arXiv subjects

Shafkat Islam

Publications and source records attributed to Shafkat Islam.

5 recordsLinked to original sources

arXiv2026

UAV-CAS: A Calibrated Digital-Twin Dataset for Intrusion Detection in UAV Swarm Networks

Intrusion detection systems (IDS) trained on wired-network benchmarks degrade sharply in real-world unmanned aerial vehicle (UAV) swarms, where mobility, fluctuating link quality, and decentralized routing reshape traffic distributions. Existing UAV-specific datasets also do not systematically vary these conditions, leaving no way to train or test an IDS against the very shift that defeats it. We present UAV-CAS, a large-scale labeled flow dataset for UAV-network intrusion detection, generated by a Containernet digital twin that is systematically calibrated against AERPAW testbed measurements. We have a four-layer calibration pipeline spanning altitude-dependent path loss, mission-specific mobility, the link-level performance chain, and end-to-end trace fidelity. UAV-CAS comprises 99,492 flows drawn from 1,024 configurations that span five attack families (DoS, DDoS, blackhole, wormhole, replay) and nine collaborative attack compositions. A diversity analysis shows that high-rate attacks separate from benign traffic up to an order of magnitude more strongly than in any prior benchmark, while stealth attacks deliberately blend with benign traffic. Across ten baseline IDS, binary attack detection saturates above $0.98$, confirming the dataset is learnable, whereas full attack-class identification remains hard -- per-class $F_1$ ranges from near zero to $0.82$ and falls into the single digits for stealth attacks. We release the dataset, simulator, and calibration data to support reproducible UAV intrusion-detection research.

cs.NI↗

arXiv2026

Pre-Deployment Complexity Estimation for Federated Perception Systems

Edge AI systems increasingly rely on federated learning to train perception models in distributed, privacy-preserving, and resource-constrained environments. Before training, however, practitioners often lack practical tools for estimating task difficulty in terms of expected accuracy and communication effort. We present a classifier-agnostic, pre-deployment framework that combines intrinsic data properties such as dimensionality, sparsity, and heterogeneity, with client-distribution composition to estimate learning complexity in federated perception systems. Using federated learning as a representative distributed training setting, we examine how learning difficulty varies across different federated configurations. Experiments on three MNIST variants show strong negative correlations between the combined complexity metric and maximum and average federated accuracy, while the intrinsic and distributed components exhibit consistent relationships with communication effort. These findings suggest that complexity estimation can serve as a practical diagnostic tool for resource planning, dataset assessment, and feasibility evaluation in edge-deployed perception systems.

cs.LG↗

arXiv2022

A Resource Allocation Scheme for Energy Demand Management in 6G-enabled Smart Grid

Smart grid (SG) systems enhance grid resilience and efficient operation, leveraging the bidirectional flow of energy and information between generation facilities and prosumers. For energy demand management (EDM), the SG network requires computing a large amount of data generated by massive Internet-of-things sensors and advanced metering infrastructure (AMI) with minimal latency. This paper proposes a deep reinforcement learning (DRL)-based resource allocation scheme in a 6G-enabled SG