topology demonstrates a significant improvement compared to state-of-the-art differential temperature sensors for on-chip monitoring of power dissipation.
@article{yang2025low,
title={A Low-Power Differential Temperature Sensor with Chopped Cascode Transistors and Switched-Capacitor Integration},
author={Yang, Junyi and Gourousis, Thomas and Yan, Mengting and Ding, Ruyi and Mittal, Ankit and Zhang, Milin and Restuccia, Francesco and Shrivastava, Aatmesh and Fei, Yunsi and Onabajo, Marvin},
journal={Electronics},
volume={14},
number={12},
pages={2381},
year={2025},
publisher={MDPI}
} 2025
Journal
RF Anti-Jamming via Multi-Level Howells-Applebaum Null-Forming: 32-Channels, 5.8 GHz/100 MHz/Beam on Xilinx Sx475T FPGA
Umesha Kumarasiri , Sivakumar Sivasankar , Hasitha Weerasooriya, Hiruni Silva, Chamira Edussooriya , Viduneth Ariyarathna , Francesco Restuccia and Arjuna Madanayake
Real-time sensing and perception of the radio spectrum based on artificial intelligence (AI) is crucial for emerging intelligent wireless and electronic warfare systems. However, sensing can be greatly impacted by harmful radio frequency interference (RFI). Emerging drone warfare allows many RFI sources/jammers to be distributed across a wide field-of-view which necessitates real-time measurement, adaptation and aperture nulling to remove the RFI before AI-based sensing and perception of sources of interest can occur. This work explores algorithmic innovations that improve the computational complexity of classical Howells-Applebaum adaptive nulling algorithm to enable fast, real-time adaptive operation at significantly lower arithmetic complexity. Design examples for AI-enabled sensing and perception across a 32-element antenna receiver with 32 independent channels and a Xilinx Virtex-6 Sx475 FPGA backend are discussed. Examples show computer architecture for digital signal processing and AI algorithms operating on the FPGA, with real-time measurements for spectrum sensing and modulation recognition on the RadioML2018.a dataset with and without the proposed adaptive nullforming system. A general adversarial AI-based spectrum perception architecture that allows both jamming of opponents while simultaneously nulling out RFI and conducting AI-based radio intelligence applications is examined and demonstrated in the 5.7-5.8 GHz band using a 32 element real-time FPGA realization. Modulation recognition is demonstrated for 16/32-QAM signals under heavy RFI conditions with additional “in the wild” RFI sources present.
@article{kumarasiri2025rf,
title={RF Anti-Jamming via Multi-Level Howells-Applebaum Null-Forming: 32-Channels, 5.8 GHz/100 MHz/Beam on Xilinx Sx475T FPGA},
author={Kumarasiri, Umesha and Sivasankar, Sivakumar and Weerasooriya, Hasitha and Silva, Hiruni and Edussooriya, Chamira and Ariyarathna, Viduneth and Restuccia, Francesco and Madanayake, Arjuna},
journal={IEEE Journal of Radio Frequency Identification},
year={2025},
publisher={IEEE}
} 2025
Journal
Semantic Edge Computing and Semantic Communications in 6G Networks: A Unifying Survey and Research Challenges
Milin Zhang, Mohammad Abdi, Venkat R. Dasari and Francesco Restuccia
Computer Networks
Semantic Edge Computing (SEC) and Semantic Communications (SemComs) have been proposed as viable approaches to achieve real-time edge-enabled intelligence in sixth-generation (6G) wireless networks. On one hand, SemCom leverages the strength of Deep Neural Networks (DNNs) to encode and communicate the semantic information only, while making it robust to channel distortions by compensating for wireless effects. Ultimately, this leads to an improvement in the communication efficiency. On the other hand, SEC has leveraged distributed DNNs to divide the computation of a DNN across different devices based on their computational and networking constraints. Although significant progress has been made in both fields, the literature lacks a systematic view to connect both fields. In this work, we fill the current gap by unifying the SEC and SemCom fields. We summarize the research problems in these two fields and provide a comprehensive review of the state of the art with a focus on their technical strengths and challenges.
@article{ZHANG2025111531,
title = {Semantic Edge Computing and Semantic Communications in 6G networks: A unifying survey and research challenges},
journal = {Computer Networks},
volume = {270},
pages = {111531},
year = {2025},
issn = {1389-1286},
doi = {https://doi.org/10.1016/j.comnet.2025.111531},
url = {https://www.sciencedirect.com/science/article/pii/S1389128625004980},
author = {Milin Zhang and Mohammad Abdi and Venkat R. Dasari and Francesco Restuccia},
keywords = {Semantic Edge Computing, Semantic Communication, Split computing, Collaborative intelligence}} 2025
Journal
Finding a Needle in a (Spectrum) Haystack: Multi-Band Multi-Device Radio Fingerprinting
Ildi Alla, Milin Zhang, Jonathan Ashdown, Valeria Loscri, Francesco Restuccia
Preprint
As the spectrum becomes increasingly crowded, quick and reliable authentication of wireless devices is critical to avoid harmful interference to incumbents of the spectrum. Radio fingerprinting achieves fast waveform-level authentication by distinguishing devices based on unique hardware imperfections in the radio circuitry. However, existing approaches can fingerprint only one signal in a specific band, making them inapplicable in real-world scenarios where multiple signals coexist in spectrum bands. This paper introduces Multi-band Multi-device Radio Fingerprinting (M2RF) to address this challenge. Specifically, we propose a learning-driven segmentation algorithm to directly process in-phase/quadrature (I/Q) samples coming from the receiver and assign each I/Q sample to a specific radio. In contrast to existing approaches, M2RF simultaneously identifies and locates in the spectrum multiple devices that emit overlapping signals and avoids the burden of processing data, making the overall approach with reduced overhead and faster. Our approach can be generalized to different channels and signal bandwidths with- out retraining, making it scalable. Experiments in three different spectrum scenarios under 2 transmission conditions and with 15 radio transmitters demonstrate the effectiveness of M2RF, achieving up to 99.56% of F1-score, and 92.44% detection rate of malicious users with only a 2.72% mean Miss Rate (MR). Dataset and code will be shared for reproducibility and a demo video is available.
@misc{alla2025finding,
title={Finding a Needle in a (Spectrum) Haystack: Multi-Band Multi-Device Radio Fingerprinting},
author={Alla, Ildi and Zhang, Milin and Ashdown, Jonathan and Loscri, Valeria and Restuccia, Francesco},
year={2025}
} 2025
Conference
Resilient Wireless Communications with Selective Deep Neural Network Classification
A. Q. M. Sazzad Sayyed, Jonathan Ashdown, Michael De Lucia, Ananthram Swami, Nathaniel D. Bastian and Francesco Restuccia
Proc. of IEEE Military Communications Conference (MILCOM)
The pervasive presence of deep neural networks (DNNs) has made artificial intelligence (AI) an integral part of tactical communication and networking systems. However, the overconfidence of modern DNNs poses significant risks in mission-critical applications. Selective Classification (SC) is a promising paradigm to mitigate this issue by enabling DNNs to abstain from unreliable predictions. Although there is a plethora of work on SC in the computer vision (CV) domain, there is no evaluation of the utility of these approaches in tactical communication scenarios. This work evaluates State-of-the-art (SOTA) SC methods within the tactical domain, focusing on their applicability to Automatic Modulation Classification (AMC) and Network Intrusion Detection System (NIDS) tasks using the RadioML and ACI IoT datasets, respectively. Our key findings show a 55% risk reduction for AMC with a 20% loss in coverage and near-zero risk for NIDS with minimal coverage loss. Furthermore, we analyze SC performance under distribution shifts, revealing limitations of traditional methods in handling covariate and semantic shifts. Finally, we explore training strategies to enhance SC performance.
@inproceedings{sayyed2025selective,
title={Resilient Wireless Communications with Selective Deep Neural Network Classification},
author={A. Q. M. Sazzad Sayyed, Jonathan Ashdown, Michael De Lucia, Ananthram Swami, Nathaniel D. Bastian and Francesco Restuccia},
booktitle={MILCOM 2025-2025 IEEE Military Communications Conference (MILCOM)},
year={2025}
} 2025
Conference
NI-Diff: Zero-Day and Adversarial Network Intrusion Detection with Diffusion Models
Milin Zhang, Michael De Lucia, Ananthram Swami, Jonathan Ashdown, Nathaniel D. Bastian and Francesco Restuccia
Proc. of IEEE Military Communications Conference (MILCOM)
While Deep Learning (DL) has achieved remarkable success in Network Intrusion Detection System (NIDS), its inherent data-driven nature makes it vulnerable to distribution shift. This limitation exposes DL-based NIDS to both adversarial attacks that are crafted by adding subtle change to original samples and zero-day attacks that are out-of-distribution (OOD) data unseen during training. However, existing work focusing on adversarial detection often fails to identify zero-day attacks and vice versa, leaving a security gap in DL-based NIDS. We propose NI-Diff, a novel detection approach that can effectively identify both adversarial network flow as well as zero-day intrusion by estimating their distribution with generative models. More specifically, we leverage a variational auto-encoder to map the network flow into a latent space and use a diffusion model to reconstruct the likely-hood from noise. Our key intuition is that the in-distribution data and the reconstructed data will have a similar likelyhood which results in similar inference output in the DL classifier. Extensive experiments on two large-scale NIDS datasets demonstrate that our approach can effectively identify 97% adversarial network flow and 92% zero-day threat with less than 2% false positive rate, outperforming state-of-the-art adversarial detection and OOD detection baselines.
@inproceedings{zhang202nidiff,
title={NI-Diff: Zero-Day and Adversarial Network Intrusion Detection with Diffusion Models},
author={Milin Zhang, Michael De Lucia, Ananthram Swami, Jonathan Ashdown, Nathaniel D. Bastian and Francesco},
booktitle={MILCOM 2025-2025 IEEE Military Communications Conference (MILCOM)},
} 2025
Conference
Method for Multi-Antenna Networking Through Multi-path Feedback Extraction and Reporting
Francesco Restuccia, Arman Elyasi, Francesca Meneghello
Technical Report
In this technical report, we present a new mechanism for channel sounding feedback in wireless multiple-input, multiple-output (MIMO) networks that significantly reduces feed-back overhead while not degrading communication performance. Our new approach consists of using the multi-path representation of the wireless propagation as an antenna-and bandwidth-independent—and thus scalable—feedback instance instead of relying on the channel frequency response (CFR) which should be fed back for each transmitter and receiver antenna pair over the entire operational bandwidth. Previous approaches propose to compress the CFR with heuristic or learning-based approaches for efficient feedback. Yet, the resulting feedback does not scale with the number of antennas in the MIMO system and the operational bandwidth. Thanks to the sparse nature of wireless multi-path channels, our approach provides a highly compressed representation of the CFR. Indeed, the number of multi-path components is less than 30 in typical indoor environments. Moreover, we show that the overhead can be further reduced by transmitting only the dominant multi-path components. Our preliminary evaluation demonstrates that the proposed multi-path parameter-based feedback allows maintaining near-optimal performance with minimal bit error rate (BER) degradation compared to CFR feedback, while drastically reducing the feedback spectrum usage. This scalable approach addresses the critical feedback overhead bottleneck of MIMO wireless networks. This will enable the use of more antennas and wider bandwidths, thus improving spectrum efficiency and enhancing multi-user performance in next-generation scenarios where channel resources are expected to be scarce due to the growing number of connected users.
@inproceedings{elyasi2025multipath,
title={{Method for Multi-Antenna Networking Through Multi-path Feedback Extraction and Reporting}},
author={Arman Elyasi and Francesca Meneghello and Francesco Restuccia},
booktitle={Technical Report},
year={2025}
} 2025
Conference
SINF: Semantic Inference in Neural Networks with Semantic Subgraphs
AQM Sazzad Sayyed, Francesco Restuccia
ICML 2025 Workshop on Machine Learning for Wireless Communication and Networks (ML4Wireless)
This paper proposes Semantic Inference (SINF) that creates semantic subgraphs in a deep neural network (DNN) based on a new Discriminative Capability Score (DCS) to drastically reduce the DNN computational load with limited performance loss. We evaluate the performance of SINF on VGG16, VGG19, and ResNet50 DNNs trained on CIFAR100 and a subset of the ImageNet dataset. Moreover, we compare its performance against 6 state-of-the-art pruning approaches. Our results show that (i) on average, SINF reduces the inference time of VGG16, VGG19, and ResNet50 respectively by up to 29%, 35%, and 15% with only 3.75%, 0.17%, and 6.75% accuracy loss for CIFAR100 while for ImageNet benchmark, the reduction in inference time is 18%, 22%, and 9% for accuracy drop of 3%, 2.5%, and 6%; (ii) DCS achieves respectively up to 3.65%, 4.25%, and 2.36% better accuracy with VGG16, VGG19, and ResNet50 with respect to existing discriminative scores for CIFAR100 and the same for ImageNet is 8.9%, 5.8%, and 5.2% respectively. Through experimental evaluation on Raspberry Pi and NVIDIA Jetson Nano, we show SINF is about 51% and 38% more energy efficient and takes about 25% and 17% less inference time than the base model for CIFAR100 and ImageNet.
@inproceedings{sayyed2025sinf,
title={SINF: Semantic Inference in Neural Networks with Semantic Subgraphs},
author={Sayyed, AQM Sazzad and Restuccia, Francesco},
booktitle={ICML 2025 Workshop on Machine Learning for Wireless Communication and Networks (ML4Wireless)}
} 2025
Conference
Semantic Subgraph Extraction Attacks To Convolutional Neural Networks
Gabriele Restuccia, Ilenia Tinnirello, Francesco Restuccia
ICML 2025 Workshop on Machine Learning for Wireless Communication and Networks (ML4Wireless)
Existing adversarial machine learning (AML) methods typically require substantial computational resources. In this work, we investigate how class-level semantic relationships can be exploited to influence adversarial attack performance. Specifically, we first aim to quantify the effect of known factors, such as semantic similarity–defined as the degree of shared intrinsic attributes–on attack efficiency. Experiments on CIFAR-100 and Tiny-ImageNet using VGG and ResNet architectures show that targeting semantically similar classes can reduce perturbation magnitudes and iterations by up to 41% and 23%, respectively. Motivated by these findings, we introduce a lightweight, one-shot, semantic subgraph extraction attack (SSEA), which constructs semantic subgraphs by leveraging class-level semantic relationships between source and adversarial target classes. Our method extracts subgraphs in a single inference pass, requires no fine-tuning or external models, preserves the original network weights, and integrates seamlessly into any white-box attack scenario. On CIFAR-100, SSEA improves the Top-1 attack success rate (ASR) by up to 8.17% for PGD on VGG-19 and nearly doubles the effectiveness of the Jitter attack. Additionally, our approach reduces floating-point operations and model size by up to 18% and 42%, respectively.
@inproceedings{restuccia2025semantic,
title={Semantic Subgraph Extraction Attacks To Convolutional Neural Networks},
author={Restuccia, Gabriele and Tinnirello, Ilenia and Restuccia, Francesco},
booktitle={ICML 2025 Workshop on Machine Learning for Wireless Communication and Networks (ML4Wireless)}
}
2025
Journal
Out-of-Distribution Detection in Computer Vision: A Comprehensive Survey and Research Challenges
AQM Sazzad Sayyed, Shahriar Rifat, Milin Zhang, Ananthram Swami, Michael De Lucia, Nathaniel D. Bastian, Francesco Restuccia
ACM Computer Surveys (Under Review)
Neural networks deployed in high-stakes scenarios need to provide inference that is robust to out-of-distribution (OOD) inputs. Although the resilience of neural networks to such inputs has been widely investigated in the uncertainty estimation and OOD detection communities, a unified vision of these inherently intertwined problem domains is still missing. In this work, we fill this gap by providing a survey of the state of the art that brings together research on both uncertainty estimation and OOD detection, ultimately offering a consolidated view of methods, tools, and benchmarks. We categorize existing OOD techniques by the task they address (multi-class and multi-label classification, segmentation, object detection), thus exposing progress and remaining challenges beyond the usual multi-class setting. Moreover, we highlight emerging themes such as test-time adaptation for OOD detection and new theoretical results that probe its fundamental limits, thereby outlining promising and under-explored directions for future research.
@article{sayyed2025ood,
author = {AMQ Sazzad Sayyed and Shahriar Rifat and Milin Zhang and Ananthram Swami and Michael De Lucia and Nathaniel D. Bastian and Francesco Restuccia},
title = {{Out-of-Distribution Detection in Computer Vision: A Comprehensive Survey and Research Challenges}},
year = {2025}
}
2025
Journal
DECOR: Multi-Modal Decentralized Cluster-based Energy Efficient Covert Routing in HetNets
K. F. Haque , J. H. Kong , T. J. Moore , K. Chan , F. Restuccia , and F. T. Dagefu
IEEE (Under Review)
State-of-the-art covert routing in heterogeneous net- work (HetNet) focuses on balancing covertness and through- put, but often overlooks explicit energy optimization. While covert communication inherently limits transmit power, meet- ing throughput demands without coordinated design can still lead to high energy consumption. In contrast, our approach jointly considers covertness, throughput, and energy efficiency, addressing all three objectives simultaneously. To this end, we propose DECOR, a Decentralized Energy-efficient COvert Routing framework that jointly optimizes covertness, throughput, and energy efficiency. Unlike traditional methods that use a single wireless technology, DECOR leverages the diversity of available wireless communication technologies in HetNet to enable si- multaneous multi-modal routing. The core idea behind DECOR is that optimal simultaneous utilization of multiple modalities improves throughput and overall energy efficiency. It minimizes the end-to-end energy consumption while satisfying stringent constraints on throughput and covertness through two core steps: (1) link-level optimization using sequential least squares pro- gramming (SLSQP), and (2) network-level optimization through a custom cluster-based routing strategy. DECOR introduces a novel clustering-based strategy that aggregates intra-cluster link information and delegates routing decisions to cluster heads, significantly reducing control overhead and enabling scalable, energy-efficient covert communication. Extensive numerical anal- ysis demonstrates that DECOR significantly outperforms existing approaches in terms of energy-efficiency and data overhead.
@article{haque_decor2025,
author = {Haque, Khandaker Foysal and Kong, Justin and Moore, Terrence J and Chen, Kevin and Restuccia, Francesco and Dagefu, Fikadu T},
title = {DECOR: Multi-Modal Decentralized Cluster-based Energy Efficient Covert Routing in HetNets},
journal = {mentis.info},
year = {2025}
} 2025
Journal
How to Poison an xApp: Dissecting Backdoor Attacks to Deep Reinforcement Learning in Open Radio Access Networks
Andrea Lacava, Stefano Maxenti, Leonardo Bonati, Salvatore D’Oro, Alina Oprea, Tommaso Melodia, Francesco Restuccia
Computer Networks (Under Review)
The development of Open Radio Access Network (RAN) cellular systems is being propelled by the integration of Artificial Intelligence (AI) techniques. While AI can enhance network performance, it expands the attack surface of the RAN. For instance, the need for datasets to train AI algorithms and the use of open interface to retrieve data in real time paves the way to data tampering during both training and inference phases. In this work, we propose MalO-RAN, a framework to evaluate the impact of data poisoning on O-RAN intelligent applications. We focus on AI-based xApps taking control decisions via Deep Reinforcement Learning (DRL), and investigate backdoor attacks, where tampered data is added to training datasets to include a backdoor in the final model that can be used by the attacker to trigger potentially harmful or inefficient pre-defined control decisions. We leverage an extensive O-RAN dataset collected on the Colosseum network emulator and show how an attacker may tamper with the training of AI models embedded in xApps, with the goal of favoring specific tenants after the application deployment on the network. We experimentally evaluate the impact of the SleeperNets and TrojDRL attacks and show that backdoor attacks achieve up to a 0.9 attack success rate. Moreover, we demonstrate the impact of these attacks on a live O-RAN deployment implemented on Colosseum, where we instantiate the xApps poisoned with MalO-RAN on an O-RAN-compliant Near-real-time RAN Intelligent Controller (RIC). Results show that these attacks cause an average network performance degradation of 87%. We pledge to share our complete codebase.
@article{lacava2025how,
author = {Andrea Lacava and Stefano Maxenti and Leonardo Bonati and Salvatore D’Oro and Alina Oprea and Tommaso Melodia and Francesco Restuccia},
title = {{How to Poison an xApp: Dissecting Backdoor Attacks to Deep Reinforcement Learning in Open Radio Access Networks}},
year = {2025}
}
2025
Journal
SCOPE: Cooperative Integrated Communications and Sensing for Material Classification at Sub-Terahertz Frequencies
K. F. Haque, X. Cantos-Roman, F. Meneghello, J. M. Jornet and F. Restuccia
IEEE Wireless Communications Letters
In this letter, we propose SCOPE—a novel, entropy-weighted ensembling approach for material classification at sub-Terahertz (THz) frequencies. Unlike existing methods that primarily use dedicated radars, SCOPE builds upon an integrated communication and sensing system and leverages information from both penetrating and reflected signals to enhance spatial resolution and detection accuracy across environments. We adopted spatial variability augmentation (SVA) to address the challenge of generalization across varying transmission distances and antenna gains. While most prior works are limited to radar systems or simulations, SCOPE is implemented and validated in a real sub-THz system working with a 10 GHz bandwidth. Our assessments across different sensing distances, antenna gains, and channel conditions demonstrate the efficacy of SCOPE, which reaches up to 99% accuracy in detecting five materials—glass, wood, metal, air, and plastic—outperforming existing techniques. To facilitate reproducibility, our dataset and code are available at: https://github.com/kfoysalhaque/SCOPE.
@article{haque2025scope,
author = {Haque, Khandaker Foysal and Cantos-Roman, Xavier and Meneghello, Francesca and Jornet, Josep Miquel and Restuccia, Francesco},
title = {SCOPE: Cooperative Integrated Communications and Sensing for Material Classification at Sub-Terahertz Frequencies},
journal = {IEEE Wireless Communications Letters},
year = {2025},
doi = {10.1109/LWC.2025.3576010},
publisher = {IEEE}
} 2025
Conference
MAGIC: Meta-Learning Adaptive Gesture Recognition with mmWave MIMO CSI
K. F. Haque, K. M. Rumman, A. Elyasi, F. Meneghello, F. Restuccia
Proc. of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (IEEE WoWMoM)
In this paper, we present MAGIC, a novel ap- proach to gesture recognition utilizing mmWave multiple-input multiple-output (MIMO) Channel State Information (CSI). Un- like existing mmWave gesture recognition methods that often rely on radar signals, MAGIC leverages CSI extracted from mmWave MIMO integrated sensing and communication (ISAC) systems. While advanced radar systems, such as those operating in frequency-modulated continuous wave (FMCW) mode, can achieve high frequency and spatial resolution, they typically require dedicated sensing infrastructure, which increases system complexity. In contrast, MAGIC utilizes high-granular CSI from orthogonal frequency-division multiplexing (OFDM) systems, enabling fine spatial, temporal, and frequency-domain informa- tion for robust gesture recognition. This eliminates the need for dedicated radar transceivers, simplifying the system and reducing transmission overhead. MAGIC employs a learning- based architecture, integrating a temporal convolutional net- work (TCN) to classify gestures by capturing long-range tem- poral dependencies. To address the critical challenge of domain adaptation in gesture recognition, we propose adaptive temporal embedding network (ATEN), a meta-learning framework that combines the temporal modeling capabilities of TCN with task- specific adaptation mechanisms. We evaluate MAGIC through a comprehensive data collection campaign involving two subjects performing 10 micro gestures across three different environ- ments, with synchronized video streams providing the ground truth. The proposed system achieves a baseline accuracy of 99.24% using TCN. The system continues to perform well – achieving up to 98.82% accuracy – when adapting to new do- mains using ATEN, outperforming other state-of-the-art domain adaptation methods by 14% on average.
@article{haque2025magic,
title={{MAGIC: Meta-Learning Adaptive Gesture Recognition with mmWave MIMO CSI}},
author={Haque, Khandaker Foysal and Rumman, K M and Elyasi, Arman and Meneghello, Francesca and Restuccia, Francesco},
journal={Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (IEEE WoWMoM)},
year={2025}
} 2025
Conference
Metacognitive Artificial Intelligence in Vision Foundation Models: Research Challenges
S. Rifat, A.Q.M. S. Sayyed, N. D. Bastian, and F. Restuccia
Workshop on Metacognitive Prediction of AI Behavior at SDM 2025
The adoption of Vision Foundation Models (VFM) in high-stakes scenarios has spurred the the demand for task-specific high-performance models. On the other hand, the lack of explainability of VFMs makes it challenging to ensure that these systems remain safe, reliable and resilient when encountering data that has a different distribution from the one encountered during training. Recently, approaches based on metacognition – the human ability to regulate cognitive processes – has emerged as a way to understand large models. This paper surveys the interdisciplinary connection between metacognition and state-of-the-art VFMs, and further examines its relationship with knowledge distillation (KD), a widely used technique in VFMs. The paper concludes by defining possible avenues for future research on the topic.
@article{rifat2025metacognitive,
author = {Shahriar Rifat and AMQ Sazzad Sayyed and Nathaniel D. Bastian and Francesco Restuccia},
journal = {Workshop on Metacognitive Prediction of AI Behavior (METACOG)},
title = {{Metacognitive Artificial Intelligence in Vision Foundation Models: Research Challenges}},
year = 2025
} 2025
Conference
SOAR: Semantic Multi-User MIMO Communications for Reliable Wireless Edge Computing
S. L.G. Contreras, K. F. Haque, F. Restuccia, and M. Levorato
Proc. of International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT)
Ensuring reliability in Mobile Edge Computing Sys- tems (MECS) is challenging due to wireless channel fluctuations, which affect packet delivery rate and real-time task performance. Existing methods, such as packet replication strategies and deep neural network (DNN) splitting, often lack task-specific adaptabil- ity, oversimplifying the effects of channel dynamics in its perfor- mance. To address this issues, we propose Semantic Offloading through Reliability (SOAR), a task-oriented multi-user MIMO framework for wireless edge computing executing vision tasks, e.g. object detection and image classification. SOAR pipeline uses distributional deep reinforcement learning (DDRL) agents with a multi-branched context-aware neural network. Two neural gates analyze onboard features to identify the context and contextual features, enabling a DDRL agent to optimize resource usage and task-specific packet-loss objectives. We evaluated SOAR in real- world vehicular system under line-of-sight (LoS) and non-line- of-sight (NLoS) propagation scenarios, SOAR reduces resource utilization by 35–40% compared to fixed antenna configuration benchmarks.
@article{haque2025magic,
title={{SOAR: Semantic Multi-User MIMO Communicationsfor Reliable Wireless Edge Computing}},
author={Contreras, Sharon L.G. and Haque, Khandaker Foysal and Restuccia, Francesco and Levorato, Marco },
journal={2025 21st International Conference on Distributed Computing in Smart Systems and the Internet of Things (DCOSS-IoT)},
year={2025}
} 2025
Conference
DEER: Simultaneous Multi-Modal Decentralized Energy Efficient Covert Routing
K. F. Haque, J. Kong, T. J. Moore, F. Restuccia, F. T. Dagefu
Proc. of IEEE Wireless Communications and Networking Conference (WCNC)
A fundamental challenge in covert routing is that meeting both covertness and throughput requirements often leads to increased transmit power, which can significantly elevate the overall energy consumption of the network. Therefore, it is important to achieve higher throughput and better energy efficiency while maintaining the required covertness. To this end, we propose a novel simultaneous multi-modal Decentralized Energy- Efficient covert Routing approach - DEER for a multi-hop heterogeneous network (HetNet). Unlike the prevailing single-modal approaches, DEER leverages the diversity of the available wireless communication technologies for simultaneous multi-modal routing. DEER aims to minimize the end-to-end total transmit power of the whole route in a decentralized fashion while maintaining the constraints on required throughput and covertness. DEER stems into two main steps: node-level optimization followed by network-level optimization using the proposed custom-tailored Dijkstra's based link state routing protocol to meet the constraints while minimizing the end-to-end total transmit power. We demonstrate by numerical analysis that DEER improves the energy efficiency by 23.5x and 2.9x times in comparison to the baseline single-modal and naive simultaneous multimodal approaches respectively.
@inproceedings{haque2025deer,
title={DEER: Simultaneous Multi-Modal Decentralized Energy Efficient Covert Routing},
author={Haque, Khandaker Foysal and Kong, Justin and Moore, Terrence J and Restuccia, Francesco and Dagefu, Fikadu T},
booktitle={2025 IEEE Wireless Communications and Networking Conference (WCNC)},
pages={1--6},
year={2025},
organization={IEEE}
} 2025
Journal
O-RAN xApps: Survey and Research Challenges
Arman Elyasi, Andrew Ashdown, K M Rumman, Francesco Restuccia
As the Open Radio Access Network (O-RAN) paradigm is transforming the cellular landscape, third-party Extensible Applications (xApps) deployed in the Near-Real-Time RAN Intelligent Controller will play a critical role in facilitating next-generation cellular networks. Such networks will rely heavily upon robust data-driven Artificial Intelligence and Machine Learning(AI/ML) solutions that can perform fine-grained RAN control at the millisecond timescale. As xApps represent the point of contact between these advanced AI/ML solutions and the RAN itself, understanding these important O-RAN applications is essential for researchers in the wireless domain and will be critical to the full-scale integration of AI/ML into the next generation of cellular networks. Motivated by a lack of relevant and up-to-date surveys on this topic, we provide a comprehensive overview of xApps, including important background information, current xApp development techniques, a discussion of xApps and AI/ML, state-of-the-art use cases, vulnerabilities in the xApp architecture, and other critical research challenges. We summarize the plethora of available information on this topic by highlighting the keyissues and presenting a cohesive message that provides the necessary back-ground to the reader and isolates the critical research issues that currently remain unaddressed. As such, this survey is a one-stop-shop for researchers seeking a solid grasp of the state-of-the-art and a clear outline of the central research challenges relevant to xApps in cellular networks.
@article{Elyasi_xApps,
author = {Arman Elyasi and Andrew Ashdown and K M Rumman and Francesco Restuccia},
title = {O-RAN xApps: Survey and Research Challenges},
journal = {Computer Networks},
year = {2025}
} 2025
Conference
DARDA: Domain-Aware Real-Time Dynamic Neural Network Adaptation
S. Rifat , J. Ashdown and F. Restuccia
Proc. of IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
Test Time Adaptation (TTA) has emerged as a practical solution to mitigate the performance degradation of Deep Neural Networks (DNNs) in the presence of corruption/ noise affecting inputs. Existing approaches in TTA continuously adapt the DNN, leading to excessive resource consumption and performance degradation due to accumulation of error stemming from lack of supervision. In this work, we propose Domain-Aware Real-Time Dynamic Adaptation (DARDA) to address such issues. Our key approach is to proactively learn latent representations of some corruption types, each one associated with a sub-network state tailored to correctly classify inputs affected by that corruption. After deployment, DARDA adapts the DNN to previously unseen corruptions in an unsupervised fashion by (i) estimating the latent representation of the ongoing corruption; (ii) selecting the sub-network whose associated corruption is the closest in the latent space to the ongoing corruption; and (iii) adapting DNN state, so that its representation matches the ongoing corruption. This way, DARDA is more resource efficient and can swiftly adapt to new distributions caused by different corruptions without requiring a large variety of input data. Through experiments with two popular mobile edge devices - Raspberry Pi and NVIDIA Jetson Nano - we show that DARDA reduces energy consumption and average cache memory footprint respectively by 1.74x and 2.64x with respect to the state of the art, while increasing the performance by 10.4%, 5.7% and 4.4% on CIFAR-10, CIFAR-100 and TinyImagenet.
@article{rifat2025darda,
title={DARDA: Domain-Aware Real-Time Dynamic Neural Network Adaptation},
author={Rifat, Shahriar and Ashdown, Jonathan and Restuccia, Francesco},
journal={Proceedings of IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
year={2025}
}
2025
Journal
BeamSense: Rethinking Wireless Sensing with MU-MIMO Wi-Fi Beamforming Feedback
K. F. Haque, M. Zhang, F. Meneghello, and F. Restuccia
Computer Networks
In this paper, we propose BeamSense, a completely novel approach to implement standard-compliant Wi-Fi sensing applications. Existing work leverages the manual extraction of the uncompressed channel state information (CSI) from Wi-Fi chips, which is not supported by the 802.11 standards and hence requires the usage of specialized equipment. On the contrary, BeamSense leverages the standard-compliant compressed beamforming feedback information (BFI) (beamforming feedback angles (BFAs)) to characterize the propagation environment. Conversely from the uncompressed CSI, the compressed BFAs (i) can be recorded without any firmware modification, and (ii) simultaneously captures the channels between the access point and all the stations, thus providing much better sensitivity. BeamSense features a novel cross-domain few-shot learning (FSL) algorithm for human activity recognition to handle unseen environments and subjects with a few additional data samples. We evaluate BeamSense through an extensive data collection campaign with three subjects performing twenty different activities in three different environments. We show that our BFAs-based approach achieves about 10% more accuracy when compared to CSI-based prior work, while our FSL strategy improves accuracy by up to 30% when compared with state-of-the-art cross-domain algorithms. Additionally, to demonstrate its versatility, we apply BeamSense to another smart home application – gesture recognition – achieving over 98% accuracy across various orientations and subjects. We share the collected datasets and BeamSense implementation code for reproducibility – https://github.com/kfoysalhaque/BeamSense
@article{haque2025beamsense,
title={Beamsense: Rethinking wireless sensing with mu-mimo wi-fi beamforming feedback},
author={Haque, Khandaker Foysal and Zhang, Milin and Meneghello, Francesca and Restuccia, Francesco},
journal={Computer Networks},
volume={258},
pages={111020},
year={2025},
publisher={Elsevier}
} 2025
Conference
On the Adversarial Vulnerability of Label-Free Test-Time Adaptation
S. Rifat, J. Ashdown, M. J. De Lucia, A. Swami, F. Restuccia
Proc. of International Conference on Learning Representations (ICLR)
Despite the success of Test-time adaptation (TTA), recent work has shown that adding relatively small adversarial perturbations to a limited number of samples leads to significant performance degradation. Therefore, it is crucial to rigorously evaluate existing TTA algorithms against relevant threats and implement appropriate security countermeasures. Importantly, existing threat models assume test-time samples will be labeled, which is impractical in real-world scenarios. To address this gap, we propose a new attack algorithm that does not rely on access to labeled test samples, thus providing a concrete way to assess the security vulnerabilities of TTA algorithms. Our attack design is grounded in theoretical foundations and can generate strong attacks against different state of the art TTA methods. In addition, we show that existing defense mechanisms are almost ineffective, which emphasizes the need for further research on TTA security. Through extensive experiments on CIFAR10-C, CIFAR100-C, and ImageNet-C, we demonstrate that our proposed approach closely matches the performance of state-of-the-art attack benchmarks, even without access to labeled samples. In certain cases, our approach generates stronger attacks, e.g., more than 4% higher error rate on CIFAR10-C.
@article{rifat2025on,
author = {Shahriar Rifat and Jonathan Ashdown and Michael J. De Lucia and Ananthram Swami and Francesco Restuccia },
journal = {Proceedings of International Conference on Learning Representations (ICLR)},
title = {{On the Adversarial Vulnerability of Label-Free Test-Time Adaptation}},
year = 2025
} 2025
Conference
PhyDNNs: Bringing Deep Neural Networks to the Physical Layer
M. Abdi and K. F. Haque and F. Meneghello and J. Ashdown and F. Restuccia
Proc. of IEEE International Conference on Computer Communications (IEEE INFOCOM)
Emerging applications require mobile devices to continuously execute complex deep neural networks (DNNs). While mobile edge computing (MEC) may reduce the computation burden of mobile devices, it exhibits excessive latency as it relies on encapsulating and decapsulating frames through the network protocol stack. To address this issue, we propose PhyDNNs, an approach where DNNs are modified to operate directly at the physical layer (PHY), thus significantly decreasing latency, energy consumption, and network overhead. Conversely from recent work in Joint Source and Channel Coding (JSCC), PhyDNNs adapt already trained DNNs to work at the PHY. To this end, we developed a novel information-theoretical framework to fine-tune PhyDNNs based on the trade-off between communication efficiency and task performance. We have prototyped PhyDNNs with an experimental testbed using a Jetson Orin Nano as the mobile device and two USRP software-defined radios (SDRs) for wireless communication. We evaluated PhyDNNs performance considering various channel conditions, DNN models, and datasets. We also tested PhyDNNs on the Colosseum network emulator considering two different propagation scenarios. Experimental results show that PhyDNNs can reduce the end-to-end inference latency, amount of transmitted data, and power consumption by up to 48×, 1385×, and 13×while keeping the accuracy within 7% of the state-of-the-art approaches. Moreover, we show that PhyDNNs experience 4.3 times less latency than the most recent JSCC method while incurring in only 1.79% performance loss. For replicability, we shared the source code for the PhyDNNs implementation.
@article{abdi2025phydnns,
author = {Mohammad Abdi and Khandaker Foysal Haque and Francesca Meneghello and Jonathan Ashdown and Francesco Restuccia},
journal = {Proceedings of IEEE International Conference on Computer Communications (INFOCOM)},
title = {{PhyDNNs: Bringing Deep Neural Networks to the Physical Layer}},
year = 2025
} 2025
Conference
How to BREAK MU-MIMO Precoding in IEEE 802.11 Wi-Fi Networks
F. Meneghello, F. Gringoli, M. Cominelli, M. Rossi and F. Restuccia
Proc. of IEEE International Conference on Computer Communications (IEEE INFOCOM) — Best Paper Award
This work reveals a critical vulnerability of the Wi-Fi standard that if unaddressed, might lead to serious security issues and compromise the performance of several billions of Wi-Fi devices. Specifically, this paper introduces and validates with commercial off-the-shelf Wi-Fi devices a new Beamforming Report Eavesdropping Attack (BREAK), which leverages the MU-MIMO channel estimation procedure used by Wi-Fi to decrease the throughput of the entire network without being detected. Through rigorous mathematical optimization, we compute the poisoned feedback that a BREAK adversary eavesdropping the feedback angles needs to send to the access point to reduce the throughput of legitimate users. Through extensive experimental evaluation with Beamformee commercial Wi-Fi routers and smartphones in multiple network configurations, we show that through BREAK, an adversary may victim beamforming feedback angles decrease the throughput at legitimate stations by 65% modifying only about 17% of its feedback without being detected. For replicability, we shared the code implementing the attack together NDP for channel sounding with the modified firmware to be used at the adversary node.
@article{meneghello2025how,
author = {Meneghello, Francesca and Gringoli, Francesco and Cominelli, Marco and Rossi, Michele and Restuccia, Francesco},
journal = {Proceedings of IEEE International Conference on Computer Communications (INFOCOM)},
title = {{How to BREAK MU-MIMO Precoding in IEEE 802.11 Wi-Fi Networks}},
year = 2025
} 2025
Conference
m3MIMO: An 8×8 mmWave Multi-User MIMO Testbed for Wireless Research
K. F. Haque , F. Meneghello, K. M. Rumman and F. Restuccia
Proc. of ACM Workshop on Wireless Network Testbeds, Experimental Evaluation & Characterization (ACM WINTECH)
In this paper, we present m3MIMO a mmWave fully-digital multi-user multi-input multi-output (MU-MIMO) testbed for advanced wireless research. m3MIMO operates in the 57-64 GHz frequency range and supports up to 1 GHz of bandwidth enabling large data multiplexing in the frequency domain through orthogonal frequency-division multiplexing (OFDM). The testbed features three custom-designed Zynq UltraScale+ RFSoC-based Software Defined Radios (SDRs) empowered with the Pi-Radio fully digital transceivers. Two of these SDRs support eight transmit and receive streams each (8 × 8 MIMO), while the third SDR supports up to four channels. m3MIMO supports three different communication modes: (i) point-to-point (P2P) transmissions; (ii) single-user multi-input multi-output (SU-MIMO), where multiple streams are transmitted to a single end-device; and (iii) MU-MIMO, where two devices are simultaneously served by a single transmitter. To showcase the m3MIMO's versatility, we present two research use cases: tracking-based beamforming and mmWave-based sensing. We will open-source the m3MIMO code along with the relevant use-case datasets, facilitating further analysis.
@inproceedings{haque2024m3mimo,
title={m3MIMO: An 8$\times$ 8 mmWave Multi-User MIMO Testbed for Wireless Research},
author={Haque, Khandaker Foysal and Meneghello, Francesca and Rumman, KM and Restuccia, Francesco},
booktitle={Proceedings of the 30th Annual International Conference on Mobile Computing and Networking},
pages={1922--1929},
year={2024}
} 2024
Conference
HyperAdv: Dynamic Defense Against Adversarial Radio Frequency Machine Learning Systems
M. Zhang , M. De Lucia, A. Swami, J. Ashdown, K. Turck and F. Restuccia
Proc. of IEEE Military Communications Conference (IEEE MILCOM)
2024
Conference
Blind Spectrum Sensing for Dynamic Spectrum Awareness in Next G Open Radio Access Networks
A. Ashdown , J. Kelly, J. Ashdown, I. Kurzrock, C. Turck, and F. Restuccia
Proc. of IEEE Military Communications Conference (IEEE MILCOM)
2024
Conference
ADA: Adversarial Dynamic Test Time Adaptation in Radio Frequency Machine Learning Systems
S. Rifat, M. De Lucia, A. Swami, J. Ashdown, K. Turck, F. Restuccia
Proc. of IEEE Military Communications Conference (IEEE MILCOM)
2024
Conference
Integrated Sensing and Communication for Efficient Edge Computing
K. F. Haque , F. Meneghello and F. Restuccia
Proc. of International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob)
Emerging mobile virtual reality (VR) systems are required to continuously perform complex computer vision tasks needing computational power that is excessive for mobile devices. Thus, techniques based on wireless edge computing (WEC) have been recently proposed. However, existing WEC methods require the transmission and processing of a high amount of video data which may ultimately saturate the wireless link. In this paper, we propose a novel sensing-assisted edge computing (ISAC-EC) approach to address this issue. ISAC-EC leverages knowledge about the physical environment to reduce the end-to-end latency and overall computational burden by transmitting to the edge server only the relevant data for the delivery of the service. Our intuition is that the transmission of the portion of the video frames where there are no changes with respect to the previous frames can be avoided. Through wireless sensing, only the part of the frames where any environmental change is detected is transmitted and processed. We evaluated ISAC-EC by using a 10K 360°camera with a Wi-Fi 6 sensing system operating at 160 MHz and performing localization and tracking. Experimental results show that ISAC-EC reduces both the channel occupation and end-to-end latency by more than 90% while improving the instance segmentation and object detection performance with respect to state-of-the-art WEC approaches. For reproducibility purposes, we pledge to share our dataset and code repository.
@INPROCEEDINGS{10770523,
author={Haque, Khandaker Foysal and Meneghello, Francesca and Restuccia, Francesco},
booktitle={2024 20th International Conference on Wireless and Mobile Computing, Networking and Communications (WiMob)},
title={Integrated Sensing and Communication for Efficient Edge Computing},
year={2024},
volume={},
number={},
pages={611-614},
keywords={Wireless communication;Location awareness;Performance evaluation;Wireless sensor networks;Computer vision;Image edge detection;Virtual reality;Reproducibility of results;Servers;Edge computing},
doi={10.1109/WiMob61911.2024.10770523}}
2024
Conference
Resilience of Entropy Model in Distributed Neural Networks
M. Zhang, M. Abdi, S. Rifat, F. Restuccia
Proc. of European Conference on Computer Vision (ECCV)
Distributed deep neural networks (DNNs) have emerged as a key technique to reduce communication overhead without sacrificing performance in edge computing systems. Recently, entropy coding has been introduced to further reduce the communication overhead. The key idea is to train the distributed DNN jointly with an entropy model, which is used as side information during inference time to adaptively encode latent representations into bit streams with variable length. To the best of our knowledge, the resilience of entropy models is yet to be investigated. As such, in this paper we formulate and investigate the resilience of entropy models to intentional interference (e.g., adversarial attacks) and unintentional interference (e.g., weather changes and motion blur). Through an extensive experimental campaign with 3 different DNN architectures, 2 entropy models and 4 rate-distortion trade-off factors, we demonstrate that the entropy attacks can increase the communication overhead by up to 95%. By separating compression features in frequency and spatial domain, we propose a new defense mechanism that can reduce the transmission overhead of the attacked input by about 9% compared to unperturbed data, with only about 2% accuracy loss. Importantly, the proposed defense mechanism is a standalone approach which can be applied in conjunction with approaches such as adversarial training to further improve robustness. Code will be shared for reproducibility.
@article{zhang2024resilience,
title={Resilience of Entropy Model in Distributed Neural Networks},
author={Zhang, Milin and Abdi, Mohammad and Rifat, Shahriar and Restuccia, Francesco},
journal={arXiv preprint arXiv:2403.00942},
year={2024}
}
2024
Journal
Evaluating the Impact of Channel Feedback Quantization and Grouping in IEEE 802.11 MIMO Wi-Fi Networks
F. Meneghello , K. F. Haque and F. Restuccia
IEEE Wireless Communications Letters
In this letter, we shed light on the impact of mimo beamforming feedback quantization and ofdm sub-channel grouping on communication performance, for IEEE 802.11ac/ax Wi-Fi networks. We performed an extensive data collection campaign with commercial Wi-Fi devices deployed in different propagation environments considering several network configurations. Our objective is to provide a benchmark for research on efficient feedback compression mechanisms and to enable further analysis. As such, we pledge to share the datasets we collected and the emulation framework we developed.
@article{meneghello2024evaluating,
title={Evaluating the Impact of Channel Feedback Quantization and Grouping in IEEE 802.11 MIMO Wi-Fi Networks},
author={Meneghello, Francesca and Haque, Khandaker Foysal and Restuccia, Francesco},
journal={IEEE Wireless Communications Letters},
year={2024},
publisher={IEEE}
}
2024
Conference
FedLoRa: IoT Spectrum Sensing Through Fast and Energy-Efficient Federated Learning in LoRa Networks
F. Busacca, S. Mangione, G. Neglia, I. Tinnirello, S. Palazzo, F. Restuccia
IEEE 21st International Conference on Mobile Ad-Hoc and Smart Systems (MASS)
Effective spectrum sensing is quintessential to decrease spectrum congestion across time, space and frequency in Internet of Things (IoT) networks. To circumvent the severe bandwidth constraints of IoT networks, federated machine learning (FML) can be used, but it is still unclear whether FML can be successfully performed in resource-constrained bandwidth-limited IoT networks. In this paper, we demonstrate for the first time that FML can tolerate losses up to a certain percentage and still converge. Then, we leverage this key result to design FedLoRa, an optimization networks for LoRa that is (i) fast, as it reduces the FML round time in comparison with other resource allocation schemes; (ii) energy-efficient, as the time reduction does not imply a higher energy consumption. The key idea is to balance the network load over the available spreading factors, and to exploit sequential polling of nodes to maximize the number of simultaneous non-interfering transmissions, leading to a shorter FML round time. As the problem is NP-Hard, we provide an approximation algorithm. We evaluate the performance of FedLoRa through experimental evaluation on the Colosseum channel emulator, as well as with real-world data collection with off-the-shelf LoRa devices in an 5kmx5km urban setting in Portland, Maine. Our results show that FedLoRa reduces the round time by up to about 35%, as compared to the baselines.
@inproceedings{busacca2024fedlora,
title={FedLoRa: IoT Spectrum Sensing Through Fast and Energy-Efficient Federated Learning in LoRa Networks},
author={Busacca, Fabio and Mangione, Stefano and Neglia, Giovanni and Tinnirello, Ilenia and Palazzo, Sergio and Restuccia, Francesco},
booktitle={2024 IEEE 21st International Conference on Mobile Ad-Hoc and Smart Systems (MASS)},
pages={295--303},
year={2024},
organization={IEEE}
}
2024
Journal
WHACK: Adversarial Beamforming in MU-MIMO Through Compressed Feedback Poisoning
F. Meneghello, F. Restuccia, M. Rossi
IEEE Transactions on Wireless Communications
Multi-user MIMO is a key component of modern wireless networks. As such, investigating the related security weaknesses is a compelling necessity. A major issue unveiled by existing work is that adversaries can “poison” the channel information feedback reported to the beamformer to decrease the performance experienced by a legitimate user. Prior work, however, assumes that the feedback is reported in an uncompressed fashion, which is not the case in current wireless standards such as Wi-Fi or 5G. In this work, we first show that assuming uncompressed feedback leads to overestimating the attack effectiveness by up to 60%. Next, we formulate ACFP (Adversarial Compressed Feedback Problem), a novel non-convex constrained optimization problem to find the compressed feedback that maximizes a victim’s bit error rate (BER) while satisfying maximum power constraints. We propose WHACK (Wireless Harmful Adversarial Compressed feedbacK), a new algorithm to solve ACFP and find the malicious compressed feedback based on the convexity of the objective function and constraint using a nonlinear conjugate gradient method. WHACK has been prototyped and extensively evaluated with off-the-shelf Wi-Fi devices. Experimental results show that it maximizes the victim’s BER, while modifying less than 60% of the feedback. Our dataset and code are available.
@article{meneghello2024whack,
title={WHACK: Adversarial Beamforming in MU-MIMO Through Compressed Feedback Poisoning},
author={Meneghello, Francesca and Restuccia, Francesco and Rossi, Michele},
journal={IEEE Transactions on Wireless Communications},
year={2024},
publisher={IEEE}
}
2024
Conference
A Digitally Reconfigurable Low-Noise Amplifier with Robust Input Impedance for Machine Learning-Based Receiver Optimizations
M. Liu, D. Das, M. Abdi, F. Restuccia, M. Onabajo
IEEE 67th International Midwest Symposium on Circuits and Systems (MWSCAS)
The surge in demand for wireless connectivity has strongly incentivized advancements in reconfigurable radio frequency (RF) circuits. Although these circuits offer promising opportunities for machine learning (ML)- based optimization when devices are operating in the field, there is still an increasing need to adjust performance and power consumption over wider ranges, especially to dynamically minimize receiver power consumption when possible. In this paper, we present a novel low-noise amplifier (LNA) topology to dynamically scale power and performance to facilitate the realization of real-time ML methods for receiver optimization. This LNA is designed to avoid any significant input impedance matching degradation despite of a wide bias current tuning range to scale the gain, noise figure (NF) and input third-order intermodulation intercept point (IIP3). Simulations of the 2.4 GHz LNA design in 65 nm CMOS technology show its digitally-programmable gain from 17.07 dB to 28.15 dB, NF from 2.56 dB 5.18 dB, and IIP3 from -14.98 dBm to -9.85 dBm, while maintaining consistent input impedance matching with S11<−13 dB.
@inproceedings{liu2024digitally,
title={A Digitally Reconfigurable Low-Noise Amplifier with Robust Input Impedance for Machine Learning-Based Receiver Optimizations},
author={Liu, Minghan and Das, Diptashree and Abdi, Mohammad and Restuccia, Francesco and Onabajo, Marvin},
booktitle={2024 IEEE 67th International Midwest Symposium on Circuits and Systems (MWSCAS)},
pages={392--396},
year={2024},
organization={IEEE}
} 2024
Conference
OffloaDNN: Shaping DNNs for Scalable Offloading of Computer Vision Tasks at the Edge
C. Puligheddu, N. Varshney, T. Hassan, J. Ashdown, F. Restuccia, C. Fabiana Chiasserini
IEEE 44th International Conference on Distributed Computing Systems (ICDCS)
Emerging mobile applications often require the execution of computer vision (CV) tasks based on compute-and memory-intensive deep neural networks (DNNs). Although offloading CV tasks to edge servers can decrease resource consumption at the mobile devices, it poses the challenge of handling multiple concurrent tasks with limited computing and memory capacity. In stark opposition with the existing state of the art, we tackle this challenge by jointly optimizing (i) the utilization of resources at the edge, among which memory - so far widely overlooked - and the radio resources used for task offloading; (ii) which and how many offloaded tasks should be executed; and (iii) the structure of the DNNs. First, we formulate the DNN for scalable Offloading of Tasks (DOT) problem, prove that it is NP-hard, and envision a weighted-tree-based heuristic solution, named OffloaDNN, that efficiently solves the DOT problem. We evaluate OffloaDNN through extensive numerical analysis using state-of-the-art image classification ResNet-18, as well as real-world experiments on the Colosseum emulator. The numerical results show that, in small-scale scenarios, OffloaDNN matches the optimum very closely, and, in larger-scale scenarios, increases the number of admitted offloaded tasks by 26.9 % with respect to the state of the art, while saving 82.5 % memory and 77.4% per-inference computing time. The numerical results are confirmed by the real-world validation on Colosseum.
@inproceedings{puligheddu2024offloadnn,
title={OffloaDNN: Shaping DNNs for Scalable Offloading of Computer Vision Tasks at the Edge},
author={Puligheddu, Corrado and Varshney, Nancy and Hassan, Tanzil and Ashdown, Jonathan and Restuccia, Francesco and Chiasserini, Carla Fabiana},
booktitle={2024 IEEE 44th International Conference on Distributed Computing Systems (ICDCS)},
pages={624--634},
year={2024},
organization={IEEE}
} 2024
Conference
Poster: Transport-Aware Resource Block Allocation in 5G Slicing
A. Pinto, TB Hassan, F. Restuccia, F. Esposito
IEEE 10th International Conference on Network Softwarization (NetSoft)
Network slicing in next-generation wireless networks is a mechanism that ensures isolation and optimal resource distribution among users under constraints imposed by limited information availability at Base Stations (BS). While several strategies to optimize wireless slices have been proposed, this poster introduces an approach to resource allocation in 5G networks that integrates detailed end-to-end flow-level data at the transport layer to refine the NextG resource block allocation process. In particular, we present an architecture that leverages the host’s network stack and a novel in-network processing and scheduling component to optimize the dynamic resource allocation process in slicing. The proposed system design, illustrated through a comprehensive system architecture, aims to distribute Resource Blocks (RBs) effectively, accounting for the unique transport-level metrics of each flow. We present some initial evaluation results with an event-driven simulator, reflecting diverse traffic types and leveraging the Rayleigh fading channel model to simulate dynamic user movement. Our findings demonstrate a promising improvement in flow completion times and a reduction in packet loss, compared to traditional allocation methods such as Round Robin and Proportional Fair schemes, typically deployed in 5G production networks.
@inproceedings{pinto2024poster,
title={Poster: Transport-Aware Resource Block Allocation in 5G Slicing},
author={Pinto, Andrea and Hassan, Tanzil Bin and Restuccia, Francesco and Esposito, Flavio},
booktitle={2024 IEEE 10th International Conference on Network Softwarization (NetSoft)},
pages={322--323},
year={2024},
organization={IEEE}
}
2024
Patent
Private 5G Cellular Connectivity as a Service Through Full-Stack Wireless Steganographyation
T. Melodia, L. Bonati, S. D’Oro and F. Restuccia
U.S. Patent No.: 17/316,773
2024
Patent
Methods for Multi-Access Edge Computing Network Slicing in 5G Networks
S. D’Oro, T. Melodia, F. Restuccia and L. Bonati
U.S. Patent No.: 17/365,070
2024
Conference
Stitching the Spectrum: Semantic Spectrum Segmentation with Wideband Signal Stitching
D. Uvaydov, M. Zhang, CP. Robinson, S. D’Oro, T. Melodia, F. Restuccia
Proc. of IEEE International Conference on Computer Communications (INFOCOM).
Spectrum has become an extremely scarce and congested resource. As a consequence, spectrum sensing enables the coexistence of different wireless technologies in shared spectrum bands. Most existing work requires spectrograms to classify signals. Ultimately, this implies that images need to be continuously created from I/Q samples, thus creating unacceptable latency for real-time operations. In addition, spectrogram-based approaches do not achieve sufficient granularity level as they are based on object detection performed on pixels and are based on rectangular bounding boxes. For this reason, we propose a completely novel approach based on semantic spectrum segmentation, where multiple signals are simultaneously classified and localized in both time and frequency at the I/Q level. Conversely from the state-of-the-art computer vision algorithm, we add non-local blocks to combine the spatial features of signals, and thus achieve better performance. In addition, we propose a novel data generation approach where a limited set of easy-to-collect real-world wireless signals are "stitched together" to generate large-scale, wideband, and diverse datasets. Experimental results obtained on multiple testbeds (including the Arena testbed) using multiple antennas, multiple sampling frequencies, and multiple radios over the course of 3 days show that our approach classifies and localizes signals with a mean intersection over union (IOU) of 96.70% across 5 wireless protocols while performing in real-time with a latency of 2.6 ms. Moreover, we demonstrate that our approach based on non-local blocks achieves 7% more accuracy when segmenting the most challenging signals with respect to the state-of-the-art U-Net algorithm. We will release our 17 GB dataset and code.
@article{uvaydov2024stitching,
title={Stitching the Spectrum: Semantic Spectrum Segmentation with Wideband Signal},
author={Uvaydov, Daniel and Zhang, Milin and Robinson, Clifton Paul and D'Oro, Salvatore and Melodia, Tommaso and Restuccia, Francesco},
journal={arXiv preprint arXiv:2402.03465},
year={2024}
}
2024
Conference
RF-SoC Platforms for RF-AI Spectrum Perception
H. Weerasooriya, G. Rathnasekara, F. Restuccia, A. Madanayake
International Applied Computational Electromagnetics Society Symposium (ACES)
Sensing and perception of RF signals using AI is fundamental for future wireless systems. This paper describes measurable progress towards a truly wideband RF-AIIML spectrum perception testbed. We provide an approach for processing the 6–7 GHz range using 8-phase multirate processing on an RF-SoC. Up to 16 such 1 GHz channels can be accommodated on a ZCU-1275/1285 RF-SoC leading to an aggregate bandwidth of 16 GHz across the FRI/FR3 range. We aim to parallelize 16 GHz of aggregate bandwidth across the 100 MHz - 24 GHz spectrum into 512 data paths each having 32 MHz of bandwidth, such that a variety of AIIML based signal perception can operate in a massively parallel architecture on the FPGA.
@inproceedings{weerasooriya2024rf,
title={RF-SoC Platforms for RF-AI Spectrum Perception},
author={Weerasooriya, Hasitha and Rathnasekara, Gayani and Restuccia, Francesco and Madanayake, Arjuna},
booktitle={2024 International Applied Computational Electromagnetics Society Symposium (ACES)},
pages={1--2},
year={2024},
organization={IEEE}
}
2024
Journal
Toward Wireless System and Circuit Co-Design for the Internet of Self-Adaptive Things
D. Das, M. Abdi, M. Liu, M. Onabajo, F. Restuccia
IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)
The deployment of a growing number of devices in Internet of Things (IoT) networks implies that uninterrupted and seamless adaptation of wireless communication parameters (e.g., carrier frequency, bandwidth and modulation) will become essential. To utilize wireless devices capable of switching several communication parameters requires real-time self-optimizations at the radio frequency integrated circuit (RFIC) level based on system level performance metrics during the processing of complex modulated signals. This article introduces a novel design verification approach for reconfigurable RFICs based on end-to-end wireless system-level performance metrics while operating in a dynamically changing communication environment. In contrast to prior work, this framework includes two modules that simulate a wireless channel and decode waveforms. These are connected to circuit-level modules that capture device- and circuit-level non-idealities of RFICs for design validation and optimization, such as transistor noises, intermodulation/harmonic distortions, and memory effects from parasitic capacitances. We demonstrate this framework with a receiver (RX) consisting of a reconfigurable complementary metal-oxide semiconductor (CMOS) low-noise amplifier (LNA) designed at the transistor level, a behavioral model of a mixer, and an ideal filter model. The seamless integration between system-level wireless models with circuit-level and behavioral models (such as VerilogA-based models) for RFIC blocks enables to preemptively evaluate circuit and system designs, and to optimize for different communication scenarios with adaptive circuits having extensive tuning ranges. An exemplary case study is presented, in which simulation results reveal that the LNA power consumption can be reduced up to 16x depending on system-level requirements.
@inproceedings{das2024toward,
title={Toward Wireless System and Circuit Co-Design for the Internet of Self-Adaptive Things},
author={Das, Diptashree and Abdi, Mohammad and Liu, Minghan and Onabajo, Marvin and Restuccia, Francesco},
booktitle={2024 IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)},
pages={540--547},
year={2024},
organization={IEEE}
}
2024
Conference
Det-RAN: Data-driven cross-layer real-time attack detection in 5G open RANs
A. Scalingi, S. D’Oro, F. Restuccia, T. Melodia, D. Giustiniano
Proc. of IEEE International Conference on Computer Communications (INFOCOM).
Fifth generation (5G) and beyond cellular networks are vulnerable to security threats, primarily due to the lack of integrity protection in the Radio Resource Control (RRC) layer. In order to address this problem, we propose a real-time anomaly detection framework that leverages the concept of distributed applications in 5G Open RAN networks. Specifically, we identify Physical Layer (PHY) features that can generate a reliable fingerprint, infer in a novel way the time of arrival of uplink packets lacking integrity protection, and handle cross-layer features. By identifying legitimate message sources and detecting suspicious activities through an Artificial Intelligence (AI) design, we demonstrate that Open RAN-based applications that run at the edge can be designed to provide additional security to the network. Our solution is first validated in extensive emulation environments achieving over 85% accuracy in predicting potential attacks on unseen test scenarios. We then integrate our approach into a real-world prototype with a large channel emulator to assess its real-time performance and costs. Our solution meets the low-latency real-time constraints of 2 ms, making it well-suited for real-world deployments.
@inproceedings{scalingi2024det,
title={Det-RAN: Data-driven cross-layer real-time attack detection in 5G open RANs},
author={Scalingi, Alessio and D’Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso and Giustiniano, Domenico},
booktitle={IEEE INFOCOM 2024-IEEE Conference on Computer Communications},
pages={41--50},
year={2024},
organization={IEEE}
}
2024
Patent
Embedded networked deep learning for implanted medical devices
D. Uvaydov, R. Guida, F. Restuccia, T. Melodia
U.S. Patent No.: 17/176,229
2024
Journal
Sub-6 GHz Energy Detection-Based Fast On-Chip Analog Spectrum Sensing with Learning-Driven Signal Classification
A. Mittal, M. Zhang, T. Gourousis, Z. Zhang, Y. Fei, M. Onabajo, F. Restuccia, A. Shrivastava
IEEE Internet of Things Journal
Cognitive communication utilizes transient openings in the spectrum to communicate opportunistically, which is a promising technique to enable more efficient spectrum usage in an increasingly congested spectrum environment. We aim to address two main challenges associated with cognitive communication: 1) spectrum sensing should be fast and energy efficient for processing a large bandwidth in a short time and 2) the spectrum sensing approach should be able to simultaneously recognize multiple signals that are present. In this article, we propose to address these challenges with a novel design framework that consists of a fast on-chip spectrum sensing in conjunction with a novel learning-based spectrum analysis model at the edge to enhance the optimizations for spectrum agility. We first utilize a model of a programmable analog-based high-quality factor (Q) on-chip spectrum sensor that is capable of scanning the sub-6-GHz band to detect the spectrum usage in less than 1 μ s. The proposed spectrum sensor also enhances the energy efficiency of the sensing. To complement the on-chip spectrum sensor, a deep learning (DL) model is deployed for a fine-grained signal detection between channels in the 400 MHz to 6-GHz range, which is intended to be executed on edge devices. Simulation results show that the DL model can detect multiple different modulated signals with a mean Intersection-over-Union (IoU) of 86.8% in highly variable bandwidth and center frequency scenarios. Finally, we present a system-level model of our framework to demonstrate the spectrum sensing and classification in the sub-6-GHz frequency band.
@article{mittal2024sub,
title={Sub-6 GHz Energy Detection-Based Fast On-Chip Analog Spectrum Sensing with Learning-Driven Signal Classification},
author={Mittal, Ankit and Zhang, Milin and Gourousis, Thomas and Zhang, Ziyue and Fei, Yunsi and Onabajo, Marvin and Restuccia, Francesco and Shrivastava, Aatmesh},
journal={IEEE Internet of Things Journal},
year={2024},
publisher={IEEE}
}
2024
Conference
BFA-Sense: Learning Beamforming Feedback Angles for Wi-Fi Sensing
K. Foysal Haque, F. Meneghello, and F. Restuccia
IEEE International Conference on Pervasive Computing and Communications Workshops
In this paper, we propose BFA-Sense, a completelynovel approach to implement standard-compliant Wi-Fi sensingapplications. Wi-Fi sensing enables game-changing applicationsin remote healthcare, home entertainment, and home surveil-lance, among others. However, existing work leverages the man-ual extraction of the uncompressed channel state information(CSI) from Wi-Fi chips, which is not supported by the 802.11standard-compliant devices and hence requires the use of special-ized equipment. On the contrary, BFA-Sense leverages the com-pressed beamforming feedback angles (BFAs) transmitted dur-ing the standard-compliant sounding procedure to characterizethe propagation environment. Conversely from the uncompressedCSI, BFAs (i) can be recorded without any firmware modifica-tion, and (ii) allows a single monitor device to simultaneouslycapture the channels between the access point and all the stations,thus providing much better sensitivity. We evaluate BFA-Sensethrough an extensive data collection campaign with three subjectsperforming twenty different activities in three different environ-ments. We assess the cross-domain adaptability of BFA-Sensethrough embedding learning for tackling unseen environmentswith a few samples from the new environment. The results showthat the proposed BFAs-based approach achieves about 11% moreaccuracy when compared to CSI-based prior work.
@inproceedings{haque2024bfa,
title={BFA-Sense: Learning Beamforming Feedback Angles for Wi-Fi Sensing},
author={Haque, Khandaker Foysal and Meneghello, Francesca and Restuccia, Francesco},
booktitle={2024 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops)},
pages={575--580},
year={2024},
organization={IEEE}
} 2024
Patent
Methods for the enforcement of network slicing policies in virtualized cellular networks
DORO. Salvatore, F. Restuccia, T. Melodia
U.S. Patent No.: 17/424,522
2024
Patent
Deep learning-based polymorphic platform
T. Melodia, F. Restuccia
U.S. Patent No.: 17/604,476
2024
Conference
AXI-REALM: A Lightweight and Modular Interconnect Extension for Traffic Regulation and Monitoring of Heterogeneous Real-Time SoCs
T. Benz, A. Ottaviano, R. Balas, A. Garofalo, F. Restuccia, A. Biondi, L. Benini
2024 Design, Automation & Test in Europe Conference & Exhibition (DATE)
The increasing demand for heterogeneous functionality in the automotive industry and the evolution of chip manufac-turing processes have led to the transition from federated to integrated critical real-time embedded systems (CRTESs). This leads to higher integration challenges of conventional timing predictability techniques due to access contention on shared resources, which can be resolved by providing system-level observability and controllability in hardware. We focus on the interconnect as a shared resource and propose AXI-REALM, a lightweight, modular, and technology - independent real-time extension to industry-standard AXI4 interconnects, available open-source. AXI-REALM uses a credit-based mechanism to distribute and control the bandwidth in a multi-subordinate system on periodic time windows, proactively prevents denial of service from malicious actors in the system, and tracks each manager's access and interference statistics for optimal budget and period selection. We provide detailed performance and implementation cost assessment in a 12nm node and an end-to-end functional case study implementing AXI-REALM into an open-source Linux-capable RISC-V SoC. In a system with a general-purpose core and a hardware accelerator's DMA engine causing interference on the interconnect, AXI-REALM achieves fair bandwidth distribution among managers, allowing the core to recover 68.2 % of its performance compared to the case without contention. Moreover, near-ideal performance (above 95 %) can be achieved by distributing the available bandwidth in favor of the core, improving the worst-case memory access latency from 264 to below eight cycles. Our approach minimizes buffering compared to other solutions and introduces only 2.45 % area overhead compared to the original SoC.
@inproceedings{benz2024axi,
title={AXI-REALM: A Lightweight and Modular Interconnect Extension for Traffic Regulation and Monitoring of Heterogeneous Real-Time SoCs},
author={Benz, Thomas and Ottaviano, Alessandro and Balas, Robert and Garofalo, Angelo and Restuccia, Francesco and Biondi, Alessandro and Benini, Luca},
booktitle={2024 Design, Automation \& Test in Europe Conference \& Exhibition (DATE)},
pages={1--6},
year={2024},
organization={IEEE}
}
2024
Journal
SDR-LoRa, an open-source, full-fledged implementation of LoRa on Software-Defined Radios: Design and potential exploitation
F. Busacca, S. Mangione, S. Palazzo, F. Restuccia, and I. Tinnirello
Computer Networks (COMNET)
In this paper, we present SDR-LoRa, an open-source, full-fledged Software Defined Radio (SDR) implementation of a LoRa transceiver. First, we conduct a thorough analysis of the LoRa physical layer (PHY) functionalities, encompassing processes such as packet modulation, demodulation, and preamble detection. Then, we leverage on this analysis to create a pioneering SDR-based LoRa PHY implementation. Accordingly, we thoroughly describe all the implementation details. Moreover, we illustrate how SDR-LoRa can help boost research on the LoRa protocol by presenting three exemplary key applications that can be built on top of our implementation, namely fine-grained localization, interference cancellation, and enhanced link reliability. To validate SDR-LoRa and its applications, we test it on two different platforms: (i) a physical setup involving USRP radios and off-the-shelf commercial devices, and (ii) the Colosseum wireless channel emulator. Our experimental findings reveal that (i) SDR-LoRa performs comparably to conventional commercial LoRa systems, and (ii) all the aforementioned applications can be successfully implemented on top of SDR-LoRa with remarkable results. The complete details of the SDR-LoRa implementation code have been publicly shared online, together with a plug-and-play Colosseum container.
@article{busacca2024sdr,
title={SDR-LoRa, an open-source, full-fledged implementation of LoRa on Software-Defined-Radios: Design and potential exploitation},
author={Busacca, Fabio and Mangione, Stefano and Palazzo, Sergio and Restuccia, Francesco and Tinnirello, Ilenia},
journal={Computer Networks},
volume={241},
pages={110194},
year={2024},
publisher={Elsevier}
}
2024
Journal
Sensing in Bi-Static ISAC Systems with Clock Asynchronism: A Signal Processing Perspective
K. Wu, J. Pegoraro, F. Meneghello, JA. Zhang, JO. Lacruz, J. Widmer, F. Restuccia, M. Rossi, X. Huang, D. Zhang, G. Caire, Y. Guo
IEEE Signal Processing Magazine (IEEE SPM)
Integrated Sensing and Communication (ISAC) has been identified as a pillar usage scenario for the impending 6G era. Bi-static sensing, a major type of sensing in ISAC, is promising to expedite ISAC in the near future, as it requires minimal changes to the existing network infrastructure. However, a critical challenge for bi-static sensing is clock asynchronism due to the use of different clocks at far-separated transmitters and receivers. This causes the received signal to be affected by time-varying random phase offsets, severely degrading, or even failing, direct sensing. Hence, to effectively enable ISAC, considerable research has been directed toward addressing the clock asynchronism issue in bi-static sensing. This paper provides an overview of the issue and existing techniques developed in an ISAC background. Based on the review and comparison, we also draw insights into the future research directions and open problems, aiming to nurture the maturation of bi-static sensing in ISAC.
@article{wu2024sensing,
title={Sensing in Bi-Static ISAC Systems with Clock Asynchronism: A Signal Processing Perspective},
author={Wu, Kai and Pegoraro, Jacopo and Meneghello, Francesca and Zhang, J Andrew and Lacruz, Jesus O and Widmer, Joerg and Restuccia, Francesco and Rossi, Michele and Huang, Xiaojing and Zhang, Daqing and others},
journal={arXiv preprint arXiv:2402.09048},
year={2024}
}
2024
Conference
Federated Learning with Packet Losses
A. Rodio, G. Neglia, F. Busacca, S. Mangione, S. Palazzo, I. Tinnirello
Proc. of the International Symposium On Wireless Personal Multimedia Communications (WPMC)
This paper tackles the problem of training Federated Learning (FL) algorithms over real-world wireless networks with packet losses. Lossy communication channels between the orchestrating server and the clients affect the convergence of FL training as well as the quality of the learned model. Although many previous works investigated how to mitigate the adverse effects of packet losses, this paper demonstrates that FL algorithms over asymmetric lossy channels can still learn the optimal model, the same model that would have been trained in a lossless scenario by classic FL algorithms like FedAvg. Convergence to the optimum only requires slight changes to FedAvg: i) while FedAvg computes a new global model by averaging the received clients' models, our algorithm, UPGA-PL, updates the global model by a pseudo-gradient step; ii) UPGA-PL accounts for the potentially heterogeneous packet losses experienced by the clients to unbias the pseudo-gradient step. Still, UPGA-PL maintains the same computational and communication complexity as FedAvg. In our experiments, UPGA-PL not only outperforms existing state-of-the-art solutions for lossy channels (by more than 5 percentage points on test accuracy) but also matches FedAvg's performance in lossless scenarios after less than 150 communication rounds.
@inproceedings{rodio2023federated,
title={Federated Learning with Packet Losses},
author={Rodio, Angelo and Neglia, Giovanni and Busacca, Fabio and Mangione, Stefano and Palazzo, Sergio and Restuccia, Francesco and Tinnirello, Ilenia},
booktitle={2023 26th International Symposium on Wireless Personal Multimedia Communications (WPMC)},
pages={1--6},
year={2023},
organization={IEEE}
}
2023
Conference
Zero-Shot Dynamic Neural Network Adaptation in Tactical Wireless Systems
Shahriar Rifat, J. Ashdown, K. Turck, and F. Restuccia
Proc. of IEEE Military Communications Conference (IEEE MILCOM)
Models based on deep neural networks have attracted interest in a myriad of applications in the wireless landscape, including spectrum intelligence, modulation recognition, and radio fingerprinting, among others. One key challenge that is currently inhibiting the application of such models in real-world tactical scenarios is that their performance radically changes with continuously changing dynamic wireless channel conditions. Existing work tackle this by performing efficient fine-tuning or adaptation using as few samples as possible. However, some prior knowledge about the new conditions is involved in the process. In this paper, we propose for the first time zero-shot dynamic neural network adaptation (zDNA) (i.e., without any additional training samples) of wireless classification models after deployment in unseen conditions. Specifically, we show that by changing only the affine transformation parameters and normalizing the learned features in different layers of the neural network in an online manner without any labeled samples, we can achieve superior performance in dynamic conditions. Our proposed approach is evaluated on the publicly available RadioML 2018.01A dataset, to test its adaptability to dynamically changing signal-to-noise ratio (SNR) conditions. Performance improvement consistency in all unseen test scenarios (up to 24% in low SNR regime) demonstrates the applicability of our framework in real-world contexts.
@inproceedings{rifat2023zero,
title={Zero-Shot Dynamic Neural Network Adaptation in Tactical Wireless Systems},
author={Rifat, Shahriar and Ashdown, Jonathan and Turck, Kurt and Restuccia, Francesco},
booktitle={MILCOM 2023-2023 IEEE Military Communications Conference (MILCOM)},
pages={418--423},
year={2023},
organization={IEEE}
}
2023
Conference
Channel-Adaptive Dynamic Neural Networks for Low-Complexity Distributed Signal Analysis
M. Abdi, J. Ashdown, K. Turck, and F. Restuccia
Proc. of IEEE Military Communications Conference (IEEE MILCOM)
Deep neural networks (DNNs) are being increasingly used to achieve signal awareness. Usually, DNNs have been entirely executed at the mobile device. However, as DNNs become more complex, edge computing techniques will become necessary for DNN deployment. Notably, it has been shown that the original DNN architectures can be modified by introducing a "bottleneck" producing a compressed representation of the input. The DNN split computation is then orchestrated between the mobile device and the edge, which respectively execute the layers before and after the bottleneck. As yet, the concept of split computing has not been applied for signal awareness. Moreover, in existing work, the dynamic nature of the wireless channel is not considered in the bottleneck design. In this paper, we propose Split Computing Enhancing Signal Awareness (SpliCES), a new approach where the bottleneck is adapted at inference time according to ongoing channel conditions. Moreover, we present a novel training strategy which optimizes the bottleneck-injected ensemble parameters to generalize to all channel conditions in its first stage, and then specializes each bottleneck separately to a certain level of channel impairment in the second stage. We evaluate SpliCES on the well-known RadioML dataset for modulation recognition, and show that on average, our proposed method transmits 12x less data while experiencing only about 3% accuracy drop in high signal-to-noise ratio (SNR) regimes. With low SNR, SpliCES achieves 5% better accuracy thanks to its dynamic structure.
@inproceedings{abdi2023channel,
title={Channel-Adaptive Dynamic Neural Networks for Low-Complexity Distributed Signal Awareness},
author={Abdi, Mohammad and Ashdown, Jonathan and Turck, Kurt and Restuccia, Francesco},
booktitle={MILCOM 2023-2023 IEEE Military Communications Conference (MILCOM)},
pages={27--32},
year={2023},
organization={IEEE}
}
2023
Conference
Hercules: An Emulation-Based Framework for Transport Layer Measurements over 5G Wireless Networks
A. Pinto, A. Ashdown, T. B. Hassan, H. Cheng, F. Esposito, L. Bonati, S. D’Oro, T. Melodia, F. Restuccia
Proceedings of ACM Workshop on Wireless Network Testbeds, Experimental evaluation & Characterization (ACM WINTECH)
The adoption of Next-Generation cellular networks is rapidly increasing, together with their achievable throughput and their latency demands. Optimizing existing transport protocols for such networks is challenging, as the wireless channel becomes critical for performance and reliability studies. The performance assessment of transport protocols for wireless networks has mostly relied on simulation-based environments. While providing valuable insights, such studies are influenced by the simulator's specific settings. Employing more advanced and flexible methods for collecting and analyzing end-to-end transport layer datasets in realistic wireless environments is crucial to the design, implementation and evaluation of transport protocols that are effective when employed in real-world 5G networks. We present Hercules, a containerized 5G standalone framework that collects data employing the OpenAirInterface 5G protocol stack. We illustrate its potential with an initial transport layer and 5G stack measurement campaign on the Colosseum wireless network testbed. In addition, we present preliminary post-processing results from testing various TCP Congestion Control techniques over multiple wireless channels.
@inproceedings{pinto2023hercules,
title={{Hercules: An Emulation-Based Framework for Transport Layer Measurements over 5G Wireless Networks}},
author={Pinto, Andrea and Ashdown, Andrew and Bin Hassan, Tanzil and Cheng, Hai and Esposito, Flavio and Bonati, Leonardo and D'Oro, Salvatore and Melodia, Tommaso and Restuccia, Francesco},
booktitle={Proceedings of the ACM Workshop on Wireless Network Testbeds, Experimental evaluation & Characterization},
pages={72--79},
year={2023}
}
2023
Conference
Wi-BFI: Extracting the IEEE 802.11 Beamforming Feedback Information from Commercial Wi-Fi Devices
K. Foysal Haque, F. Meneghello, and F. Restuccia
Proc. of the 17th ACM Workshop on Wireless Network Testbeds, Experimental Evaluation & Characterization (ACM WINTECH)
Recently, researchers have shown that the beamforming feedback angles (BFAs) used for Wi-Fi multiple-input multiple-output (MIMO) operations can be effectively leveraged as a proxy of the channel frequency response (CFR) for different purposes. Examples are passive human activity recognition and device fingerprinting. However, even though the BFAs report frames are sent in clear text, there is not yet a unified open-source tool to extract and decode the BFAs from the frames. To fill this gap, we developed Wi-BFI, the first tool that allows retrieving Wi-Fi BFAs and reconstructing the beamforming feedback information (BFI) -- a compressed representation of the CFR -- from the BFAs frames captured over the air. The tool supports BFAs extraction within both IEEE 802.11ac and 802.11ax networks operating on radio channels with 160/80/40/20 MHz bandwidth. Both multi-user and single-user MIMO feedback can be decoded through Wi-BFI. The tool supports real-time and offline extraction and storage of BFAs and BFI. The real-time mode also includes a visual representation of the channel state that continuously updates based on the collected data. Wi-BFI code is open source and the tool is also available as a pip package1.
@inproceedings{haque2023wi,
title={Wi-BFI: Extracting the IEEE 802.11 Beamforming Feedback Information from Commercial Wi-Fi Devices},
author={Haque, Khandaker Foysal and Meneghello, Francesca and Restuccia, Francesco},
booktitle={Proceedings of the 17th ACM Workshop on Wireless Network Testbeds, Experimental evaluation \& Characterization},
pages={104--111},
year={2023}
} 2023
Conference
Identification of Stealthy Hardware Trojans through On-Chip Temperature Sensing and an Autoencoder-Based Machine Learning Algorithm
T. Gourousis, Z. Zhang, M. Yan, M. Zhang, A. Mittal, A. Shrivastava, F. Restuccia, Y. Fei, M. Onabajo
Proc. of IEEE International Midwest Symposium on Circuits and Systems (IEEE MWSCAS)
This paper presents an anomaly detection approach with non-invasive on-chip temperature sensing for hardware Trojan detection, which is coupled with a proposed anomaly detection technique using an autoencoder-based machine learning (ML) algorithm. In a case study, the developed algorithm identifies the Hardware Trojan with over 90% accuracy for a Trojan power consumption that is as low as 2.5% of the circuit under test. Besides identifying the Trojan, the algorithm can also provide information about the location of the Trojan on the chip. The proposed on-chip anomaly detection approach with machine learning is under development as a solution for enhanced hardware security in modern electronic systems, particularly for Internet of Things applications.
@inproceedings{gourousis2023identification,
title={Identification of Stealthy Hardware Trojans through On-Chip Temperature Sensing and an Autoencoder-Based Machine Learning Algorithm},
author={Gourousis, Thomas and Zhang, Ziyue and Yan, Mengting and Zhang, Millin and Mittal, Ankit and Shrivastava, Aatmesh and Restuccia, Francesco and Fei, Yunsi and Onabajo, Marvin},
booktitle={2023 IEEE 66th International Midwest Symposium on Circuits and Systems (MWSCAS)},
pages={30--34},
year={2023},
organization={IEEE}
}
2023
Patent
Neural Network for Adversarial Deep Learning in Wireless Systems
F. Restuccia, S. D’Oro and T. Melodia
U.S. Patent No.: 17/128,437
2023
Conference
SplitBeam: Effective and Efficient Beamforming in Wi-Fi Networks Through Split Computing
N. Bahadori, Y. Matsubara, M. Levorato and F. Restuccia
Proc. of the IEEE International Conference on Distributed Computing Systems (IEEE ICDCS)
Modern IEEE 802.11 (Wi-Fi) networks extensively rely on multiple-input multiple-output (MIMO) to significantly improve throughput. To correctly beamform MIMO transmissions, the access point needs to frequently acquire a beamforming matrix (BM) from each connected station. However, the size of the matrix grows with the number of antennas and subcarriers, resulting in an increasing amount of airtime overhead and computational load at the station. Conventional approaches come with either excessive computational load or loss of beamforming precision. For this reason, we propose SplitBeam, a new framework where we train a split deep neural network (DNN) to directly output the BM given the channel state information (CSI) matrix as input. The DNN is designed with an additional “bottleneck” layer to “split” the original DNN into a head model and a tail model, respectively executed by the station and the access point. The head model generates a compressed representation of the BM, which is then used by the AP to produce the BM using the tail model. We formulate and solve a bottleneck optimization problem (BOP) to keep computation, airtime overhead, and bit error rate (BER) below application requirements. We perform extensive experimental CSI collection with off-the-shelf Wi-Fi devices in two distinct environments and compare the performance of SplitBeam with the standard IEEE 802.11 algorithm for BM feedback and the state-of-the-art DNN-based approach LB-SciFi. Our experimental results show that SplitBeam reduces the beamforming feedback size and computational complexity by respectively up to 81 % and 84 % while maintaining BER within about 10 −3 of existing approaches. We also implement the SplitBeam DNNs on FPGA hardware to estimate the end-to-end BM reporting delay, and show that the latter is less than 10 milliseconds in the most complex scenario, which is the target channel sounding frequency in realistic multi-user MIMO scenarios. To allow full reproducibility, we will release our code and datasets to the community.
@inproceedings{baadori2023splitbeam,
title={{SplitBeam: Effective and Efficient Beamforming in Wi-Fi Networks Through Split Computing}},
author={Bahadori, Niloofar and Matsubara, Yoshitomo and Levorato, Marco and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Conference on Distributed Computing Systems (IEEE ICDCS)},
year={2023},
}
2023
Journal
Toward Integrated Sensing and Communications in IEEE 802.11bf Wi-Fi Networks
F. Meneghello, C. Chen, C. Cordeiro, and F. Restuccia
IEEE Communications Magazine
As Wi-Fi becomes ubiquitous in public and private spaces, it becomes natural to leverage its intrinsic ability to sense the surrounding environment to implement groundbreaking wireless sensing applications such as human presence detection, activity recognition, and object tracking. For this reason, the IEEE 802.11bf Task Group is defining the appropriate modifications to existing Wi-Fi standards to enhance sensing capabilities through 802.11-compliant devices. However, the new standard is expected to leave the specific sensing algorithms open to implementation. To fill this gap, this article explores the practical implications of integrating sensing into Wi-Fi networks. We provide an overview of the physical and medium access control layers sensing enablers, together with the application layer perspective. We analyze the impact of communication parameters on sensing performance and detail the main research challenges. To make our evaluation replicable, we pledge to release all of our dataset and code to the community. The dataset and code are available at https://github.com/francescamen/SHARPax.
@article{meneghello2023toward,
title={{Toward Integrated Sensing and Communications in IEEE 802.11 bf Wi-Fi Networks}},
author={Meneghello, Francesca and Chen, Cheng and Cordeiro, Carlos and Restuccia, Francesco},
journal={IEEE Communications Magazine},
volume={61},
number={7},
pages={128--133},
year={2023},
publisher={IEEE}
}
2023
Conference
Shot Boundary Detection and Convolutional Neural Network for Video Classification
F. Candela, A. Giordano, M. Campolo, F. Restuccia and F. C. Morabito
Proceedings of the International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME 2023)
The classification of videos is a crucial computer vision task. Convolutional Neural Networks (CNN) have become a popular choice for video classification due to their ability to learn hierarchical representations of the input data. Shot boundary detection is another technique that is often used in video classification to segment the video into shots, which are temporal segments that contain a continuous action or scene. In the long-term video of a television broadcast there are continuous scene changes, some of which better define the beginning or end of a program, an advertisement, a type of human action and much more, supporting information for classification made only on specific video segments. Shot boundary detection can be incorporated into CNN models to segment the video into shots, and then feeding each shot into the CNN-based model for classification of television programs and their content. In this paper the implementation of a shot boundary technique will be described together with a pre-trained CNN on ImageNet for the video classification of some types of programs and television contents according to a schedule provided by the Italian national communications agency, the presented framework succeeds in achieving accuracy levels of 96%.
@inproceedings{candela2023shot,
title={{Shot Boundary Detection and Convolutional Neural Network for Video Classification}},
author={Candela, Federico and Giordano, Angelo and Campolo, Maurizio and Restuccia, Francesco and Morabito, Francesco Carlo},
booktitle={Proceedings of International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME)},
pages={1--5},
year={2023},
organization={IEEE}
}
2023
Conference
SiMWiSense: Simultaneous Multi-Subject Activity Classification Through Wi-Fi Signals
K. Foysal Haque, M. Zhang and F. Restuccia
Proceedings of the IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks
Recent advances in Wi-Fi sensing have ushered in a plethora of pervasive applications in home surveillance, remote healthcare, road safety, and home entertainment, among others. Most of the existing works are limited to the activity classification of a single human subject at a given time. Conversely, a more realistic scenario is to achieve simultaneous, multi-subject activity classification. The first key challenge in that context is that the number of classes grows exponentially with the number of subjects and activities. Moreover, it is known that Wi-Fi sensing systems struggle to adapt to new environments and subjects. To address both issues, we propose SiMWiSense, the first framework for simultaneous multi-subject activity classification based on Wi-Fi that generalizes to multiple environments and subjects. We address the scalability issue by using the Channel State Information (CSI) computed from the device positioned closest to the subject. We experimentally prove this intuition by confirming that the best accuracy is experienced when the CSI computed by the transceiver positioned closest to the subject is used for classification. To address the generalization issue, we develop a brand-new few-shot learning algorithm named Feature Reusable Embedding Learning (FREL). Through an extensive data collection campaign in 3 different environments and 3 subjects performing 20 different activities simultaneously, we demonstrate that SiMWiSense achieves classification accuracy of up to 97%, while FREL improves the accuracy by 85% in comparison to a traditional Convolutional Neural Network (CNN) and up to 20% when compared to the state-of-the-art few-shot embedding learning (FSEL), by using only 15 seconds of additional data for each class. For reproducibility purposes, we share our 1TB dataset and code repository.
@article{haque2023simwisense,
title={{SiMWiSense: Simultaneous Multi-Subject Activity Classification Through Wi-Fi Signals}},
author={Haque, Khandaker Foysal and Zhang, Milin and Restuccia, Francesco},
journal={Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (IEEE WoWMoM)},
year={2023}
}
2023
Conference
Edge-V: Enabling Vehicular Edge Intelligence in Unlicensed Spectrum Bands
F. Raviglione, C. Casetti and F. Restuccia
Proceedings of IEEE Vehicular Technology Conference (IEEE VTC2023-Spring)
Cutting-edge advances in wireless networking will soon enable a new generation of safer, smarter, and more autonomous vehicles. These vehicles will rely on real-time execution of complex Deep Learning (DL) tasks as well as high-speed multimedia streaming between road users for navigation purposes. Relying entirely on cellular networks (i) puts an unnecessary burden on an already overcrowded and expensive licensed spectrum; (ii) increases the latency of edge-offloaded tasks to intolerable levels for vehicular applications. Alongside the usage of a proper network infrastructure, vehicles will need to support on-board and offloaded cooperative intelligence. On this basis, we propose Edge-V , the first framework enabling practical vehicular edge intelligence and high-speed vehicular connectivity, using only unlicensed spectrum bands. Through a DSRC link, Edge-V acquires real-time localized knowledge, and coordinates the use of point-to-point millimeter Wave (mmWave) technologies to deliver high-bandwidth connectivity between vehicles. Edge-V also foresees smart offloading if on-board computing resources are insufficient. We prototype and evaluate Edge-V in a real-world laboratory testbed, showing its advantages with respect to cellular and cloud-based approaches.
@inproceedings{raviglione2023edge,
title={{Edge-V: Enabling Vehicular Edge Intelligence in Unlicensed Spectrum Bands}},
author={Raviglione, Francesco and Casetti, Claudio and Restuccia, Francesco},
booktitle={Proceedings of IEEE Vehicular Technology Conference (IEEE VTC2023-Spring)},
pages={1--5},
year={2023},
organization={IEEE}
}
2023
Conference
Mobile RF Scenario Design for Massive-Scale Wireless Channel Emulators
R. Rusca, F. Raviglione, C. Casetti, P. Giaccone, F. Restuccia
Proceedings of Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit)
Large-scale wireless emulation is gaining momen-tum nowadays, thanks to its potential in the development and deployment of advanced use cases for next-generation wireless networks. Several novel use cases are indeed emerging, including massive MIMO, millimeter wave beamforming and AI-based Vehicle-to-Everything (V2X) optimized communication. The de-velopment and testing of a wireless application, especially at a large scale and when dealing with mobile nodes, faces several challenges that cannot be solved by simulation frameworks alone. Thus, massive-scale channel emulators are emerging, enabling the emulation of realistic scenarios which leverage real hardware and radio signals. However, this is a complex task due to the lack of realistic scenarios based on real datasets. We thus propose a novel framework for the design and generation of channel emulation scenarios starting from real mobility traces, either generated by means of dedicated tools, or collected on the field. Our framework provides a practical way of generating mobility scenarios with vehicles, pedestrians, drones and other mobile entities. We detail all the steps foreseen by our framework, from the provision of the traces and radio parameters, to the generation of a matrix describing the delay and IQ samples for each time instant and node in the scenario. We also showcase the potentiality of our proposal by designing and creating a vehicular 5G scenario with 13 vehicles, starting from a recently-disclosed open dataset. This scenario is then validated on the Colosseum channel emulator, proving how our framework can provide an effective tool for large-scale wireless networking evaluation.
@inproceedings{rusca2023mobile,
title={{Mobile RF Scenario Design for Massive-Scale Wireless Channel Emulators}},
author={Rusca, Riccardo and Raviglione, Francesco and Casetti, Claudio and Giaccone, Paolo and Restuccia, Francesco},
booktitle={Proceedings of Joint European Conference on Networks and Communications & 6G Summit (EuCNC/6G Summit)},
pages={675--680},
year={2023},
organization={IEEE}
} 2023
Conference
A2-UAV: Application-Aware Content and Network Optimization of Edge-Assisted UAV Systems
A. Coletta, F. Giorgi, G. Maselli, M. Prata, D. Silvestri, J. Ashdown, and F. Restuccia
Proceedings of IEEE Conference on Computer Communications (IEEE INFOCOM)
To perform advanced surveillance, Unmanned Aerial Vehicles (UAVs) require the execution of edge-assisted computer vision (CV) tasks. In multi-hop UAV networks, the successful transmission of these tasks to the edge is severely challenged due to severe bandwidth constraints. For this reason, we propose a novel A2-UAV framework to optimize the number of correctly executed tasks at the edge. In stark contrast with existing art, we take an application-aware approach and formulate a novel pplication-Aware Task Planning Problem (A2-TPP) that takes into account (i) the relationship between deep neural network (DNN) accuracy and image compression for the classes of interest based on the available dataset, (ii) the target positions, (iii) the current energy/position of the UAVs to optimize routing, data pre-processing and target assignment for each UAV. We demonstrate A2-TPP is NP-Hard and propose a polynomial-time algorithm to solve it efficiently. We extensively evaluate A2-UAV through real-world experiments with a testbed composed by four DJI Mavic Air 2 UAVs. We consider state-of-the-art image classification tasks with four different DNN models (i.e., DenseNet, ResNet152, ResNet50 and MobileNet-V2) and object detection tasks using YoloV4 trained on the ImageNet dataset. Results show that A2-UAV attains on average around 38% more accomplished tasks than the state-of-the-art, with 400% more accomplished tasks when the number of targets increases significantly. To allow full reproducibility, we pledge to share datasets and code with the research community.
@article{coletta20232,
title={{A}$^{2}${-UAV: Application-Aware Content and Network Optimization of Edge-Assisted UAV Systems}},
author={Coletta, Andrea and Giorgi, Flavio and Maselli, Gaia and Prata, Matteo and Silvestri, Domenicomichele and Ashdown, Jonathan and Restuccia, Francesco},
journal={Proceedings of IEEE International Conference on Computer Communications (IEEE INFOCOM)},
year={2023}
}
2023
Conference
SEM-O-RAN: Semantic and Flexible O-RAN Slicing for NextG Edge-Assisted Mobile Systems
C. Puligheddu, J. Ashdown, C. F. Chiasserini, and F. Restuccia
Proceedings of IEEE Conference on Computer Communications (IEEE INFOCOM)
5G and beyond cellular networks (NextG) will support the continuous execution of resource-expensive edge-assisted deep learning (DL) tasks. To this end, Radio Access Network (RAN) resources will need to be carefully "sliced" to satisfy heterogeneous application requirements while minimizing RAN usage. Existing slicing frameworks treat each DL task as equal and inflexibly define the resources to assign to each task, which leads to sub-optimal performance. In this paper, we propose SEM-O-RAN, the first semantic and flexible slicing framework for NextG Open RANs. Our key intuition is that different DL classifiers can tolerate different levels of image compression, due to the semantic nature of the target classes. Therefore, compression can be semantically applied so that the networking load can be minimized. Moreover, flexibility allows SEM-O-RAN to consider multiple edge allocations leading to the same task-related performance, which significantly improves system-wide performance as more tasks can be allocated. First, we mathematically formulate the Semantic Flexible Edge Slicing Problem (SF-ESP), demonstrate that it is NP-hard, and provide an approximation algorithm to solve it efficiently. Then, we evaluate the performance of SEM-O-RAN through extensive numerical analysis with state-of-the-art multi-object detection (YOLOX) and image segmentation (BiSeNet V2), as well as real-world experiments on the Colosseum testbed. Our results show that SEM-O-RAN improves the number of allocated tasks by up to 169% with respect to the state of the art.
@inproceedings{puligheddu2023sem,
title={{SEM-O-RAN: Semantic and Flexible O-RAN Slicing for NextG Edge-Assisted Mobile Systems}},
author={Puligheddu, Corrado and Ashdown, Jonathan and Chiasserini, Carla Fabiana and Restuccia, Francesco},
booktitle={Proceedings of IEEE Conference on Computer Communications (IEEE INFOCOM)},
pages={1--10},
year={2023},
organization={IEEE}
}
2023
Patent
Real-Time Cognitive Wireless Networking Through Deep Learning in Transmission and Reception Communication Paths
F. Restuccia and T. Melodia
U.S. Patent No.: 11,610,111
2023
Patent
Device and Method for Reliable Classification of Wireless Signals
S. D’Oro, F. Restuccia, and T. Melodia
U.S. Patent No.: 11,611,457
2023
Journal
Deep Learning at the Physical Layer for Adaptive Terahertz Communications
J. Hall, J.M. Jornet, N. Thawdar, T. Melodia and F. Restuccia
IEEE Transactions on Terahertz Science and Technology
Wireless communications in the terahertz (THz) band will become a cornerstone of sixth-generation (6G) networks. The THz channel, however, presents several challenges, such as distance-dependent absorption coefficients that can change the bandwidth significantly in case of mobility. Thus, future THz transmitters will have to switch modulation and bandwidth almost continuously. Moreover, using the same transmission scheme can enable adversaries to leverage smart interfering to inflict more damage with less energy expense. To help enable adaptive and secure THz communications, this article presents the first ever experimental study of modulation and bandwidth classification (MBC) at THz frequencies through deep learning (DL) techniques. We have performed an extensive experimental data collection campaign at 120 GHz with different modulation schemes, signal bandwidth (up to 20 GHz), and different signal-to-noise ratio (SNR) levels. We prove for the first time the feasibility and effectiveness of MBC at THz frequencies, with our DL models reaching accuracy up to 78% and 90% in low- and high-SNR conditions. Furthermore, we investigate the memory and latency constraints that need to be satisfied as a function of the signal bandwidth, and propose a boosting technique to improve the inference quality by trading off latency for accuracy. Finally, we experimentally evaluate the latency of our CNN models through FPGA implementation.
@article{hall2023deep,
title={{Deep Learning at the Physical Layer for Adaptive Terahertz Communications}},
author={Hall, Jacob and Jornet, Josep Miquel and Thawdar, Ngwe and Melodia, Tommaso and Restuccia, Francesco},
journal={IEEE Transactions on Terahertz Science and Technology},
volume={13},
number={2},
pages={102--112},
year={2023},
publisher={IEEE}
}
2023
Conference
Exposing the CSI: A Systematic Investigation of CSI-based Wi-Fi Sensing Capabilities and Limitations
M. Cominelli, F. Gringoli and F. Restuccia
Proceedings of IEEE Conference on Pervasive Computing and Communications (IEEE PerCom)
Thanks to the ubiquitous deployment of Wi-Fi hotspots, channel state information (CSI)-based Wi-Fi sensing can unleash game-changing applications in many fields, such as healthcare, security, and entertainment. However, despite one decade of active research on Wi-Fi sensing, most existing work only considers legacy IEEE 802.11n devices, often in particular and strictly-controlled environments. Worse yet, there is a fundamental lack of understanding of the impact on CSI-based sensing of modern Wi-Fi features, such as 160-MHz bandwidth, multiple-input multiple-output (MIMO) transmissions, and increased spectral resolution in IEEE 802.11ax (Wi-Fi 6). This work aims to shed light on the impact of Wi-Fi 6 features on the sensing performance and to create a benchmark for future research on Wi-Fi sensing. To this end, we perform an extensive CSI data collection campaign involving 3 individuals, 3 environments, and 12 activities, using Wi-Fi 6 signals. An anonymized ground truth obtained through video recording accompanies our 80-GB dataset, which contains almost two hours of CSI data from three collectors. We leverage our dataset to dissect the performance of a state-of-the-art sensing framework across different environments and individuals. Our key findings suggest that (i) MIMO transmissions and higher spectral resolution might be more beneficial than larger bandwidth for sensing applications; (ii) there is a pressing need to standardize research on Wi-Fi sensing because the path towards a truly environment-independent framework is still uncertain. To ease the experiments' replicability and address the current lack of Wi-Fi 6 CSI datasets, we release our 80-GB dataset to the community.
@inproceedings{cominelli2023exposing,
title={{Exposing the CSI: A Systematic Investigation of CSI-based Wi-Fi Sensing Capabilities and Limitations}},
author={Cominelli, Marco and Gringoli, Francesco and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Conference on Pervasive Computing and Communications (IEEE PerCom)},
pages={81--90},
year={2023},
organization={IEEE}
} 2023
Patent
Coordination-Free mmWave Beam Management With Deep Waveform Learning
M. Polese, F. Restuccia and T. Melodia
U.S. Patent No.: 11,588,539
2023
Patent
Underwater Ultrasonic Communication System and Method
F. Restuccia, E. Demirors, and T. Melodia
U.S. Patent No.: 11,575,987
2023
Journal
Wi-Fi Sensing Based on IEEE 802.11bf
C. Chen, H. Song, Q. Li, F. Meneghello, F. Restuccia and C. Cordeiro
IEEE Communications Magazine
Conventionally, Wi-Fi radio signals are widely used for data transmissions in a wireless local area network (WLAN). Recently, it has been an interesting topic to also apply Wi-Fi radio signals to sense the environment where these signals propagate and identify changes associated with certain activities. This technique is referred to as Wi-Fi sensing, and it has been proven effective in a variety of use cases, such as proximity detection, gesture recognition, target counting, and health monitoring. As a result, the IEEE 802.11 working group has formed a new Task Group, 802.11 bf, to develop a new amendment to define necessary PHY and MAC protocols to support Wi-Fi sensing in all spectrum bands, including sub-7 GHz bands (2.4 GHz, 5 GHz, and 6 GHz band), as well as 60 GHz millimeter-wave band. In this article, our primary goal is to identify and describe the basic elements that have been developed in IEEE 802.11 bf to enable Wi-Fi sensing applications in different WLAN scenarios.
@article{chen2022wi,
title={{Wi-Fi sensing based on IEEE 802.11 bf}},
author={Chen, Cheng and Song, Hao and Li, Qinghua and Meneghello, Francesca and Restuccia, Francesco and Cordeiro, Carlos},
journal={IEEE Communications Magazine},
volume={61},
number={1},
pages={121--127},
year={2022},
publisher={IEEE}
} 2023
Journal
IEEE 802.11 Artificial Intelligence Machine Learning (AIML) Technical Interest Group
F. Restuccia
IEEE Communications Standards Magazine
IEEE 802.11 Artificial Intelligence Machine Learning (AIML) Technical Interest Group
@ARTICLE{10034506,
author={Francesco Restuccia},
journal={IEEE Communications Standards Magazine},
title={{IEEE 802.11 Artificial Intelligence Machine Learning (AIML) Technical Interest Group (in Standards News)}},
year={2022},
volume={6},
number={4},
pages={4-5},
doi={10.1109/MCOMSTD.2022.10034506}}
2022
Journal
Split Computing and Early Exiting for Deep Learning Applications: Survey and Research Challenges
Y. Matsubara, M. Levorato and F. Restuccia
ACM Computing Surveys
Mobile devices such as smartphones and autonomous vehicles increasingly rely on deep neural networks (DNNs) to execute complex inference tasks such as image classification and speech recognition, among others. However, continuously executing the entire DNN on mobile devices can quickly deplete their battery. Although task offloading to cloud/edge servers may decrease the mobile device’s computational burden, erratic patterns in channel quality, network, and edge server load can lead to a significant delay in task execution. Recently, approaches based on split computing (SC) have been proposed, where the DNN is split into a head and a tail model, executed respectively on the mobile device and on the edge server. Ultimately, this may reduce bandwidth usage as well as energy consumption. Another approach, called early exiting (EE), trains models to embed multiple “exits” earlier in the architecture, each providing increasingly higher target accuracy. Therefore, the tradeoff between accuracy and delay can be tuned according to the current conditions or application demands. In this article, we provide a comprehensive survey of the state of the art in SC and EE strategies by presenting a comparison of the most relevant approaches. We conclude the article by providing a set of compelling research challenges.
@article{matsubara2022split,
title={{Split Computing And Early Exiting For Deep Learning Applications: Survey And Research Challenges}},
author={Matsubara, Yoshitomo and Levorato, Marco and Restuccia, Francesco},
journal={ACM Computing Surveys},
volume={55},
number={5},
pages={1--30},
year={2022},
publisher={ACM New York, NY}
} 2022
Conference
3D-O-RAN: Dynamic Data Driven Open Radio Access Network Systems
F. Restuccia, E. Blasch, A. Ashdown, J. Ashdown, and K. Turck
Proceedings of IEEE Military Communications Conference (IEEE MILCOM)
This position paper introduces a Dynamic Data Driven Open Radio Access Network System (3D-O-RAN). The key objective of 3D-O-RAN is to support congested, contested and contaminated tactical settings where multimedia sensors, application constraints and operating wireless conditions may frequently change over space, time and frequency. 3D-O-RAN is compliant with the O-RAN specification for beyond 5G cellular systems to reduce costs and guarantee interoperability among vendors. Moreover, 3D-O-RAN integrates computational, sensing, and cellular networking components in a highly-dynamic, feedback-based, data-driven control loop. Specifically, 3D-O-RAN is designed to incorporate heterogeneous data into the network control loop to achieve a system-wide optimal operating point. Moreover, 3D-O-RAN steers the multimedia sensor measurement process in real time according to the required application needs and current physical and/or environmental constraints. 3D-O-RAN uses (i) a semantic slicing engine, which takes into account the semantic of the application to optimally compress the multimedia stream without losing in classification accuracy; (ii) a dynamic data driven neural network certification system that translates mission-level constraints into technical-level constraints on neural network latency/accuracy, and occupation of hardware/software resources. Realistic use-case scenarios of 3D-O-RAN in a tactical context demonstrate system performance.
@inproceedings{restuccia20223d,
title={{3D-O-RAN: Dynamic Data Driven Open Radio Access Network Systems}},
author={Restuccia, Francesco and Blasch, Erik and Ashdown, Andrew and Ashdown, Jonathan and Turck, Kurt},
booktitle={Proceedings of IEEE Military Communications Conference (IEEE MILCOM)},
pages={19--24},
year={2022},
organization={IEEE}
}
2022
Conference
SDR-LoRa: Dissecting and Implementing LoRa on Software-Defined Radios to Advance Experimental IoT Research
F. Busacca, S. Mangione, I. Tinnirello, S. Palazzo, and F. Restuccia
Proceedings of ACM Workshop on Wireless Network Testbeds, Experimental Evaluation & Characterization (ACM WINTECH)
In this paper, we present SDR-LoRa, a full-fledged SDR implementation of a LoRa transmitter and receiver. First, we reverse-engineer the LoRa physical layer (PHY) functionalities, including the procedures of packet modulation, demodulation, and preamble detection. Based on this analysis, we develop the first Software Defined Radio (SDR) implementation of the LoRa PHY. Furthermore, we integrate LoRa with an Automatic Repeat Request (ARQ) error detection protocol. SDR-LoRa has been validated on (i) the Colosseum wireless channel emulator; and (ii) a real testbed with USRP radios and commercial-off-the-shelf (COTS) devices. Our experimental results demonstrate that the performance of SDR-LoRa is in line with commercial LoRa systems. We pledge to share the entirety of the SDR-LoRa code.
@inproceedings{busacca2022sdr,
title={{SDR-LoRa: Dissecting and Implementing LoRa on Software-Defined Radios to Advance Experimental IoT Research}},
author={Busacca, Fabio and Mangione, Stefano and Tinnirello, Ilenia and Palazzo, Sergio and Restuccia, Francesco},
booktitle={Proceedings of the ACM Workshop on Wireless Network Testbeds, Experimental evaluation & CHaracterization},
pages={24--31},
year={2022}
} 2022
Journal
Generalized Wireless Adversarial Deep Learning
F. Restuccia, S. D’Oro, A. Al-Shawabka, B. Costa Rendon, K. Chowdhury, S. Ioannidis and T. Melodia
Computer Networks
Deep learning techniques can classify spectrum phenomena (e.g., waveform modulation) with accuracy levels that were once thought impossible. Although we have recently seen many advances in this field, extensive work in computer vision has demonstrated that an adversary can "crack" a classifier by designing inputs that "steer" the classifier away from the ground truth. This paper advances the state of the art by proposing a generalized analysis and evaluation of adversarial machine learning (AML) attacks to deep learning systems in the wireless domain. We postulate a series of adversarial attacks, and formulate a Generalized Wireless Adversarial Machine Learning Problem (GWAP) where we analyze the combined effect of the wireless channel and the adversarial waveform on the efficacy of the attacks. We extensively evaluate the performance of our attacks on a state-of-the-art 1,000-device radio fingerprinting dataset, and a 24-class modulation dataset. Results show that our algorithms can decrease the classifiers' accuracy up to 3x while keeping the waveform distortion to a minimum.
@inproceedings{restuccia2020generalized,
title={{Generalized Wireless Adversarial Deep Learning}},
author={Restuccia, Francesco and D'Oro, Salvatore and Al-Shawabka, Amani and Rendon, Bruno Costa and Chowdhury, Kaushik and Ioannidis, Stratis and Melodia, Tommaso},
booktitle={Proceedings of the 2nd ACM Workshop on Wireless Security and Machine Learning},
pages={49--54},
year={2020}
} 2022
Conference
Terahertz Communications Can Work in Rain and Snow: Impact of Adverse Weather Conditions on Channels at 140 GHz
J. Hall, M. Polese, V. Petrov, D. Bodet, F. Restuccia, T. Melodia, and J. Jornet
Proceedings of the ACM Workshop on Millimeter-Wave and Terahertz Networks and Sensing Systems (mmNets)
Next-generation wireless networks will leverage the spectrum above 100 GHz to enable ultra-high data rate communications over multi-GHz-wide bandwidths. The propagation environment at such high frequencies, however, introduces challenges throughout the whole protocol stack design, from physical layer signal processing to application design. Therefore, it is fundamental to develop a holistic understanding of the channel propagation and fading characteristics over realistic deployment scenarios and ultra-wide bands. In this paper, we conduct an extensive measurement campaign to evaluate the impact of weather conditions on a wireless link in the 130-150 GHz band through a channel sounding campaign with clear weather, rain, and snow in a typical urban backhaul scenario. We present a novel channel sounder design that captures signals with -82 dBm sensitivity and 20 GHz of bandwidth. We analyze link budget, capacity, as well as channel parameters such as the delay spread and the K-factor. Our experimental results indicate that in the considered context the adverse weather does not interrupt the link, but introduces some additional constraints (e.g., high delay spread and increase in path loss in snow conditions) that need to be accounted for in the design of reliable Sixth Generation (6G) communication links above 100 GHz.
@inproceedings{sen2022terahertz,
title={{Terahertz Communications Can Work in Rain and Snow: Impact of Adverse Weather Conditions on Channels at 140 GHz}},
author={Sen, Priyangshu and Hall, Jacob and Polese, Michele and Petrov, Vitaly and Bodet, Duschia and Restuccia, Francesco and Melodia, Tommaso and Jornet, Josep M},
booktitle={Proceedings of the ACM Workshop on Millimeter-Wave and Terahertz Networks and Sensing Systems},
pages={13--18},
year={2022}
}
2022
Conference
DeepCSI: Rethinking Wi-Fi Radio Fingerprinting Through MU-MIMO CSI Feedback Deep Learning
F. Meneghello, M. Rossi and F. Restuccia
Proceedings of the IEEE International Conference on Distributed Computing Systems (ICDCS)
We present DeepCSI, a novel approach to Wi-Fi radio fingerprinting (RFP) which leverages standard-compliant beamforming feedback matrices to authenticate MU-MIMO Wi-Fi devices on the move. By capturing unique imperfections in off-the-shelf radio circuitry, RFP techniques can identify wireless devices directly at the physical layer, allowing low-latency low-energy cryptography-free authentication. However, existing Wi-Fi RFP techniques are based on software-defined radio (SDRs), which may ultimately prevent their widespread adoption. Moreover, it is unclear whether existing strategies can work in the presence of MU-MIMO transmitters – a key technology in modern Wi-Fi standards. Conversely from prior work, DeepCSI does not require SDR technologies and can be run on any low-cost Wi-Fi device to authenticate MU-MIMO transmitters. Our key intuition is that imperfections in the transmitter’s radio circuitry percolate onto the beamforming feedback matrix, and thus RFP can be performed without explicit channel state information (CSI) computation. DeepCSI is robust to inter-stream and inter-user interference being the beamforming feedback not affected by those phenomena. We extensively evaluate the performance of DeepCSI through a massive data collection campaign performed in the wild with off-the-shelf equipment, where 10 MU-MIMO Wi-Fi radios emit signals in different positions. Experimental results indicate that DeepCSI correctly identifies the transmitter with an accuracy of up to 98%. The identification accuracy remains above 82% when the device moves within the environment. To allow replicability and provide a performance benchmark, we pledge to share the 800 GB datasets – collected in static and, for the first time, dynamic conditions – and the code database with the community.
@inproceedings{meneghello2022deepcsi,
title={{DeepCSI: Rethinking Wi-Fi Radio Fingerprinting Through MU-MIMO CSI Feedback Deep Learning}},
author={Meneghello, Francesca and Rossi, Michele and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Conference on Distributed Computing Systems (ICDCS)},
pages={1062--1072},
year={2022},
organization={IEEE}
}
2022
Conference
ReWiS: Reliable Wi-Fi Sensing Through Few-Shot Multi-Antenna Multi-Receiver CSI Learning
N. Bahadori, J. Ashdown and F. Restuccia
Proceedings of the IEEE International Symposium on a World of Wireless, Mobile, and Multimedia Networks (IEEE WoWMoM) — Best Paper Award
Thanks to the ubiquitousness of Wi-Fi access points and devices, Wi-Fi sensing enables transformative applications in remote health care, home/office security, and surveillance, just to name a few. Existing work has explored the usage of machine learning on channel state information (CSI) computed from Wi-Fi packets to classify events of interest. However, most of these algorithms require a significant amount of data collection, as well as extensive computational power for additional CSI feature extraction. Moreover, the majority of these models suffer from poor accuracy when tested in a new/untrained environment. In this paper, we propose ReWiS, a novel framework for robust and environment-independent Wi-Fi sensing. The key innovation of ReWiS is to leverage few-shot learning (FSL) as the inference engine, which (i) reduces the need for extensive data collection and application-specific feature extraction; (ii) can rapidly generalize to new environments by leveraging only a few new samples. Moreover, ReWiS leverages multi-antenna, multi-receiver diversity, as well as fine-grained frequency resolution, to improve the overall robustness of the algorithms. Finally, we propose a technique based on singular value decomposition (SVD) to make the FSL input constant irrespective of the number of receive antennas. We prototype the ReWiS using off-the-shelf Wi-Fi equipment and showcase its performance by considering a compelling use case of human activity recognition. Thus, we perform an extensive data collection campaign in three different propagation environments with two human subjects. We evaluate the impact of each diversity component on the performance and compare ReWiS with an existing convolutional neural network (CNN)-based approach. Experimental results show that ReWiS improves the performance by about 40% with respect to existing single-antenna low-resolution approaches. Moreover, when compared to a CNN-based approach, ReWiS shows a 35% more accuracy and less than 10% drop in accuracy when tested in different environments, while the CNN drops by more than 45%. To allow reproducibility of our results and to address the current dearth of Wi-Fi sensing datasets, we pledge to release our 60 GB dataset and the entire code repository to the community.
@inproceedings{bahadori2022rewis,
title={{ReWiS: Reliable Wi-Fi Sensing Through Few-Shot Multi-Antenna Multi-Receiver CSI Learning}},
author={Bahadori, Niloofar and Ashdown, Jonathan and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)},
pages={50--59},
year={2022},
organization={IEEE}
} 2022
Conference
SmartDet: Context-Aware Dynamic Control of Edge Task Offloading for Mobile Object Detection
D. Callegaro, M. Levorato, and F. Restuccia
Proceedings of the IEEE International Symposium on a World of Wireless, Mobile, and Multimedia Networks (WoWMoM)
Mobile devices such as drones and autonomous vehicles increasingly rely on object detection (OD) through deep neural networks (DNNs) to perform critical tasks such as navigation, target-tracking and surveillance, just to name a few. Due to their high complexity, the execution of these DNNs requires excessive time and energy. Low-complexity object tracking (OT) is thus used along with OD, where the latter is periodically applied to generate "fresh" references for tracking. However, the frames processed with OD incur large delays, which does not comply with real-time applications requirements. Offloading OD to edge servers can mitigate this issue, but existing work focuses on the optimization of the offloading process in systems where the wireless channel has a very large capacity. Herein, we consider systems with constrained and erratic channel capacity, and establish parallel OT (at the mobile device) and OD (at the edge server) processes that are resilient to large OD latency. We propose Katch-Up, a novel tracking mechanism that improves the system resilience to excessive OD delay. We show that this technique greatly improves the quality of the reference available to tracking, and boosts performance up to 33%. However, while Katch-Up significantly improves performance, it also increases the computing load of the mobile device. Hence, we design SmartDet, a low-complexity controller based on deep reinforcement learning (DRL) that learns to achieve the right trade-off between resource utilization and OD performance. SmartDet takes as input highly-heterogeneous context-related information related to the current video content and the current network conditions to optimize frequency and type of OD offloading, as well as Katch-Up utilization. We extensively evaluate SmartDet on a real-world testbed composed by a JetSon Nano as mobile device and a GTX 980 Ti as edge server, connected through a Wi-Fi link, to collect several network-related traces, as well as energy measurements. We consider a state-of-the-art video dataset (ILSVRC 2015 - VID) and state-of-the-art OD models (EfficientDet 0, 2 and 4). Experimental results show that SmartDet achieves an optimal balance between tracking performance – mean Average Recall (mAR) and resource usage. With respect to a baseline with full Katch-Up usage and maximum channel usage, we still increase mAR by 4% while using 50% less of the channel and 30% power resources associated with Katch-Up. With respect to a fixed strategy using minimal resources, we increase mAR by 20% while using Katch-Up on 1/3 of the frames.
@inproceedings{callegaro2022smartdet,
title={{SmartDet: Context-Aware Dynamic Control of Edge Task Offloading for Mobile Object Detection}},
author={Callegaro, Davide and Levorato, Marco and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)},
pages={357--366},
year={2022},
organization={IEEE}
} 2022
Conference
BottleFit: Learning Compressed Representations in Deep Neural Networks for Effective and Efficient Split Computing
Y. Matsubara, D. Callegaro, S. Singh, M. Levorato, and F. Restuccia
Proceedings of the IEEE International Symposium on a World of Wireless, Mobile, and Multimedia Networks (WoWMoM)
Although mission-critical applications require the use of deep neural networks (DNNs), their continuous execution at mobile devices results in a significant increase in energy consumption. While edge offloading can decrease energy consumption, erratic patterns in channel quality, network and edge server load can lead to severe disruption of the system’s key operations. An alternative approach, called split computing, generates compressed representations within the model (called "bottlenecks"), to reduce bandwidth usage and energy consumption. Prior work has proposed approaches that introduce additional layers, to the detriment of energy consumption and latency. For this reason, we propose a new framework called BottleFit, which, in addition to targeted DNN architecture modifications, includes a novel training strategy to achieve high accuracy even with strong compression rates. We apply BottleFit on cutting-edge DNN models in image classification, and show that BottleFit achieves 77.1% data compression with up to 0.6% accuracy loss on ImageNet dataset, while state of the art such as SPINN loses up to 6% in accuracy. We experimentally measure the power consumption and latency of an image classification application running on an NVIDIA Jetson Nano board (GPU-based) and a Raspberry PI board (GPU-less). We show that BottleFit decreases power consumption and latency respectively by up to 49% and 89% with respect to (w.r.t.) local computing and by 37% and 55% w.r.t. edge offloading. We also compare BottleFit with state-of-the-art autoencoders-based approaches, and show that (i) BottleFit reduces power consumption and execution time respectively by up to 54% and 44% on the Jetson and 40% and 62% on Raspberry PI; (ii) the size of the head model executed on the mobile device is 83 times smaller. We publish the code repository for reproducibility of the results in this study.
@inproceedings{matsubara2022bottlefit,
title={{BottleFit: Learning Compressed Representations in Deep Neural Networks for Effective and Efficient Split Computing}},
author={Matsubara, Yoshitomo and Callegaro, Davide and Singh, Sameer and Levorato, Marco and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)},
pages={337--346},
year={2022},
organization={IEEE}
} 2022
Conference
ChARM: NextG Spectrum Sharing Through Data-Driven Real-Time O-RAN Dynamic Control
L. Baldesi, F. Restuccia and T. Melodia
Proceedings of IEEE Conference on Computer Communications (INFOCOM) — Best Paper Award
Today’s radio access networks (RANs) are monolithic entities which often operate statically on a given set of parameters for the entirety of their operations. To implement realistic and effective spectrum sharing policies, RANs will need to seamlessly and intelligently change their operational parameters. In stark contrast with existing paradigms, the new O-RAN architectures for 5G-and-beyond networks (NextG) separate the logic that controls the RAN from its hardware substrate, allowing unprecedented real-time fine-grained control of RAN components. In this context, we propose the Channel-Aware Reactive Mechanism (ChARM), a data-driven O-RAN-compliant framework that allows (i) sensing the spectrum to infer the presence of interference and (ii) reacting in real time by switching the distributed unit (DU) and radio unit (RU) operational parameters according to a specified spectrum access policy. ChARM is based on neural networks operating directly on unprocessed I/Q waveforms to determine the current spectrum context. ChARM does not require any modification to the existing 3GPP standards. It is designed to operate within the O-RAN specifications, and can be used in conjunction with other spectrum sharing mechanisms (e.g., LTE-U, LTE-LAA or MulteFire). We demonstrate the performance of ChARM in the context of spectrum sharing among LTE and Wi-Fi in unlicensed bands, where a controller operating over a RAN Intelligent Controller (RIC) senses the spectrum and switches cell frequency to avoid Wi-Fi. We develop a prototype of ChARM using srsRAN, and leverage the Colosseum channel emulator to collect a large-scale waveform dataset to train our neural networks with. To collect standard-compliant Wi-Fi data, we extended the Colosseum testbed using system-on-chip (SoC) boards running a modified version of the OpenWiFi architecture. Experimental results show that ChARM achieves accuracy of up to 96% on Colosseum and 85% on an over-the-air testbed, demonstrating the capacity of ChARM to exploit the considered spectrum channels.
@inproceedings{baldesi2022charm,
title={{ChARM: NextG Spectrum Sharing Through Data-Driven Real-Time O-RAN Dynamic Control}},
author={Baldesi, Luca and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (INFOCOM)},
pages={240--249},
year={2022},
organization={IEEE}
} 2022
Journal
Dynamic Spectrum Sharing Between Active and Passive Users Above 100 GHz
M. Polese, V. Ariyarathna, P. Sen, J. V. Siles, F. Restuccia, T. Melodia, J. M. Jornet
Nature Communications Engineering
Sixth-generation wireless networks will aggregate higher-than-ever mobile traffic into ultra-high capacity backhaul links, which could be deployed on the largely untapped spectrum above 100 GHz. Current regulations however prevent the allocation of large contiguous bands for communications at these frequencies, since several narrow bands are reserved to protect passive sensing services. These include radio astronomy and Earth exploration satellites using sensors that suffer from harmful interference from active transmitters. Here we show that active and passive spectrum sharing above 100 GHz is feasible by introducing and experimentally evaluating a real-time, dual-band backhaul prototype that tracks the presence of passive users (in this case the NASA satellite Aura) and avoids interference by automatically switching bands (123.5–140 GHz and 210–225 GHz). Our system enables wide-band transmissions in the above-100-GHz spectrum, while avoiding harmful interference to satellite systems, paving the way for innovative spectrum policy and technologies in these crucial bands.
@article{polese2022dynamic,
title={{Dynamic Spectrum Sharing Between Active And Passive Users Above 100 GHz}},
author={Polese, Michele and Ariyarathna, Viduneth and Sen, Priyangshu and Siles, Jose V and Restuccia, Francesco and Melodia, Tommaso and Jornet, Josep M},
journal={Communications Engineering},
volume={1},
number={1},
pages={6},
year={2022},
publisher={Nature Publishing Group UK London}
} 2022
Conference
AiEEG: Personalized Seizure Prediction Through Partially-Reconfigurable Deep Neural Networks
D. Uvaydov, R. Guida, P. Johari, F. Restuccia and T. Melodia
Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom)
With more than 65M people affected by epilepsy worldwide, early prediction and response to seizure onsets have become more important than ever. Cutting-edge research in implantable medical devices (IMDs) has shown that deep neural networks (DNNs) applied to intracranial electroencephalogram (iEEG) data can predict seizures up to an hour before onset. However, offloading of iEEG data to the edge/cloud is highly prohibitive, due to the sheer size of the generated data. Existing work either focuses on the DNN training phase only, or does not consider the severe energy/space limitations of IMDs. Moreover, the technical aspects of patient personalization, which allows for patient-specific hyper-parameter tuning, still remain unaddressed. In this paper, we propose a platform called AiEEG for in vivo early seizure prediction, whose DNN hardware circuitry can be reconfigured remotely without surgery. We prototype AiEEG on a system on chip (SoC) platform and demonstrate its end-to-end capabilities in seizure prediction with a population of 30 epileptic patients, with iEEG signals coming from a real-world dataset. Extensive experimental results shows that (i) our embedded and personalized DNN has an area under the curve (AUC) averaging at 0.97 and as low as zero false positives per hour (FPH) for over half the patients, an improvement of about 3.5x with respect to a non-personalized prediction method and the best for a dataset of this size when compared to the state-of-the-art; (ii) our AiEEG platform consumes 4.2x less energy than a cloud-based approach, leading to a 4x battery lifetime improvement; (iii) we are able to remotely fine-tune the DNN through partial reconfiguration as needed in about 10s.
@inproceedings{uvaydov2022aieeg,
title={{AiEEG: Personalized Seizure Prediction Through Partially-Reconfigurable Deep Neural Networks}},
author={Uvaydov, Daniel and Guida, Raffaele and Johari, Pedram and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE International Conference on Pervasive Computing and Communications (PerCom)},
pages={77--88},
year={2022},
organization={IEEE}
}
2022
Conference
Colosseum: Large-Scale Wireless Experimentation Through Hardware-in-the-Loop Network Emulation
L. Bonati, P. Johari, M. Polese, S. D’Oro, S. Mohanti, M. Tehrani-Moayyed, D. Villa, S. Shrivastava, C. Tassie, K. Yoder, A. Bagga, P. Patel, V. Petkov, M. Seltser, F. Restuccia, A. Gosain, K.R. Chowdhury, S. Basagni, T. Melodia
Proceedings of IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)
Colosseum is an open-access and publicly-available large-scale wireless testbed for experimental research via virtualized and softwarized waveforms and protocol stacks on a fully programmable, “white-box” platform. Through 256 state-of-the-art software-defined radios and a massive channel emulator core, Colosseum can model virtually any scenario, enabling the design, development and testing of solutions at scale in a variety of deployments and channel conditions. These Colosseum radio-frequency scenarios are reproduced through high-fidelity FPGAbased emulation with finite-impulse response filters. Filters model the taps of desired wireless channels and apply them to the signals generated by the radio nodes, faithfully mimicking the conditions of real-world wireless environments. In this paper, we introduce Colosseum as a testbed that is for the first time open to the research community. We describe the architecture of Colosseum and its experimentation and emulation capabilities. We then demonstrate the effectiveness of Colosseum for experimental research at scale through exemplary use cases including prevailing wireless technologies (e.g., cellular and Wi-Fi) in spectrum sharing and unmanned aerial vehicle scenarios. A roadmap for Colosseum future updates concludes the paper.
@inproceedings{bonati2021colosseum,
author = {Bonati, Leonardo and Johari, Pedram and Polese, Michele and D'Oro, Salvatore and Mohanti, Subhramoy and Tehrani-Moayyed, Miead and Villa, Davide and Shrivastava, Shweta and Tassie, Chinenye and Yoder, Kurt and Bagga, Ajeet and Patel, Paresh and Petkov, Ventz and Seltser, Michael and Restuccia, Francesco and Gosain, Abhimanyu and Chowdhury, Kaushik R. and Basagni, Stefano and Melodia, Tommaso},
booktitle = {Proceedings of IEEE International Symposium on Dynamic Spectrum Access Networks (DySPAN)},
organization = {IEEE},
pages = {105--113},
title = {{Colosseum: Large-Scale Wireless Experimentation Through Hardware-in-the-Loop Network Emulation}},
year = {2021}} 2021
Journal
DeepFIR: Channel-Robust Physical-Layer Deep Learning Through Adaptive Wave-form Filtering
F. Restuccia, S. D’Oro, A. Al-Shawabka, B. Costa Rendon, S. Ioannidis and T. Melodia
IEEE Transactions on Wireless Communications
Deep learning can be used to classify waveform characteristics ( e.g. , modulation) with accuracy levels that are hardly attainable with traditional techniques. Recent research has demonstrated that one of the most crucial challenges in wireless deep learning is to counteract the channel action, which may significantly alter the waveform features. The problem is further exacerbated by the fact that deep learning algorithms are hardly re-trainable in real time due to their sheer size. This paper proposes DeepFIR , a framework to counteract the channel action in wireless deep learning algorithms without retraining the underlying deep learning model . The key intuition is that through the application of a carefully-optimized digital finite input response filter (FIR) at the transmitter’s side, we can apply tiny modifications to the waveform to strengthen its features according to the current channel conditions. We mathematically formulate the Waveform Optimization Problem as the problem of finding the optimum FIR to be used on a waveform to improve the classifier’s accuracy. We also propose a data-driven methodology to train the FIRs directly with dataset inputs. We extensively evaluate DeepFIR on an experimental testbed of 20 software-defined radios, as well as on two datasets made up by 500 ADS-B devices and by 500 WiFi devices and a 24-class modulation dataset. Experimental results show that our approach (i) increases the accuracy of the radio fingerprinting models by about 35%, 50% and 58%; (ii) decreases an adversary’s accuracy by about 54% when trying to imitate other device’s fingerprints by using their filters; (iii) achieves 27% improvement over the state of the art on a 100-device dataset; (iv) increases by the accuracy of the modulation dataset.
@article{restuccia2021deepfir,
title={{DeepFIR: Channel-Robust Physical-Layer Deep Learning Through Adaptive Waveform Filtering}},
author={Restuccia, Francesco and D’Oro, Salvatore and Al-Shawabka, Amani and Rendon, Bruno Costa and Ioannidis, Stratis and Melodia, Tommaso},
journal={IEEE Transactions on Wireless Communications},
volume={20},
number={12},
pages={8054--8066},
year={2021},
publisher={IEEE}
} 2021
Patent
Real-Time Channel-Resilient Optimization of Radio Fingerprinting
F. Restuccia, S. D’Oro and T. Melodia
U.S. Patent No.: 11,184,783
2021
Journal
Coordinated 5G Network Slicing: How Constructive Interference Can Boost Network Throughput
S. D’Oro, L. Bonati, F. Restuccia, and T. Melodia
IEEE/ACM Transactions on Networking
Radio access network (RAN) slicing is a virtualization technology that partitions radio resources into multiple autonomous virtual networks. Since RAN slicing can be tailored to provide diverse performance requirements, it will be pivotal to achieve the high-throughput and low-latency communications that next-generation (5G) systems have long yearned for. To this end, effective RAN slicing algorithms must (i) partition radio resources so as to leverage coordination among multiple base stations and thus boost network throughput; and (ii) reduce interference across different slices to guarantee slice isolation and avoid performance degradation. The ultimate goal of this paper is to design RAN slicing algorithms that address the above two requirements. First, we show that the RAN slicing problem can be formulated as a 0-1 Quadratic Programming problem, and we prove its NP-hardness. Second, we propose an optimal solution for small-scale 5G network deployments, and we present three approximation algorithms to make the optimization problem tractable when the network size increases. We first analyze the performance of our algorithms through simulations, and then demonstrate their performance through experiments on a standard-compliant LTE testbed with 2 base stations and 6 smartphones. Our results show that not only do our algorithms efficiently partition RAN resources, but also improve network throughput by 27% and increase by 2× the signal-to-interference-plus-noise ratio.
@article{d2021coordinated,
title={{Coordinated 5G Network Slicing: How Constructive Interference Can Boost Network Throughput},
author={D’Oro, Salvatore and Bonati, Leonardo and Restuccia, Francesco and Melodia, Tommaso},
journal={IEEE/ACM Transactions on Networking},
volume={29},
number={4},
pages={1881--1894},
year={2021},
publisher={IEEE}
} 2021
Conference
SeReMAS: Self-Resilient Mobile Autonomous Systems Through Predictive Edge Computing
D. Callegaro, M. Levorato, and F. Restuccia
Proceedings of IEEE International Conference on Sensing, Communication and Networking (SECON)
Edge computing enables Mobile Autonomous Systems (MASs) to execute continuous streams of heavy-duty mission-critical processing tasks, such as real-time obstacle detection and navigation. However, in practical applications, erratic patterns in channel quality, network load, and edge server load can interrupt the task flow's execution, which necessarily leads to severe disruption of the system's key operations. Existing work has mostly tackled the problem with reactive approaches, which cannot guarantee task-level reliability. Conversely, in this paper we focus on learning-based predictive edge computing to achieve self-resilient task offloading. By conducting a preliminary experimental evaluation, we show that there is no dominant feature that can predict the edge-MAS system reliability, which calls for an ensemble and selection of weaker features. To tackle the complexity of the problem, we propose SeReMAS, a data-driven optimization framework. We first mathematically formulate a Redundant Task Offloading Problem (RTOP), where a MAS may connect to multiple edge servers for redundancy, and needs to select which server(s) to transmit its computing tasks in order to maximize the probability of task execution while minimizing channel and edge resource utilization. We then create a predictor based on Deep Reinforcement Learning (DRL), which produces the optimum task assignment based on application-, network- and telemetry-based features. We prototype SeReMAS on a testbed composed by a Tarot650 quadcopter drone, mounting a PixHawk flight controller, a Jetson Nano board, and three 802.11n WiFi interfaces. We extensively evaluate SeReMAS by considering an application where one drone offloads high-resolution images for real-time analysis to three edge servers on the ground. Experimental results show that SeReMAS improves the task execution probability by 17% with respect to existing reactive-based approaches. To allow full reproducibility of results, we share the dataset and code with the research community.
@inproceedings{callegaro2021seremas,
title={{SeReMAS: Self-Resilient Mobile Autonomous Systems Through Predictive Edge Computing}},
author={Callegaro, Davide and Levorato, Marco and Restuccia, Francesco},
booktitle={Proceedings of IEEE International Conference on Sensing, Communication, and Networking (SECON)},
pages={1--9},
year={2021},
organization={IEEE}
} 2021
Conference
DeepLoRa: Fingerprinting LoRa Devices at Scale Through Deep Learning and Data Augmentation
A. Al-Shawabka, P. Pietraski, S.B. Pattar, F. Restuccia and T. Melodia
Proceedings of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)
The Long Range (LoRa) protocol for low-power wide-area networks (LPWANs) is a strong candidate to enable the massive roll-out of the Internet of Things (IoT) because of its low cost, impressive sensitivity (-137dBm), and massive scalability potential. As tens of thousands of tiny LoRa devices are deployed over large geographic areas, a key component to the success of LoRa will be the development of reliable and robust authentication mechanisms. To this end, Radio Frequency Fingerprinting (RFFP) through deep learning (DL) has been heralded as an effective zero-power supplement or alternative to energy-hungry cryptography. Existing work on LoRa RFFP has mostly focused on small-scale testbeds and low-dimensional learning techniques; however, many challenges remain. Key among them are authentication techniques robust to a wide variety of channel variations over time and supporting a vast population of devices. In this work, we advance the state of the art by presenting (i) the first massive experimental evaluation of DL RFFP and (ii) new data augmentation techniques for LoRa designed to counter the degradation introduced by the wireless channel. Specifically, we collected and publicly shared more than 1TB of waveform data from 100 bit-similar devices (with identical manufacturing processes) over different deployment scenarios (outdoor vs. indoor) and spanning several days. We train and test diverse DL models (convolutional and recurrent neural networks) using either preamble or payload data slices. We compare three different representations of the received signal: (i) IQ, (ii) amplitude-phase, and (iii) spectrogram. Finally, we propose a novel data augmentation technique called DeepLoRa to enhance the LoRa RFFP performance. Results show that (i) training the CNN models with IQ representation is not always the best combo in fingerprinting LoRa radios; training CNNs and RNN-LSTMs with amplitude-phase and spectrogram representations may increase the fingerprinting performance in small and medium-scale testbeds; (ii) using only payload data in the fingerprinting process outperforms preamble only data, and (iii) DeepLoRa data augmentation technique improves the classification accuracy from 19% to 36% in the RFFP challenging case of training on data collected on a different day than the testing data. Moreover, DeepLoRa raises the accuracy from 82% to 91% when training and testing 100 devices with data collected on the same day.
@inproceedings{al2021deeplora,
title={{DeepLoRa: Fingerprinting LoRa Devices at Scale Through Deep Learning and Data Augmentation}},
author={Al-Shawabka, Amani and Pietraski, Philip and Pattar, Sudhir B and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)},
pages={251--260},
year={2021}
} 2021
Conference
DeepBeam: Deep Waveform Learning for Coordination-Free Beam Management in mmWave Networks
M. Polese, F. Restuccia, and T. Melodia
Proceedings of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)
Highly directional millimeter wave (mmWave) radios need to perform beam management to establish and maintain reliable links. To achieve this objective, existing solutions mostly rely on explicit coordination between the transmitter (TX) and the receiver (RX), which significantly reduces the airtime available for communication and further complicates the network protocol design. This paper advances the state of the art by presenting DeepBeam, a framework for beam management that does not require pilot sequences from the TX, nor any beam sweeping or synchronization from the RX. This is achieved by inferring (i) the Angle of Arrival (AoA) of the beam and (ii) the actual beam being used by the transmitter through waveform-level deep learning on ongoing transmissions between the TX to other receivers. In this way, the RX can associate Signal-to-Noise-Ratio (SNR) levels to beams without explicit coordination with the TX. This is possible because different beam patterns introduce different "impairments" to the waveform, which can be subsequently learned by a convolutional neural network (CNN). To demonstrate the generality of DeepBeam, we conduct an extensive experimental data collection campaign where we collect more than 4 TB of mmWave waveforms with (i) 4 phased array antennas at 60.48 GHz, (ii) 2 codebooks containing 24 one-dimensional beams and 12 two-dimensional beams; (iii) 3 receiver gains; (iv) 3 different AoAs; (v) multiple TX and RX locations. Moreover, we collect waveform data with two custom-designed mmWave software-defined radios with fully-digital beamforming architectures at 58 GHz. We also implement our learning models in FPGA to evaluate latency performance. Results show that DeepBeam (i) achieves accuracy of up to 96%, 84% and 77% with a 5-beam, 12-beam and 24-beam codebook, respectively; (ii) reduces latency by up to 7x with respect to the 5G NR initial beam sweep in a default configuration and with a 12-beam codebook. The waveform dataset and the full DeepBeam code repository are publicly available.
@inproceedings{polese2021deepbeam,
title={{DeepBeam: Deep Waveform Learning for Coordination-Free Beam Management in mmWave Networks}},
author={Polese, Michele and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)},
pages={61--70},
year={2021}
}
2021
Conference
The Tags Are Alright: Robust Large-Scale RFID Clone Detection Through Federated Data-Augmented Radio Fingerprinting
M. Piva, G. Maselli, and F. Restuccia
Proceedings of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)
Millions of RFID tags are pervasively used all around the globe to inexpensively identify a wide variety of everyday-use objects. One of the key issues of RFID is that tags cannot use energy-hungry cryptography, and thus can be easily cloned. For this reason, radio fingerprinting (RFP) is a compelling approach that leverages the unique imperfections in the tag's wireless circuitry to achieve large-scale RFID clone detection. Recent work, however, has unveiled that time-varying channel conditions can significantly decrease the accuracy of the RFP process. Prior art in RFID identification does not consider this critical aspect, and instead focuses on custom-tailored feature extraction techniques and data collection with static channel conditions. For this reason, we propose the first large-scale investigation into RFP of RFID tags with dynamic channel conditions. Specifically, we perform a massive data collection campaign on a testbed composed by 200 off-the-shelf identical RFID tags and a software-defined radio (SDR) tag reader. We collect data with different tag-reader distances in an over-the-air configuration. To emulate implanted RFID tags, we also collect data with two different kinds of porcine meat inserted between the tag and the reader. We use this rich dataset to train and test several convolutional neural network (CNN)-based classifiers in a variety of channel conditions. Our investigation reveals that training and testing on different channel conditions drastically degrades the classifier's accuracy. For this reason, we propose a novel training framework based on federated machine learning (FML) and data augmentation (DAG) to boost the accuracy. Extensive experimental results indicate that (i) our FML approach improves accuracy by up to 48%; (ii) our DAG approach improves the FML performance by up to 19% and the single-dataset performance by 31%. To the best of our knowledge, this is the first paper experimentally demonstrating the efficacy of FML and DAG on a large device population. To allow full replicability, we are sharing with the research community our fully-labeled 200-GB RFID waveform dataset, as well as the entirety of our code and trained models, concurrently with our submission.
@inproceedings{piva2021tags,
title={{The Tags Are Alright: Robust Large-Scale RFID Clone Detection Through Federated Data-Augmented Radio Fingerprinting}},
author={Piva, Mauro and Maselli, Gaia and Restuccia, Francesco},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)},
pages={41--50},
year={2021}
} 2021
Conference
A Blockchain Definition to Clarify Its Role for the Internet of Things
L. Ghiro, F. Restuccia, S. D’Oro, S. Basagni, T. Melodia, L. Maccari, and R. Lo Cigno
Mediterranean Communication and Computer Networking Conference (MedComNet)
The term blockchain is used for disparate projects, ranging from cryptocurrencies to applications for the Internet of Things (IoT). The concept of blockchain appears therefore blurred, as the same technology cannot empower applications with extremely different requirements, levels of security and performance. This position paper elaborates on the theory of distributed systems to advance a clear definition of blockchain allowing us to clarify its possible role in the IoT. The definition binds together three elements that, as a whole, delineate those unique features that distinguish the blockchain from other distributed ledger technologies: immutability, transparency and anonymity. We note that immutability-which is imperative for securing blockchains-imposes remarkable resource consumption. Moreover, while transparency demands no confidentiality, anonymity enhances privacy but prevents user identification. As such, we raise the concern that these blockchain features clash with the requirements of most IoT applications where devices are power-constrained, data needs to be kept confidential, and users to be clearly identifiable. We consequently downplay the role of the blockchain for the IoT: this can act as a ledger external to the IoT architecture, invoked as seldom as possible and only to record the aggregate results of myriads of local (IoT) transactions that are most of the time performed off-chain to meet performance and scalability requirements.
@inproceedings{ghiro2021blockchain,
title={{A Blockchain Definition to Clarify Its Role for the Internet of Things}},
author={Ghiro, Lorenzo and Restuccia, Francesco and D'Oro, Salvatore and Basagni, Stefano and Melodia, Tommaso and Maccari, Leonardo and Cigno, Renato Lo},
booktitle={Mediterranean Communication and Computer Networking Conference (MedComNet)},
pages={1--8},
year={2021},
organization={IEEE}
} 2021
Conference
Can You Fix My Neural Network? Real-Time Adaptive Waveform Synthesis for Resilient Wireless Signal Classification
S. D’Oro, F. Restuccia and T. Melodia
Proceedings of IEEE Conference on Computer Communications (INFOCOM)
Due to the sheer scale of the Internet of Things (IoT) and 5G, the wireless spectrum is becoming severely congested. For this reason, wireless devices will need to continuously adapt to current spectrum conditions by changing their communication parameters in real-time. Therefore, wireless signal classification (WSC) will become a compelling necessity to decode fast-changing signals from dynamic transmitters. Thanks to its capability of classifying complex phenomena without explicit mathematical modeling, deep learning (DL) has been demonstrated to be a key enabler of WSC. Although DL can achieve a very high accuracy under certain conditions, recent research has unveiled that the wireless channel can disrupt the features learned by the DL model during training, thus drastically reducing the classification performance in real-world live settings. Since retraining classifiers is cumbersome after deployment, existing work has leveraged the usage of carefully-tailored Finite Impulse Response (FIR) filters that, when applied at the transmitter’s side, can restore the features that are lost because of the the channel actions, i.e., waveform synthesis. However, these approaches compute FIRs using offline optimization strategies, which limits their efficacy in highly-dynamic channel settings. In this paper, we improve the state of the art by proposing Chares, a Deep Reinforcement Learning (DRL)-based framework for channel-resilient adaptive waveform synthesis. Chares adapts to new and unseen channel conditions by optimally computing through DRL the FIRs in real time. Chares is a DRL agent whose architecture is based upon the Twin Delayed Deep Deterministic Policy Gradients (TD3), which requires minimal feedback from the receiver and explores a continuous action space for best performance. Chares has been extensively evaluated on two well-known datasets with an extensive number of channels. We have also evaluated the real-time latency of Chares with an implementation on field-programmable gate array (FPGA). Results show that Chares increases the accuracy up to 4.1x when no waveform synthesis is performed, by 1.9x with respect to existing work, and can compute new actions within 41µs.
@inproceedings{d2021can,
title={{Can You Fix My Neural Network? Real-Time Adaptive Waveform Synthesis for Resilient Wireless Signal Classification}},
author={D’Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (INFOCOM)},
pages={1--10},
year={2021},
organization={IEEE}
} 2021
Conference
SteaLTE: Private 5G Cellular Connectivity as a Service with Full-stack Wireless Steganography
L. Bonati, S. D’Oro, F. Restuccia, S. Basagni and T. Melodia
Proceedings of IEEE Conference on Computer Communications (INFOCOM)
Fifth-generation (5G) systems will extensively employ radio access network (RAN) softwarization. This key innovation enables the instantiation of "virtual cellular networks" running on different slices of the shared physical infrastructure. In this paper, we propose the concept of Private Cellular Connectivity as a Service (PCCaaS), where infrastructure providers deploy covert network slices known only to a subset of users. We then present SteaLTE as the first realization of a PCCaaS-enabling system for cellular networks. At its core, SteaLTE utilizes wireless steganography to disguise data as noise to adversarial receivers. Differently from previous work, however, it takes a full-stack approach to steganography, contributing an LTE-compliant stegano-graphic protocol stack for PCCaaS-based communications, and packet schedulers and operations to embed covert data streams on top of traditional cellular traffic (primary traffic). SteaLTE balances undetectability and performance by mimicking channel impairments so that covert data waveforms are almost indistinguishable from noise. We evaluate the performance of SteaLTE on an indoor LTE-compliant testbed under different traffic profiles, distance and mobility patterns. We further test it on the outdoor PAWR POWDER platform over long-range cellular links. Results show that in most experiments SteaLTE imposes little loss of primary traffic throughput in presence of covert data transmissions (<; 6%), making it suitable for undetectable PCCaaS networking.
@inproceedings{bonati2021stealte,
title={{SteaLTE: Private 5G Cellular Connectivity as a Service with Full-stack Wireless Steganography}},
author={Bonati, Leonardo and D’Oro, Salvatore and Restuccia, Francesco and Basagni, Stefano and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (INFOCOM)},
pages={1--10},
year={2021},
organization={IEEE}
} 2021
Conference
DeepSense: Fast Wideband Spectrum Sensing Through Real-Time In-the-Loop Deep Learning
D. Uvaydov, S. D’Oro, F. Restuccia and T. Melodia
Proceedings of IEEE Conference on Computer Communications (INFOCOM)
Spectrum sharing will be a key technology to tackle spectrum scarcity in the sub-6 GHz bands. To fairly access the shared bandwidth, wireless users will necessarily need to quickly sense large portions of spectrum and opportunistically access unutilized bands. The key unaddressed challenges of spectrum sensing are that (i) it has to be performed with extremely low latency over large bandwidths to detect tiny spectrum holes and to guarantee strict real-time digital signal processing (DSP) constraints; (ii) its underlying algorithms need to be extremely accurate, and flexible enough to work with different wireless bands and protocols to find application in real-world settings. To the best of our knowledge, the literature lacks spectrum sensing techniques able to accomplish both requirements. In this paper, we propose DeepSense, a software/hardware framework for real-time wideband spectrum sensing that relies on real-time deep learning tightly integrated into the transceiver’s baseband processing logic to detect and exploit unutilized spectrum bands. DeepSense uses a convolutional neural network (CNN) implemented in the wireless platform’s hardware fabric to analyze a small portion of the unprocessed baseband waveform to automatically extract the maximum amount of information with the least amount of I/Q samples. We extensively validate the accuracy, latency and generality performance of DeepSense with (i) a 400 GB dataset containing hundreds of thousands of WiFi transmissions collected "in the wild" with different Signal-to-Noise-Ratio (SNR) conditions and over different days; (ii) a dataset of transmissions collected using our own software-defined radio testbed; and (iii) a synthetic dataset of LTE transmissions under controlled SNR conditions. We also measure the real-time latency of the CNNs trained on the three datasets with an FPGA implementation, and compare our approach with a fixed energy threshold mechanism. Results show that our learning-based approach can deliver a precision and recall of 98% and 97% respectively and a latency as low as 0.61ms. For reproducibility and benchmarking purposes, we pledge to share the code and the datasets used in this paper to the community.
@inproceedings{uvaydov2021deepsense,
title={{DeepSense: Fast Wideband Spectrum Sensing Through Real-Time In-the-Loop Deep Learning}},
author={Uvaydov, Daniel and D’Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (INFOCOM)},
pages={1--10},
year={2021},
organization={IEEE}
} 2021
Journal
Arena: A 64-antenna SDR-based Ceiling Grid Testing Platform for Sub-6 GHz 5G-and-Beyond Radio Spectrum Research
L. Bertizzolo, L. Bonati, E. Demirors, A. Al-Shawabka, S. D’Oro, F. Restuccia, and T. Melodia
Computer Networks
Arena is an open-access wireless testing platform based on a grid of antennas mounted on the ceiling of a large office-space environment. Each antenna is connected to programmable software-defined radios (SDR) enabling sub-6 GHz 5G-and-beyond spectrum research. With 12 computational servers, 24 SDRs synchronized at the symbol level, and a total of 64 antennas, Arena provides the computational power and the scale to foster new technology development in some of the most crowded spectrum bands. Arena is based on a three-tier design, where the servers and the SDRs are housed in a double rack in a dedicated room, while the antennas are hung off the ceiling of a 2240 square feet office space and cabled to the radios through 100 ft-long cables. This ensures a reconfigurable, scalable, and repeatable real-time experimental evaluation in a real wireless indoor environment.In this paper, we introduce the architecture, capabilities, and system design choices of Arena, and provides details of the software and hardware implementation of various testbed components. Furthermore, we describe key capabilities by providing examples of published work that employed Arena for applications as diverse as synchronized MIMO transmission schemes, multi-hop ad hoc networking, multi-cell 5G networks, AI-powered Radio-Frequency fingerprinting, secure wireless communications, and spectrum sensing for cognitive radio.
@article{bertizzolo2020arena,
title={{Arena: A 64-antenna SDR-based Ceiling Grid Testing Platform for Sub-6 GHz 5G-and-Beyond Radio Spectrum Research}},
author={Bertizzolo, Lorenzo and Bonati, Leonardo and Demirors, Emrecan and Al-Shawabka, Amani and D’Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso},
journal={Computer Networks},
volume={181},
pages={107436},
year={2020},
publisher={Elsevier}
} 2020
Conference
HyBloSE: Hybrid Blockchain for Secure-by-Design Smart Environments
G. Maselli, M. Piva and F. Restuccia
Proceedings of ACM Workshop on Cryptocurrencies and Blockchains for Distributed Systems (CryBlock)
Although smart environments are a key component of the Internet of Things (IoT), it is also clear that billions connected doors, washing machines, ovens and others will ultimately raise security and privacy concerns. Early work in this area, as well as most of commercial solutions, has adopted a centralized client/server approach, neglecting the multitude of risks that are induced by an unfair control of the server side. This has made the adoption of a decentralized and trust-less framework quintessential to guarantee devices security. Nevertheless, decentralized proposals are hardly applicable due to costs, slowness and privacy issues. In this paper, we make the use of blockchain practical for smart environments by designing HyBloSE, a secure-by-design and lightweight blockchain-based framework, able to run on low-power devices without additional hardware. HyBloSE is built by using Delegated Proof of Authority and a Moving Window Blockchain. We evaluate HyBloSE through a network emulator and real experiments with different Raspberry Pi platforms. Results show that HyBloSE guarantees a higher security level in terms of resiliency to internal and external attacks compared to centralized solutions, with overhead below 0.38s per operation and less than $4 per month for unlimited operations. Furthermore, we show how Proof of Authority is more adapt then Proof of Work in IoT private scenarios.
@inproceedings{maselli2020hyblose,
title={{HyBloSE: Hybrid Blockchain for Secure-by-Design Smart Environments}},
author={Maselli, Gaia and Piva, Mauro and Restuccia, Francesco},
booktitle={Proceedings of ACM Workshop on Cryptocurrencies and Blockchains for Distributed Systems (CryBlock)},
pages={23--28},
year={2020}
} 2020
Journal
Deep Learning at the Physical Layer: System Challenges and Applications to 5G and Beyond
F. Restuccia and T. Melodia
IEEE Communications Magazine
The unprecedented requirements of IoT have made fine-grained optimization of spectrum resources an urgent necessity. Thus, designing techniques able to extract knowledge from the spectrum in real time and select the optimal spectrum access strategy accordingly has become more important than ever. Moreover, 5G networks will require complex management schemes to deal with problems such as adaptive beam management and rate selection. Although deep learning (DL) has been successful in modeling complex phenomena, commercially available wireless devices are still very far from actually adopting learning-based techniques to optimize their spectrum usage. In this article, we first discuss the need for real-time DL at the physical layer, and then summarize the current state of the art and existing limitations. We conclude the article by discussing an agenda of research challenges and how DL can be applied to address crucial problems in 5G and beyond networks.
@article{restuccia2020deep,
title={{Deep Learning at the Physical Layer: System Challenges and Applications to 5G and Beyond}},
author={Restuccia, Francesco and Melodia, Tommaso},
journal={IEEE Communications Magazine},
volume={58},
number={10},
pages={58--64},
year={2020},
publisher={IEEE}
} 2020
Journal
The Implantable Internet of Medical Things: Toward Lifelong Remote Monitoring and Treatment of Chronic Diseases
R. Guida, N. Dave, F. Restuccia, E. Demirors, and T. Melodia
ACM GetMobile
The promise of real-time detection and response to life-crippling diseases brought by the Implantable Internet of Medical Things (IIoMT) has recently spurred substantial advances in implantable technologies. Yet, existing medical devices do not provide at once the miniaturized end-to-end body monitoring, wireless communication and remote powering capabilities to implement IIoMT applications. This paper fills the existing research gap by presenting U-Verse, the first FDA-compliant rechargeable IIoMT platform packing sensing, computation, communication, and recharging circuits into a penny-scale platform. Extensive experimental evaluation indicates that U-Verse (i) can be wirelessly recharged and can store energy several orders of magnitude more than state-of-theart capacity in tens of minutes; (ii) with one single charge, it can operate from few hours to several days. Finally, U-Verse is demonstrated through (i) a closed-loop application that sends data via ultrasounds through real porcine meat; and (ii) a real-time reconfigurable pacemaker.
@article{guida2021implantable,
title={{The Implantable Internet Of Medical Things: Toward Lifelong Remote Monitoring And Treatment Of Chronic Diseases}},
author={Guida, Raffaele and Dave, Neil and Restuccia, Francesco and Demirors, Emrecan and Melodia, Tommaso},
journal={GetMobile: Mobile Computing and Communications},
volume={24},
number={3},
pages={20--25},
year={2021},
publisher={ACM New York, NY, USA}
} 2020
Conference
Generalized Wireless Adversarial Deep Learning
F. Restuccia, S. D’Oro, A. Al-Shawabka, B. Costa Rendon, K. Chowdhury, S. Ioannidis and T. Melodia
Proceedings of ACM Workshop on Wireless Security and Machine Learning (WiseML)
Deep learning techniques can classify spectrum phenomena (e.g., waveform modulation) with accuracy levels that were once thought impossible. Although we have recently seen many advances in this field, extensive work in computer vision has demonstrated that an adversary can "crack" a classifier by designing inputs that "steer" the classifier away from the ground truth. This paper advances the state of the art by proposing a generalized analysis and evaluation of adversarial machine learning (AML) attacks to deep learning systems in the wireless domain. We postulate a series of adversarial attacks, and formulate a Generalized Wireless Adversarial Machine Learning Problem (GWAP) where we analyze the combined effect of the wireless channel and the adversarial waveform on the efficacy of the attacks. We extensively evaluate the performance of our attacks on a state-of-the-art 1,000-device radio fingerprinting dataset, and a 24-class modulation dataset. Results show that our algorithms can decrease the classifiers' accuracy up to 3x while keeping the waveform distortion to a minimum.
@inproceedings{restuccia2020generalized,
title={{Generalized Wireless Adversarial Deep Learning}},
author={Restuccia, Francesco and D'Oro, Salvatore and Al-Shawabka, Amani and Rendon, Bruno Costa and Chowdhury, Kaushik and Ioannidis, Stratis and Melodia, Tommaso},
booktitle={Proceedings of the ACM Workshop on Wireless Security and Machine Learning (WiseML)},
pages={49--54},
year={2020}
} 2020
Conference
Exposing the Fingerprint: Dissecting the Impact of the Wireless Channel on Radio Fingerprinting
A. Al-Shawabka, F. Restuccia, S. D’Oro, T. Jian, B. Costa Rendon, N. Soltani, J. Dy, K. Chowdhury, S. Ioannidis and T. Melodia
Proceedings of the IEEE Conference on Computer Communications (INFOCOM)
Radio fingerprinting uniquely identifies wireless devices by leveraging tiny hardware-level imperfections inevitably present in off-the-shelf radio circuitry. This way, devices can be directly identified at the physical layer by analyzing the unprocessed received waveform - thus avoiding energy-expensive upper-layer cryptography that resource-challenged embedded devices may not be able to afford. Recent advances have proven that convolutional neural networks (CNNs) - thanks to their multidimensional mappings - can achieve fingerprinting accuracy levels impossible to achieve by traditional low-dimensional algorithms. The same research, however, has also suggested that the wireless channel may negatively impact the accuracy of CNN-based radio fingerprinting algorithms by making device-unique hardware imperfections much harder to recognize.In spite of the growing interest in radio fingerprinting research by academia and DARPA, the wireless research community still lacks (i) a large-scale open dataset for radio fingerprinting collected in diverse environments and rich, diverse, channel conditions; and (ii) a full-fledged, systematic, quantitative investigation of the impact of the wireless channel on the accuracy of CNN-based radio fingerprinting algorithms. The key contribution of this paper is to bridge this gap by (i) collecting and sharing with the community more than 7TB of wireless data obtained from 20 wireless devices with identical RF circuitry (and thus, worst-case scenario for fingerprinting) over the course of several days in (a) an anechoic chamber, (b) in-the-wild testbed, and (c) with cable connections; and (ii) providing a first-of-its-kind evaluation of the impact of the wireless channel on CNN-based fingerprinting algorithms through (a) the 7TB experimental dataset and (b) a 400GB dataset provided by DARPA containing hundreds of thousands of transmissions from thousands of WiFi and ADS-B devices with different SNR conditions. Experimental results conclude that (i) the wireless channel impacts the classification accuracy significantly, i.e., from 85% to 9% and from 30% to 17% in the experimental and DARPA dataset, respectively; and that (ii) equalizing I/Q data can increase the accuracy to a significant extent (i.e., by up to 23%) when the number of devices increases significantly.
@inproceedings{al2020exposing,
title={{Exposing the Fingerprint: Dissecting the Impact of the Wireless Channel on Radio Fingerprinting}},
author={Al-Shawabka, Amani and Restuccia, Francesco and D’Oro, Salvatore and Jian, Tong and Rendon, Bruno Costa and Soltani, Nasim and Dy, Jennifer and Ioannidis, Stratis and Chowdhury, Kaushik and Melodia, Tommaso},
booktitle={Proceedings of the IEEE Conference on Computer Communications (INFOCOM)},
pages={646--655},
year={2020},
organization={IEEE}
}
2020
Conference
DeepWiERL: Bringing Deep Reinforcement Learning to the Internet of Self-Adaptive Things
F. Restuccia and T. Melodia
Proceedings of IEEE Conference on Computer Communications (IEEE INFOCOM)
Recent work has demonstrated that cutting-edge advances in deep reinforcement learning (DRL) may be leveraged to empower wireless devices with the much-needed ability to "sense" current spectrum and network conditions and "react" in real time by either exploiting known optimal actions or exploring new actions. Yet, understanding whether real-time DRL can be at all applied in the resource-challenged embedded IoT domain, as well as designing IoT-tailored DRL systems and architectures, still remains mostly uncharted territory. This paper bridges the existing gap between the extensive theoretical research on wireless DRL and its system-level applications by presenting Deep Wireless Embedded Reinforcement Learning (DeepWiERL), a general-purpose, hybrid software/hardware DRL framework specifically tailored for embedded IoT wireless devices. DeepWiERL provides abstractions, circuits, software structures and drivers to support the training and real-time execution of state-of-the-art DRL algorithms on the device's hardware. Moreover, DeepWiERL includes a novel supervised DRL model selection and bootstrap (S-DMSB) technique that leverages transfer learning and high-level synthesis (HLS) circuit design to orchestrate a neural network architecture that satisfies hardware and application throughput constraints and speeds up the DRL algorithm convergence. Experimental evaluation on a fully-custom software-defined radio testbed (i) proves for the first time the feasibility of real-time DRL-based algorithms on a real-world wireless platform with multiple channel conditions; (ii) shows that DeepWiERL supports 16x data rate and consumes 14x less energy than a software-based implementation, and (iii) indicates that S-DMSB may improve the DRL convergence time by 6x and increase the obtained reward by 45% if prior channel knowledge is available.
@inproceedings{restuccia2020deepwierl,
title={{DeepWiERL: Bringing Deep Reinforcement Learning to the Internet of Self-Adaptive Things}},
author={Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (IEEE INFOCOM)},
pages={844--853},
year={2020},
organization={IEEE}
}
2020
Journal
Toward Operator-to-Waveform 5G Radio Access Network Slicing
S. D’Oro, F. Restuccia, and T. Melodia
IEEE Communications Magazine
RAN slicing refers to a vision where multiple MNOs are assigned virtual networks (i.e., slices) instantiated on top of the same physical network resources. Existing work in this area has addressed RAN slicing at different levels of network abstraction, but has often neglected the multitude of tightly intertwined inter-level operations involved in practical slicing systems. For this reason, this article discusses a novel framework for operator-to-waveform 5G RAN slicing. In the proposed framework, slicing operations are treated holistically, including MNOs' selection of base stations and maximum number of users, down to the waveform-level scheduling of resource blocks. Simulation results show that the proposed framework generates RAN slices where 95 percent of allocated resources can be used to perform coordination- based 5G transmission technologies, and facilitates the coexistence of multiple RAN slices providing up to 120 percent improvement in terms of SINR experienced by mobile users.
@article{d2020toward,
title={{Toward Operator-to-Waveform 5G Radio Access Network Slicing}},
author={D'Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso},
journal={IEEE Communications Magazine},
volume={58},
number={4},
pages={18--23},
year={2020},
publisher={IEEE}
} 2020
Journal
No Radio Left Behind: Radio Fingerprinting Through Deep Learning of Physical-Layer Hardware Impairments
K. Sankhe, M. Belgiovine, F. Zhou, L. Angioloni, F. Restuccia, S. D’Oro, T. Melodia, S. Ioannidis, and K. Chowdhury
IEEE Transactions on Cognitive Communications and Networking
Due to the unprecedented scale of the Internet of Things, designing scalable, accurate, energy-efficient and tamper-proof authentication mechanisms has now become more important than ever. To this end, in this paper we present ORACLE, a novel system based on convolutional neural networks (CNNs) to “fingerprint” (i.e., identify) a unique radio from a large pool of devices by deep-learning the fine-grained hardware impairments imposed by radio circuitry on physical-layer I/Q samples. First, we show how hardware-specific imperfections are learned by the CNN framework. Then, we extensively evaluate the performance of ORACLE on several first-of-its-kind large-scale datasets of WiFi-transmissions collected “in the wild”, as well as a dataset of nominally-identical (i.e., equal baseband signals) WiFi devices, reaching 80-90% accuracy is many cases with the error gap arising due to channel-induced effects. Finally, we show through an experimental testbed, how this accuracy can reach over 99% by intentionally inserting and learning the effect of controlled impairments at the transmitter side, to completely remove the impact of the wireless channel. Furthermore, to scale this approach for classifying potential thousands of radios, we propose an impairment hopping spread spectrum (IHOP) technique that is resilient to spoofing attacks.
@article{sankhe2019no,
title={{No Radio Left Behind: Radio Fingerprinting Through Deep Learning of Physical-layer Hardware Impairments}},
author={Sankhe, Kunal and Belgiovine, Mauro and Zhou, Fan and Angioloni, Luca and Restuccia, Frank and D’Oro, Salvatore and Melodia, Tommaso and Ioannidis, Stratis and Chowdhury, Kaushik},
journal={IEEE Transactions on Cognitive Communications and Networking},
volume={6},
number={1},
pages={165--178},
year={2019},
publisher={IEEE}
} 2020
Conference
Impairment Shift Keying: Covert Signaling by Deep Learning of Controlled Radio Imperfections
K. Sankhe, F. Restuccia, S. D’Oro, T. Jian, Z. Wang, A. Al-Shawabka, J. Dy, T. Melodia, S. Ioannidis, and K. Chowdhury
Proceedings of IEEE Military Communications Conference (MILCOM)
The broadcast nature of the wireless spectrum necessarily implies the possibility of eavesdropping, as well as malicious modification of waveforms through inexpensive, widely available software-defined radios (SDRs). This paper proposes a method for covert wireless communications that can be used to authenticate a device or exchange private information between devices. Our approach, called Impairment Shift Keying (ISK), introduces small yet controlled modifications to the radio transmitter hardware, which distorts regular standards-compliant waveforms, such as WiFi, with only 1% increase in bit error rate. A deep convolutional neural network (CNN) is trained to learn these overlay signal variations, which serves as a low-overhead classifier returning a binary 0 or 1 per detected impairment pattern. By mapping device-specific injected impairment patterns to signal variations, ISK validates device IDs with only few inphase (I) and quadrature (Q) samples. Furthermore, through an experimental testbed, ISK is shown to be resilient to channel and SNR level variations, allowing a throughput of 93-1500 Kbps on the covert channel that is undetected by other receivers.
@inproceedings{sankhe2019impairment,
title={{Impairment Shift Keying: Covert Signaling by Deep Learning of Controlled Radio Imperfections}},
author={Sankhe, Kunal and Restuccia, Francesco and D'Oro, Salvatore and Jian, Tong and Wang, Zifeng and Al-Shawabka, Amani and Dy, Jennifer and Melodia, Tommaso and Ioannidis, Stratis and Chowdhury, Kaushik},
booktitle={Proceedings of the IEEE Military Communications Conference (MILCOM)},
pages={598--603},
year={2019},
organization={IEEE}
} 2019
Conference
U-Verse: A Miniaturized Platform for End-to-End Closed-Loop Implantable Internet of Medical Things Systems
R. Guida, N. Dave, F. Restuccia, E. Demirors, and T. Melodia
Proceedings of ACM Conference on Embedded Networked Sensor Systems (SenSys)
The promise of real-time detection and response to life-crippling diseases brought by the Implantable Internet of Medical Things (IIoMT) has recently spurred substantial advances in implantable technologies. Yet, existing devices do not provide at once the miniaturized end-to-end sensing-computation-communication-recharging capabilities to implement IIoMT applications. This paper fills the existing research gap by presenting U-Verse, the first FDA-compliant rechargeable IIoMT platform packing sensing, computation, communication, and recharging circuits into a penny-scale platform. U-Verse uses a single miniaturized transducer for data exchange and for wireless charging. To predict U-Verse's performance, we (i) derive and experimentally validate a mathematical model of U-Verse's charging efficiency; and (ii) experimentally calculate the resistance-reactance parameters of our ultrasonic transducer and rectifying circuit. We design a matching circuit to maximize the amount of power transferred from the outside. We also go through the challenge of fabricating a full-fledged cm-scale printed circuit board (PCB) for U-Verse. Extensive experimental evaluation indicates that U-Verse (i) is able to recharge a 330mF and 15F energy storage unit - several orders of magnitude higher than existing work - respectively under 20 and 60 minutes at a depth of 5cm; (ii) achieves stored charge duration of up to 610 and 40 hours in case of battery and supercapacitor energy storage, respectively. Finally, U-Verse is demonstrated through (i) a closed-loop application where a periodic sensing/actuation task sends data via ultrasounds through real porcine meat; and (ii) a real-time reconfigurable pacemaker.
@inproceedings{guida2019u,
title={{U-Verse: A Miniaturized Platform for End-to-End Closed-Loop Implantable Internet of Medical Things Systems}},
author={Guida, Raffaele and Dave, Neil and Restuccia, Francesco and Demirors, Emrecan and Melodia, Tommaso},
booktitle={Proceedings of ACM Conference on Embedded Networked Sensor Systems (SenSys)},
pages={311--323},
year={2019}
} 2019
Conference
MillimeTera: Toward A Large-Scale Open-Source mmWave and Terahertz Experimental Testbed
M. Polese, F. Restuccia, A. Gosain, J. M. Jornet, S Bhardwaj, V. Ariyarathna, S. Mandal, K. Zheng, A. Dhananjay, M. Mezzavilla, J. Buckwalter, M. Rodwell, X. Wang, M. Zorzi, A. Madanayake and T. Melodia
Proceedings of ACM Workshop on Millimeter-Wave Networks and Sensing Systems (mmNets)
The promise of widespread 5th generation (5G) and beyond wireless systems can only be fulfilled through extensive experimental campaigns aimed at validating the large body of theoretical findings on millimeter wave (mmWave) and Terahertz (THz) frequencies. However, experimental research efforts in this field are often stymied by the lack of open hardware, open-source software, and affordable testbeds accessible by the research community at large, who is now forced to perform simulation-based research or - if at all possible - small-scale, ad hoc experiments. After discussing existing research challenges in mmWave and THz testbeds, in this paper we propose MillimeTera, a vision for a new gen- eration of disruptive experimental platforms that will radically transform the status quo in mmWave and THz research. We next discuss our preliminary hardware and software efforts, and finally provide a roadmap of our main design and development goals in the years to come.
@inproceedings{polese2019millimetera,
title={{MillimeTera: Toward A Large-Scale Open-Source mmWave and Terahertz Experimental Testbed}},
author={Polese, Michele and Restuccia, Francesco and Gosain, Abhimanyu and Jornet, Josep and Bhardwaj, Shubhendu and Ariyarathna, Viduneth and Mandal, Soumyajit and Zheng, Kai and Dhananjay, Aditya and Mezzavilla, Marco and Buckwalter, Jim and Rodwell, Mark and Wang, Xin and Zorzi, Michele and Madanayake, Arjuna and Melodia, Tommaso},
booktitle={Proceedings of ACM Workshop on Millimeter-Wave Networks and Sensing Systems (mmNets)},
pages={27--32},
year={2019}
}
2019
Conference
Sl-EDGE: Network Slicing at the Edge
S. D’Oro, L. Bonati, F. Restuccia, M. Polese, M. Zorzi and T. Melodia
Proceedings of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)
Network slicing of multi-access edge computing (MEC) resources is expected to be a pivotal technology to the success of 5G networks and beyond. The key challenge that sets MEC slicing apart from traditional resource allocation problems is that edge nodes depend on tightly-intertwined and strictly-constrained networking, computation and storage resources. Therefore, instantiating MEC slices without incurring in resource over-provisioning is hardly addressable with existing slicing algorithms. The main innovation of this paper is Sl-EDGE, a unified MEC slicing framework that allows network operators to instantiate heterogeneous slice services (e.g., video streaming, caching, 5G network access) on edge devices. We first describe the architecture and operations of Sl-EDGE, and then show that the problem of optimally instantiating joint network-MEC slices is NP-hard. Thus, we propose near-optimal algorithms that leverage key similarities among edge nodes and resource virtualization to instantiate heterogeneous slices 7.5x faster and within 25% of the optimum. We first assess the performance of our algorithms through extensive numerical analysis, and show that Sl-EDGE instantiates slices 6x more efficiently then state-of-the-art MEC slicing algorithms. Furthermore, experimental results on a 24-radio testbed with 9 smartphones demonstrate that Sl-EDGE provides simultaneously highly-efficient slicing of joint LTE connectivity, video streaming over WiFi, and ffmpeg video transcoding.
@inproceedings{d2020sl,
title={{Sl-EDGE: Network Slicing at the Edge}},
author={D'Oro, Salvatore and Bonati, Leonardo and Restuccia, Francesco and Polese, Michele and Zorzi, Michele and Melodia, Tommaso},
booktitle={Proceedings of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)},
pages={1--10},
year={2020}
}
2019
Conference
PolymoRF: Polymorphic Wireless Receivers Through Physical Layer Deep Learning
F. Restuccia and T. Melodia
Proceedings of ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)
Today's wireless technologies are largely based on inflexible designs, which makes them inefficient and prone to a variety of wireless attacks. To address this key issue, wireless receivers will need to (i) infer on-the-fly the physical-layer parameters currently used by transmitters; and if needed, (ii) change their hardware and software structures to demodulate the incoming waveform. In this paper, we introduce PolymoRF, a deep learning-based polymorphic receiver able to reconfigure itself in real time based on the inferred waveform parameters. Our key technical innovations are (i) a novel embedded deep learning architecture, called RFNet, which enables the solution of key waveform inference problems; (ii) a generalized hardware/software architecture that integrates RFNet with radio components and signal processing. We prototype PolymoRF on a custom software-defined radio platform, and show through extensive over-the-air experiments that PolymoRF achieves throughput within 87% of a perfect-knowledge Oracle system, thus demonstrating for the first time that polymorphic receivers are feasible.
@inproceedings{restuccia2020polymorf,
title={{PolymoRF: Polymorphic Wireless Receivers Through Physical Layer Deep Learning}},
author={Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of the ACM International Symposium on Theory, Algorithmic Foundations, and Protocol Design for Mobile Networks and Mobile Computing (MobiHoc)},
pages={271--280},
year={2020}
}
2019
Journal
Machine Learning for Wireless Communications in the Internet of Things: A Comprehensive Survey
J. Jagannath, N. Polosky, A. Jagannath, F. Restuccia, and T. Melodia
Ad Hoc Networks
The Internet of Things (IoT) is expected to require more effective and efficient wireless communications than ever before. For this reason, techniques such as spectrum sharing, dynamic spectrum access, extraction of signal intelligence and optimized routing will soon become essential components of the IoT wireless communication paradigm. In this vision, IoT devices must be able to not only learn to autonomously extract spectrum knowledge on-the-fly from the network but also leverage such knowledge to dynamically change appropriate wireless parameters (e.g., frequency band, symbol modulation, coding rate, route selection, etc.) to reach the network’s optimal operating point. Given that the majority of the IoT will be composed of tiny, mobile, and energy-constrained devices, traditional techniques based on a priori network optimization may not be suitable, since (i) an accurate model of the environment may not be readily available in practical scenarios; (ii) the computational requirements of traditional optimization techniques may prove unbearable for IoT devices. To address the above challenges, much research has been devoted to exploring the use of machine learning to address problems in the IoT wireless communications domain. The reason behind machine learning’s popularity is that it provides a general framework to solve very complex problems where a model of the phenomenon being learned is too complex to derive or too dynamic to be summarized in mathematical terms.This work provides a comprehensive survey of the state of the art in the application of machine learning techniques to address key problems in IoT wireless communications with an emphasis on its ad hoc networking aspect. First, we present extensive background notions of machine learning techniques. Then, by adopting a bottom-up approach, we examine existing work on machine learning for the IoT at the physical, data-link and network layer of the protocol stack. Thereafter, we discuss directions taken by the community towards hardware implementation to ensure the feasibility of these techniques. Additionally, before concluding, we also provide a brief discussion of the application of machine learning in IoT beyond wireless communication. Finally, each of these discussions is accompanied by a detailed analysis of the related open problems and challenges.
@article{jagannath2019machine,
title={{Machine Learning for Wireless Communications in the Internet of Things: A Comprehensive Survey}},
author={Jagannath, Jithin and Polosky, Nicholas and Jagannath, Anu and Restuccia, Francesco and Melodia, Tommaso},
journal={Ad Hoc Networks},
volume={93},
pages={101913},
year={2019},
publisher={Elsevier}
} 2019
Conference
Jam Sessions: Analysis and Experimental Evaluation of Advanced Jamming Attacks in MIMO Networks
L. Zhang, F. Restuccia, T. Melodia and S.M. Pudlewski
Proceedings of ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc)
Recent research advances in wireless security have shown that advanced jamming can significantly decrease the performance of wireless communications. In advanced jamming, the adversary intentionally concentrates the available energy budget on specific critical components (e.g., pilot symbols, acknowledgement packets, etc.) to (i) increase the jamming effectiveness, as more targets can be jammed with the same energy budget; and (ii) decrease the likelihood of being detected, as the channel is jammed for a shorter period of time. These key aspects make advanced jamming very stealthy yet exceptionally effective in practical scenarios. One of the fundamental challenges in designing defense mechanisms against an advanced jammer is understanding which jamming strategies yields the lowest throughput, for a given channel condition and a given amount of energy. To the best of our knowledge, this problem still remains unsolved, as an analytic model to quantitatively compare advanced jamming schemes is still missing in existing literature. To fill this gap, in this paper we conduct a comparative analysis of several most viable advanced jamming schemes in the widely-used MIMO networks. We first mathematically model a number of advanced jamming schemes at the signal processing level, so that a quantitative relationship between the jamming energy and the jamming effect is established. Based on the model, theorems are derived on the optimal advanced jamming scheme for an arbitrary channel condition. The theoretical findings are validated through extensive simulations and experiments on a 5-radio 2x2 MIMO testbed. Our results show that the theorems are able to predict jamming efficiency with high accuracy. Moreover, to further demonstrate that the theoretical findings are applicable to address crucial real-world jamming problems, we show that the theorems can be incorporated to state-of-art reinforcement-learning based jamming algorithms and boost the action exploration phase so that a faster convergence is achieved.
@inproceedings{zhang2019jam,
title={{Jam Sessions: Analysis and Experimental Evaluation of Advanced Jamming Attacks in MIMO Networks}},
author={Zhang, Liyang and Restuccia, Francesco and Melodia, Tommaso and Pudlewski, Scott M},
booktitle={Proceedings of ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc)},
pages={61--70},
year={2019}
} 2019
Conference
DeepRadioID: Real-Time Channel-Resistent Optimization of Deep Learning-based Radio Fingerprinting Algorithms
F. Restuccia, S. D’Oro, A. Al-Shawabka, M. Belgiovine, L. Angioloni, S. Ioannidis, K. Chowdhury and T. Melodia
Proceedings of ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc)
Radio fingerprinting provides a reliable and energy-efficient IoT authentication strategy by leveraging the unique hardware-level imperfections imposed on the received wireless signal by the transmitter's radio circuitry. Most of existing approaches utilize hand-tailored protocol-specific feature extraction techniques, which can identify devices operating under a pre-defined wireless protocol only. Conversely, by mapping inputs onto a very large feature space, deep learning algorithms can be trained to fingerprint large populations of devices operating under any wireless standard. One of the most crucial challenges in radio fingerprinting is to counteract the action of the wireless channel, which decreases fingerprinting accuracy significantly by disrupting hardware impairments. On the other hand, due to their sheer size, deep learning algorithms are hardly re-trainable in real-time. Another aspect that is yet to be investigated is whether an adversary can successfully impersonate another device's fingerprint. To address these key issues, this paper proposes DeepRadioID, a system to optimize the accuracy of deep-learning-based radio fingerprinting algorithms without retraining the underlying deep learning model. The key intuition is that through the application of a carefully-optimized digital finite input response filter (FIR) at the transmitter's side, we can apply tiny modifications to the waveform to strengthen its fingerprint according to the current channel conditions. We mathematically formulate the Waveform Optimization Problem (WOP) as the problem of finding, for a given trained neural network, the optimum FIR to be used by the transmitter to improve its fingerprinting accuracy. We extensively evaluate DeepRadioID on a experimental testbed of 20 nominally-identical software-defined radios, as well as on two datasets made up by 500 ADS-B devices and by 500 WiFi devices provided by the DARPA RFMLS program. Experimental results show that DeepRadioID (i) increases fingerprinting accuracy by about 35%, 50% and 58% on the three scenarios considered; (ii) decreases an adversary's accuracy by about 54% when trying to imitate other device's fingerprints by using their filters; (iii) achieves 27% improvement over the state of the art on a 100-device dataset.
@inproceedings{restuccia2019deepradioid,
title={{DeepRadioID: Real-Time Channel-Resistent Optimization of Deep Learning-based Radio Fingerprinting Algorithms}},
author={Restuccia, Francesco and D'Oro, Salvatore and Al-Shawabka, Amani and Belgiovine, Mauro and Angioloni, Luca and Ioannidis, Stratis and Chowdhury, Kaushik and Melodia, Tommaso},
booktitle={Proceedings of ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc)},
pages={51--60},
year={2019}
} 2019
Journal
IncentMe: Effective Mechanism Design to Stimulate Crowdsensing Participants with Uncertain Mobility
F. Restuccia, P. Ferraro, S. Silvestri, S. K. Das, and G. Lo Re
IEEE Transactions on Mobile Computing
Mobile crowdsensing harnesses the sensing power of modern smartphones to collect and analyze data beyond the scale of what was previously possible with traditional sensor networks. Given the participatory nature of mobile crowdsensing, it is imperative to incentivize mobile users to provide sensing services in a timely and reliable manner. Most importantly, given sensed information is often valid for a limited period of time, the capability of smartphone users to execute sensing tasks largely depends on their mobility pattern, which is often uncertain. For this reason, in this paper, we propose IncentMe, a framework that solves this core issue by leveraging game-theoretical reverse auction mechanism design. After demonstrating that the proposed problem is NP-hard, we derive two mechanisms that are parallelizable and achieve higher approximation ratio than existing work. IncentMe has been extensively evaluated on a road traffic monitoring application implemented using mobility traces of taxi cabs in San Francisco, Rome, and Beijing. Results demonstrate that the mechanisms in IncentMe outperform the state of the art work by improving the efficiency in recruiting participants by 30 percent.
@article{restuccia2018incentme,
title={{IncentMe: Effective Mechanism Design to Stimulate Crowdsensing Participants with Uncertain Mobility}},
author={Restuccia, Francesco and Ferraro, Pierluca and Silvestri, Simone and Das, Sajal K and Re, Giuseppe Lo},
journal={IEEE Transactions on Mobile Computing},
volume={18},
number={7},
pages={1571--1584},
year={2018},
publisher={IEEE}
} 2019
Journal
Taming Cross-Layer Attacks in Wireless Networks: A Bayesian Learning Approach
L. Zhang, F. Restuccia, T. Melodia, and S. Pudlewski
IEEE Transactions on Mobile Computing
Wireless networks are extremely vulnerable to a plethora of security threats, including eavesdropping, jamming, and spoofing, to name a few. Recently, a number of next-generation cross-layer attacks have been unveiled, which leverage small changes on one network layer to stealthily and significantly compromise another target layer. Since cross-layer attacks are stealthy, dynamic, and unpredictable in nature, novel security techniques are needed. Since models of the environment and attacker's behavior may be hard to obtain in practical scenarios, machine learning techniques become the ideal choice to tackle cross-layer attacks. In this paper, we propose FORMAT, a novel framework to tackle cross-layer security attacks in wireless networks. FORMAT is based on Bayesian learning and made up by a detection and a mitigation component. On one hand, the attack detection component constructs a model of observed evidence to identify stealthy attack activities. On the other hand, the mitigation component uses optimization theory to achieve the desired trade-off between security and performance. The proposed FORMAT framework has been extensively evaluated and compared with existing work by simulations and experiments obtained with a real-world testbed made up by Ettus Universal Software Radio Peripheral (USRP) radios. Results demonstrate the effectiveness of the proposed methodology as FORMAT is able to effectively detect and mitigate the considered cross-layer attacks.
@article{zhang2018taming,
title={{Taming Cross-Layer Attacks in Wireless Networks: A Bayesian Learning Approach}},
author={Zhang, Liyang and Restuccia, Francesco and Melodia, Tommaso and Pudlewski, Scott M},
journal={IEEE Transactions on Mobile Computing},
volume={18},
number={7},
pages={1688--1702},
year={2018},
publisher={IEEE}
} 2019
Conference
Hiding Data in Plain Sight: Undetectable Wireless Communications Through Pseudo-Noise Asymmetric Shift Keying
S. D’Oro, F. Restuccia, and T. Melodia
Proceedings of IEEE Conference on Computer Communications (INFOCOM)
Undetectable wireless transmissions are fundamental to avoid eavesdroppers or censorship by authoritarian governments. To address this issue, wireless steganography “hides” covert information inside primary information by slightly modifying the transmitted waveform such that primary information will still be decodable, while covert information will be seen as noise by agnostic receivers. Since the addition of covert information inevitably decreases the SNR of the primary transmission, a key challenge in wireless steganography is to mathematically analyze and optimize the impact of the covert channel on the primary channel as a function of different channel conditions. Another core issue is to make sure that the covert channel is almost undetectable by eavesdroppers. Existing approaches are protocol-specific and thus their performance cannot be assessed and optimized in general scenarios. To address this research gap, we notice that existing wireless technologies rely on phase-keying modulations (e.g., BPSK, QPSK) that in most cases do not use the channel up to its Shannon capacity. Therefore, the residual capacity can be leveraged to implement a wireless system based on a pseudo-noise asymmetric shift keying (PN-ASK) modulation, where covert symbols are mapped by shifting the amplitude of primary symbols. This way, covert information will be undetectable, since a receiver expecting phase-modulated symbols will see their shift in amplitude as an effect of channel/path loss degradation. Through rigorous mathematical analysis, we first investigate the SER of PN-ASK as a function of the channel; then, we find the optimal PN-ASK parameters that optimize primary and covert throughput under different channel condition. We evaluate the throughput performance and undetectability of PN-ASK through extensive simulations and on an experimental testbed based on USRP N210 software-defined radios. Results indicate that PN-ASK improves the throughput by more than 8x with respect to prior art. Finally, we demonstrate through experiments that PN-ASK is able to transmit covert data on top of IEEE 802.11g frames, which are correctly decoded by an off-the-shelf laptop WiFi card without any hardware modifications.
@inproceedings{d2019hiding,
title={{Hiding Data in Plain Sight: Undetectable Wireless Communications Through Pseudo-Noise Asymmetric Shift Keying}},
author={D’Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (INFOCOM)},
pages={1585--1593},
year={2019},
organization={IEEE}
}
2019
Conference
The Slice Is Served: Enforcing Radio Access Network Slicing in Virtualized 5G Systems
Proceedings of IEEE International Conference on Computer Communications (INFOCOM)
The notions of softwarization and virtualization of the radio access network (RAN) of next-generation (5G) wireless systems are ushering in a vision where applications and services are physically decoupled from devices and network infrastructure. This crucial aspect will ultimately enable the dynamic deployment of heterogeneous services by different network operators over the same physical infrastructure. RAN slicing is a form of 5G virtualization that allows network infrastructure owners to dynamically “slice” and “serve” their network resources (i. e., spectrum, power, antennas, among others) to different mobile virtual network operators (MVNOs), according to their current needs. Once the slicing policy (i.e., the percentage of resources assigned to each MVNO) has been computed, a major challenge is how to allocate spectrum resources to MVNOs in such a way that (i) the slicing policy defined by the network owner is enforced; and (ii) the interference among different MVNOs is minimized. In this article, we mathematically formalize the RAN slicing enforcement problem (RSEP) and demonstrate its NP-hardness. For this reason, we design three approximation algorithms that render the solution scalable as the RSEP increases in size. We extensively evaluate their performance through simulations and experiments on a testbed made up of 8 software-defined radio peripherals. Experimental results reveal that not only do our algorithms enforce the slicing policies, but can also double the total network throughput when intra-MVNO power control policies are used in conjunction.
@inproceedings{d2019slice,
title={The slice is served: Enforcing radio access network slicing in virtualized 5G systems},
author={D’Oro, Salvatore and Restuccia, Francesco and Talamonti, Alessandro and Melodia, Tommaso},
booktitle={IEEE INFOCOM 2019-IEEE Conference on Computer Communications},
pages={442--450},
year={2019},
organization={IEEE}
} 2019
Conference
Big Data Goes Small: Real-Time Spectrum-Driven Embedded Wireless Networking Through Deep Learning in the RF Loop
F. Restuccia and T. Melodia
Proceedings of IEEE Conference on Computer Communications (INFOCOM)
The explosion of 5G networks and the Internet of Things will result in an exceptionally crowded RF environment, where techniques such as spectrum sharing and dynamic spectrum access will become essential components of the wireless communication process. In this vision, wireless devices must be able to (i) learn to autonomously extract knowledge from the spectrum on-the-fly; and (ii) react in real time to the inferred spectrum knowledge by appropriately changing communication parameters, including frequency band, symbol modulation, coding rate, among others. Traditional CPU-based machine learning suffers from high latency, and requires application-specific and computationally-intensive feature extraction/selection algorithms. Conversely, deep learning allows the analysis of massive amount of unprocessed spectrum data without ad-hoc feature extraction. So far, deep learning has been used for offline wireless spectrum analysis only. Therefore, additional research is needed to design systems that bring deep learning algorithms directly on the device's hardware and tightly intertwined with the RF components to enable real-time spectrum-driven decision-making at the physical layer. In this paper, we present RFLearn, the first system enabling spectrum knowledge extraction from unprocessed I/Q samples by deep learning directly in the RF loop. RFLearn provides (i) a complete hardware/software architecture where the CPU, radio transceiver and learning/actuation circuits are tightly connected for maximum performance; and (ii) a learning circuit design framework where the latency vs. hardware resource consumption trade-off can explored. We implement and evaluate the performance of RFLearn on custom software-defined radio built on a system-on-chip (SoC) ZYNQ-7000 device mounting AD9361 radio transceivers and VERT2450 antennas. We showcase the capabilities of RFLearn by applying it to solving the fundamental problems of modulation and OFDM parameter recognition. Experimental results reveal that RFLearn decreases latency and power by about 17x and 15x with respect to a software-based solution, with a comparatively low hardware resource consumption.
@inproceedings{restuccia2019big,
title={{Big Data Goes Small: Real-Time Spectrum-Driven Embedded Wireless Networking Through Deep Learning in the RF Loop}},
author={Restuccia, Francesco and Melodia, Tommaso},
booktitle={Proceedings of IEEE Conference on Computer Communications (INFOCOM)},
pages={2152--2160},
year={2019},
organization={IEEE}
} 2019
Journal
FIRST: A Framework for Optimizing Information Quality in Mobile Crowdsensing Systems
F. Restuccia, P. Ferraro, S. Silvestri, T. S. Sanders, S. K. Das, and G. Lo Re
ACM Transactions on Sensor Networks
Thanks to the collective action of participating smartphone users, mobile crowdsensing allows data collection at a scale and pace that was once impossible. The biggest challenge to overcome in mobile crowdsensing is that participants may exhibit malicious or unreliable behavior, thus compromising the accuracy of the data collection process. Therefore, it becomes imperative to design algorithms to accurately classify between reliable and unreliable sensing reports. To address this crucial issue, we propose a novel Framework for optimizing Information Reliability in Smartphone-based participaTory sensing (FIRST) that leverages mobile trusted participants (MTPs) to securely assess the reliability of sensing reports. FIRST models and solves the challenging problem of determining before deployment the minimum number of MTPs to be used to achieve desired classification accuracy. After a rigorous mathematical study of its performance, we extensively evaluate FIRST through an implementation in iOS and Android of a room occupancy monitoring system and through simulations with real-world mobility traces. Experimental results demonstrate that FIRST reduces significantly the impact of three security attacks (i.e., corruption, on/off, and collusion) by achieving a classification accuracy of almost 80% in the considered scenarios. Finally, we discuss our ongoing research efforts to test the performance of FIRST as part of the National Map Corps project.
@article{restuccia2018first,
title={{FIRST: A Framework for Optimizing Information Quality in Mobile Crowdsensing Systems}},
author={Restuccia, Francesco and Ferraro, Pierluca and Sanders, Timothy S and Silvestri, Simone and Das, Sajal K and Re, Giuseppe Lo},
journal={ACM Transactions on Sensor Networks (TOSN)},
volume={15},
number={1},
pages={1--35},
year={2018},
publisher={ACM New York, NY, USA}
} 2019
Journal
Securing the Internet of Things in the Age of Machine Learning and Software-Defined Networking
F. Restuccia, S. D’Oro, and T. Melodia
IEEE Internet of Things Journal
The Internet of Things (IoT) realizes a vision where billions of interconnected devices are deployed just about everywhere, from inside our bodies to the most remote areas of the globe. As the IoT will soon pervade every aspect of our lives and will be accessible from anywhere, addressing critical IoT security threats is now more important than ever. Traditional approaches where security is applied as an afterthought and as a “patch” against known attacks are insufficient. Indeed, next-generation IoT challenges will require a new secure-by-design vision, where threats are addressed proactively and IoT devices learn to dynamically adapt to different threats. To this end, machine learning (ML) and software-defined networking (SDN) will be key to provide both reconfigurability and intelligence to the IoT devices. In this paper, we first provide a taxonomy and survey the state of the art in IoT security research, and offer a roadmap of concrete research challenges related to the application of ML and SDN to address existing and next-generation IoT security threats.
@article{restuccia2018securing,
title={{Securing the Internet of Things in the Age of Machine Learning and Software-Defined Networking}},
author={Restuccia, Francesco and D’Oro, Salvatore and Melodia, Tommaso},
journal={IEEE Internet of Things Journal},
volume={5},
number={6},
pages={4829--4842},
year={2018},
publisher={IEEE}
} 2018
Journal
Low-Complexity Distributed Radio Access Network Slicing: Algorithms and Experimental Results
S. D’Oro, F. Restuccia, T. Melodia, and S. Palazzo
IEEE/ACM Transactions on Networking
Radio access network (RAN) slicing is an effective methodology to dynamically allocate networking resources in 5G networks. One of the main challenges of RAN slicing is that it is provably an NP-Hard problem. For this reason, we design near-optimal low-complexity distributed RAN slicing algorithms. First, we model the slicing problem as a congestion game, and demonstrate that such game admits a unique Nash equilibrium (NE). Then, we evaluate the Price of Anarchy (PoA) of the NE, i.e., the efficiency of the NE as compared with the social optimum, and demonstrate that the PoA is upper-bounded by 3/2. Next, we propose two fully-distributed algorithms that provably converge to the unique NE without revealing privacy-sensitive parameters from the slice tenants. Moreover, we introduce an adaptive pricing mechanism of the wireless resources to improve the network owner’s profit. We evaluate the performance of our algorithms through simulations and an experimental testbed deployed on the Amazon EC2 cloud, both based on a real-world dataset of base stations from the OpenCellID project. Results conclude that our algorithms converge to the NE rapidly and achieve near-optimal performance, while our pricing mechanism effectively improves the profit of the network owner.
@article{d2018low,
title={{Low-Complexity Distributed Radio Access Network Slicing: Algorithms and Experimental Results}},
author={D’Oro, Salvatore and Restuccia, Francesco and Melodia, Tommaso and Palazzo, Sergio},
journal={IEEE/ACM Transactions on Networking},
volume={26},
number={6},
pages={2815--2828},
year={2018},
publisher={IEEE}
} 2018
Journal
Quality of Information in Mobile Crowdsensing: Survey and Research Challenges
F. Restuccia, N. Ghosh, S. Bhattacharjee, S. K. Das, and T. Melodia
ACM Transactions on Sensor Networks
Smartphones have become the most pervasive devices in people’s lives and are clearly transforming the way we live and perceive technology. Today’s smartphones benefit from almost ubiquitous Internet connectivity and come equipped with a plethora of inexpensive yet powerful embedded sensors, such as an accelerometer, a gyroscope, a microphone, and a camera. This unique combination has enabled revolutionary applications based on the mobile crowdsensing paradigm, such as real-time road traffic monitoring, air and noise pollution, crime control, and wildlife monitoring, just to name a few. Differently from prior sensing paradigms, humans are now the primary actors of the sensing process, since they become fundamental in retrieving reliable and up-to-date information about the event being monitored. As humans may behave unreliably or maliciously, assessing and guaranteeing Quality of Information (QoI) becomes more important than ever. In this article, we provide a new framework for defining and enforcing the QoI in mobile crowdsensing and analyze in depth the current state of the art on the topic. We also outline novel research challenges, along with possible directions of future work.
@article{restuccia2017quality,
title={{Quality of Information in Mobile Crowdsensing: Survey and Research Challenges}},
author={Restuccia, Francesco and Ghosh, Nirnay and Bhattacharjee, Shameek and Das, Sajal K and Melodia, Tommaso},
journal={ACM Transactions on Sensor Networks (TOSN)},
volume={13},
number={4},
pages={1--43},
year={2017},
publisher={ACM New York, NY, USA}
} 2017
Conference
iSonar: Software-defined Underwater Acoustic Networking for Amphibious Smartphones
F. Restuccia, E. Demirors, and T. Melodia
Proceedings of ACM International Conference on Underwater Networks & Systems (WUWNet)
Recent technological advances have brought to the end-user market smartphones that are able to remain fully-functional even when submerged under water. This capability will soon enable the commercialization of a plethora of exciting smartphone apps, including life-saving systems such as real-time monitoring of scuba divers breathing. On the other hand, it becomes paramount to empower smartphones with end-to-end underwater communication capabilities. In this paper, we propose iSonar, the first system implementing reliable software-defined acoustic networking between water-proof smartphones. Specifically, iSonar transforms off-the-shelf smartphones into ultrasonic software "radios" that implement an orthogonal frequency division multiplexing-based communication system to exchange data under water. To this end, iSonar sends and receives information through the AUX interface and by implementing a lightweight network stack entirely in software. We have implemented a fully-functional hardware/software prototype of iSonar on Android smartphones and off-the-shelf electronic equipment, and extensively evaluated its performance through several experiments in a tank testbed. Results show that iSonar is able to achieve packet error rate (PER) of 10-3, which is significant considering the low audio sampling rate and the strong multipath effect induced by the water tank environment.
@inproceedings{restuccia2017isonar,
title={{iSonar: Software-defined Underwater Acoustic Networking for Amphibious Smartphones}},
author={Restuccia, Francesco and Demirors, Emrecan and Melodia, Tommaso},
booktitle={Proceedings of ACM International Conference on Underwater Networks & Systems (WUWNet)},
pages={1--9},
year={2017}
}
2017
Conference
Learning to Detect and Mitigate Cross-layer Attacks in Wireless Networks: Framework and Applications
L. Zhang, F. Restuccia, T. Melodia, and S. Pudlewski
Proceedings of IEEE Conference on Communications and Network Security (IEEE CNS)
Security threats such as jamming and route manipulation can have significant consequences on the performance of modern wireless networks. To increase the efficacy and stealthiness of such threats, a number of extremely challenging, next-generation cross-layer attacks have been recently unveiled. Although existing research has thoroughly addressed many single-layer attacks, the problem of detecting and mitigating cross-layer attacks still remains unsolved. For this reason, in this paper we propose a novel framework to analyze and address cross-layer attacks in wireless networks. Specifically, our framework consists of a detection and a mitigation component. The attack detection component is based on a Bayesian learning detection scheme that constructs a model of observed evidence to identify stealthy attack activities. The mitigation component comprises a scheme that achieves the desired trade-off between security and performance. We specialize and evaluate the proposed framework by considering a specific cross-layer attack that uses jamming as an auxiliary tool to achieve route manipulation. Simulations and experimental results obtained with a testbed made up by USRP software-defined radios demonstrate the effectiveness of the proposed methodology.
@inproceedings{zhang2017learning,
author={Zhang, Liyang and Restuccia, Francesco and Melodia, Tommaso and Pudlewski, Scott M.},
booktitle={Proceedings of IEEE Conference on Communications and Network Security (IEEE CNS)},
title={{Learning to Detect and Mitigate Cross-layer Attacks in Wireless Networks: Framework and Applications}},
year={2017},
volume={},
number={},
pages={1-9}
} 2017
Journal
Accurate and Efficient Modeling of 802.15.4 Unslotted CSMA/CA through Event Chains Computation
D. De Guglielmo, F. Restuccia, G. Anastasi, M. Conti, and S.K. Das
IEEE Transactions on Mobile Computing
Many analytical models have been proposed for evaluating the performance of event-driven 802.15.4 Wireless Sensor Networks (WSNs), in Non-Beacon Enabled (NBE) mode. However, existing models do not provide accurate analysis of large-scale WSNs, due to tractability issues and/or simplifying assumptions. In this paper, we propose a new approach called Event Chains Computation (ECC) to model the unslotted CSMA/CA algorithm used for channel access in NBE mode. ECC relies on the idea that outcomes of the CSMA/CA algorithm can be represented as chains of events that subsequently occur in the network. Although ECC can generate all the possible outcomes, it only considers chains with a probability to occur greater than a pre-defined threshold to reduce complexity. Furthermore, ECC parallelizes the computation by managing different chains through different threads. Our results show that, by an appropriate threshold selection, the time to derive performance metrics can be drastically reduced, with negligible impact on accuracy. We also show that the computation time decreases almost linearly with the number of employed threads. We validate our model through simulations and testbed experiments, and use it to investigate the impact of different parameters on the WSN performance, in terms of delivery ratio, latency, and energy consumption.
@article{de2016accurate,
title={{Accurate and Efficient Modeling of 802.15. 4 Unslotted CSMA/CA through Event Chains Computation}},
author={De Guglielmo, Domenico and Restuccia, Francesco and Anastasi, Giuseppe and Conti, Marco and Das, Sajal K},
journal={IEEE Transactions on Mobile Computing},
volume={15},
number={12},
pages={2954--2968},
year={2016},
publisher={IEEE}
} 2016
Journal
Optimizing the Lifetime of Sensor Networks with Uncontrollable Mobile Sinks and QoS Constraints
F. Restuccia and S. K. Das
ACM Transactions on Sensor Networks
In past literature, it has been demonstrated that the use of mobile sinks (MSs) increases dramatically the lifetime of wireless sensor networks (WSNs). In applications where the MSs are humans, animals, or transportation systems, the mobility of the MS is often random and unpredictable, implying the necessity of novel and specific algorithms able to deal with large uncertainty on the MS mobility. In this paper, we define the yet unsolved problem of optimizing the lifetime of a WSN in the presence of uncontrollable and random sink mobility with QoS constraints. Then, we present a novel Swarm-Intelligence-based Sensor Selection Algorithm (SISSA), which optimizes network lifetime and meets pre-defined QoS constraints. Next, we mathematically analyze SISSA and derive analytical bounds on energy consumption, number of messages exchanged, and convergence time. The efficiency of SISSA and the accuracy of the model are experimentally evaluated with a testbed composed by 40 sensors, and the network lifetime provided by SISSA is compared to that by an ideal scheme. Experimental and analytical results conclude that SISSA is highly scalable and energy-efficient, and provides on the average the 56% of the lifetime provided by the ideal scheme in all the considered network parameter sets.
@article{restuccia2016optimizing,
title={{Optimizing the Lifetime of Sensor Networks with Uncontrollable Mobile Sinks and QoS Constraints}},
author={Restuccia, Francesco and Das, Sajal K},
journal={ACM Transactions on Sensor Networks (TOSN)},
volume={12},
number={1},
pages={1--31},
year={2016},
publisher={ACM New York, NY, USA}
} 2016
Journal
Incentive Mechanisms for Participatory Sensing: Survey and Research Challenges
F. Restuccia, S.K. Das, and J. Payton
ACM Transactions on Sensor Networks
Participatory sensing is a powerful paradigm that takes advantage of smartphones to collect and analyze data beyond the scale of what was previously possible. Given that participatory sensing systems rely completely on the users’ willingness to submit up-to-date and accurate information, it is paramount to effectively incentivize users’ active and reliable participation. In this article, we survey existing literature on incentive mechanisms for participatory sensing systems. In particular, we present a taxonomy of existing incentive mechanisms for participatory sensing systems, which are subsequently discussed in depth by comparing and contrasting different approaches. Finally, we discuss an agenda of open research challenges in incentivizing users in participatory sensing.
@article{restuccia2016incentive,
title={{Incentive Mechanisms for Participatory Sensing: Survey and Research Challenges}},
author={Restuccia, Francesco and Das, Sajal K and Payton, Jamie},
journal={ACM Transactions on Sensor Networks (TOSN)},
volume={12},
number={2},
pages={1--40},
year={2016},
publisher={ACM New York, NY, USA}
} 2016
Conference
Lifetime Optimization with QoS of Sensor Networks with Uncontrollable Mobile Sinks
Restuccia, Francesco, and Das, Sajal K
In 2015 IEEE 16th International Symposium on A World of Wireless, Mobile and Multimedia Networks (WoWMoM)
In past literature, it has been demonstrated that the use of mobile sinks (MSs) increases dramatically the lifetime of wireless sensor networks (WSNs). In applications where the MSs are humans, animals, or transportation systems, the mobility of the MS is often random and unpredictable, implying the necessity of novel and specific algorithms able to deal with large uncertainty on the MS mobility. In this paper, we define the yet unsolved problem of optimizing the lifetime of a WSN in the presence of uncontrollable and random sink mobility with QoS constraints. Then, we present a novel Swarm-Intelligence-based Sensor Selection Algorithm (SISSA), which optimizes network lifetime and meets pre-defined QoS constraints. Next, we mathematically analyze SISSA and derive analytical bounds on energy consumption, number of messages exchanged, and convergence time. The efficiency of SISSA and the accuracy of the model are experimentally evaluated with a testbed composed by 40 sensors, and the network lifetime provided by SISSA is compared to that by an ideal scheme. Experimental and analytical results conclude that SISSA is highly scalable and energy-efficient, and provides on the average the 56% of the lifetime provided by the ideal scheme in all the considered network parameter sets.
@inproceedings{restuccia2015lifetime,
abbr = {Conference},
title = {Lifetime Optimization with QoS of Sensor Networks with Uncontrollable Mobile Sinks},
author = {Restuccia, Francesco and Das, Sajal K},
booktitle = {2015 IEEE 16th International Symposium on A World of Wireless, Mobile and Multimedia Networks (WoWMoM)},
bibtex_show = {true},
pages = {1--9},
html = {https://ieeexplore.ieee.org/abstract/document/7158130},
year = {2015},
organization = {IEEE}
} 2015
Conference
Preserving QoI in Participatory Sensing by Tackling Location-Spoofing Through Mobile WiFi Hotspots
Restuccia, Francesco, Saracino, Andrea, Das, Sajal K, and Martinelli, Fabio
In 2015 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops)
The Quality of Information (QoI) in Participatory Sensing (PS) systems largely depends on the location accuracy of participating users. However, users could easily provide false information through Location Spoofing Attacks (LSA). Existing PS systems are not able to efficiently validate the position of users in large-scale outdoor environments, thus being prone to reduced QoI. In this paper we present an efficient scheme to secure PS systems from LSAs. In particular, the user location is verified with the help of mobile WiFi hot spots (MHSs), which are users activating WiFi interface on their smart phones and waiting connections from nearby users, and thereby validating their position inside the sensing area. A reputation-based algorithm is proposed to rule out sensing reports of location-spoofing users, thereby increasing the reliability of the PS system. The effectiveness of our scheme is analyzed by real-world experiments and simulation study.
@inproceedings{restuccia2015preserving,
abbr = {Conference},
title = {Preserving QoI in Participatory Sensing by Tackling Location-Spoofing Through Mobile WiFi Hotspots},
author = {Restuccia, Francesco and Saracino, Andrea and Das, Sajal K and Martinelli, Fabio},
bibtex_show = {true},
booktitle = {2015 IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops)},
pages = {81--86},
year = {2015},
html = {https://ieeexplore.ieee.org/abstract/document/7133998},
organization = {IEEE}
} 2015
Conference
Preserving QoI in Participatory Sensing by Tackling Location-spoofing Attacks through Mobile WiFi Hotspots
F. Restuccia, A. Saracino, S. K. Das, F. Martinelli
Proceedings of the International Workshop on Information Quality and Quality of Service for Pervasive Computing (IQ2S)
The Quality of Information (QoI) in Participatory Sensing (PS) systems largely depends on the location accuracy of participating users. However, users could easily provide false information through Location Spoofing Attacks (LSA). Existing PS systems are not able to efficiently validate the position of users in large-scale outdoor environments, thus being prone to reduced QoI. In this paper we present an efficient scheme to secure PS systems from LSAs. In particular, the user location is verified with the help of mobile WiFi hot spots (MHSs), which are users activating WiFi interface on their smart phones and waiting connections from nearby users, and thereby validating their position inside the sensing area. A reputation-based algorithm is proposed to rule out sensing reports of location-spoofing users, thereby increasing the reliability of the PS system. The effectiveness of our scheme is analyzed by real-world experiments and simulation study.
@inproceedings{restuccia2015preserving,
title={{Preserving QoI in Participatory Sensing by Tackling Location-spoofing Attacks through Mobile WiFi Hotspots}},
author={Restuccia, Francesco and Saracino, Andrea and Das, Sajal K and Martinelli, Fabio},
booktitle={IEEE International Conference on Pervasive Computing and Communication Workshops (PerCom Workshops)},
pages={81--86},
year={2015},
organization={IEEE}
} 2015
Journal
Analysis and Optimization of Protocol for Mobile Element Discovery in Sensor Networks
F. Restuccia, G. Anastasi, M. Conti, S. K. Das
IEEE Transactions on Mobile Computing
Recent studies have demonstrated that mobile elements (MEs) are an efficient solution to help decrease dramatically energy consumption in wireless sensor networks (WSNs). However, in most of cases, sensors use duty cycle schemes to save energy, and unless the ME mobility pattern is deterministic, each sensor node has to discover the presence of the ME in the nearby area before starting to exchange data with it. Therefore, in such wireless sensor networks with mobile elements (in short, WSN-MEs), the definition and analysis of a protocol for efficient ME discovery becomes of fundamental importance. In this paper, we propose an extensive performance analysis of an easy-to-implement, hierarchical discovery protocol for WSN-MEs, called Dual Beacon Discovery (2BD) protocol, taking into account stochastic, multi-path, variable speed ME mobility patterns. We also derive the optimal parameter values that minimize the energy consumption of sensor nodes, while guaranteeing the minimum node throughput required by the applications under consideration. Finally, we compare the 2BD protocol with a classical solution based on Periodic Listening (PL). Our results show that 2BD can exploit its hierarchical mechanism and thus significantly increase lifetime, especially when the ME discovery phase is relatively long.
@article{restuccia2013analysis,
title={{Analysis and Optimization of a Protocol for Mobile Element Discovery in Sensor Networks}},
author={Restuccia, Francesco and Anastasi, Giuseppe and Conti, Marco and Das, Sajal K},
journal={IEEE Transactions on Mobile Computing},
volume={13},
number={9},
pages={1942--1954},
year={2013},
publisher={IEEE}
} 2014
Conference
FIDES: A Trust-based Framework for Secure User Incentivization in Participatory Sensing
F. Restuccia and S. K. Das
Proceedings of IEEE International Symposium on a World of Wireless, Mobile, and Multimedia Networks (WoWMoM)
Participatory sensing (PS) has recently attracted tremendous attention given its potential for a wide variety of sensing applications. Due to the fact that PS systems rely completely on the data provided by the users, incentivizing users' active participation while guaranteeing data reliability is paramount to effectively employ PS systems in practical scenarios. In this paper, we first define a set of attacks which compromise data reliability of existing PS applications. Next, we propose a scalable and secure trust-based framework, called FIDES, which relies on the concept of mobile security agents (MSAs) and Josang's trust model to rule out incorrect reports and reward reliable users. By simulating the FIDES framework on mobility traces of taxi cabs in San Francisco, we demonstrate that FIDES secures the PS system from the proposed attacks, guarantees high data reliability, and saves significant amount of revenue with respect to existing reward mechanisms.
@inproceedings{restuccia2014fides,
title={{FIDES: A Trust-based Framework for Secure User Incentivization in Participatory Sensing}},
author={Restuccia, Francesco and Das, Sajal K},
booktitle={Proceedings of IEEE International Symposium on a World of Wireless, Mobile, and Multimedia Networks (WoWMoM)},
pages={1--10},
year={2014},
organization={IEEE}
} 2014
Conference
Performance Analysis of a Hierarchical Discovery Protocol for WSNs with Mobile Elements
F. Restuccia, G. Anastasi, M. Conti, S. K. Das
Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)
Wireless Sensor Networks (WSNs) are emerging as an effective solution for a wide range of real-life applications. In scenarios where a fine-grain sensing is not required, sensor nodes can be sparsely deployed in strategic locations and special Mobile Elements (MEs) can be used for data collection. Since communication between a sensor node and a ME can occur only when they are in the transmission range of each other, one of the main challenges in the design of a WSN with MEs is the energy-efficient and timely discovery of MEs. In this paper, we consider a hierarchical ME discovery protocol, namely Dual beacon Discovery (2BD) protocol, based on two different beacon messages emitted by the ME (i.e., Long-Range Beacons and Short-Range Beacons). We develop a detailed analytical model of 2BD assuming a sparse network scenario, and derive the optimal parameter values that minimize the energy consumption at sensor nodes, while guaranteeing the minimum throughput required by the application. Finally, we compare the energy efficiency and performance of 2BD with those of a traditional discovery protocol based on a single beacon. Our results show that 2BD can provide significant energy savings, especially when the discovery phase is relatively long.
@inproceedings{restuccia2012performance,
title={{Performance Analysis of a Hierarchical Discovery Protocol for WSNs with Mobile Elements}},
author={Restuccia, Francesco and Anastasi, Giuseppe and Conti, Marco and Das, Sajal K},
booktitle={Proceedings of IEEE International Symposium on a World of Wireless, Mobile and Multimedia Networks (WoWMoM)},
pages={1--9},
year={2012},
organization={IEEE}
} 2012
Conference
A Hybrid and Flexible Discovery Algorithm for Wireless Sensor Networks with Mobile Elements
K. Kondepu, F. Restuccia, G. Anastasi, M. Conti
Proceedings of IEEE International Symposium on Computers and Communications (ISCC)
In sparse wireless sensor networks, data collection is carried out through specialized mobile nodes that visit sensor nodes, gather data, and transport them to the sink node. Since visit times are typically unpredictable, one of the main challenges to be faced in this kind of networks is the energy-efficient discovery of mobile collector nodes by sensor nodes. In this paper, we propose an adaptive discovery algorithm that combines a learning-based approach with a hierarchical scheme. Thanks to its hybrid nature, the proposed algorithm is very flexible, as it can adapt to very different mobility patterns of the mobile collector node(s), ranging from deterministic to completely random mobility. We have investigated the performance of the proposed approach, through simulation, and we have compared it with existing adaptive algorithms that only leverage either a learning-based or a hierarchical approach. Our results show that the proposed hybrid algorithm outperforms the considered adaptive approaches in all the analyzed scenarios.
@inproceedings{kondepu2012hybrid,
title={{A Hybrid and Flexible Discovery Algorithm for Wireless Sensor Networks with Mobile Elements}},
author={Kondepu, Koteswararao and Restuccia, Francesco and Anastasi, Giuseppe and Conti, Marco},
booktitle={Proceedings of IEEE International Symposium on Computers and Communications (ISCC)},
pages={295--300},
year={2012},
organization={IEEE}
} 2012
Showing all 179 publications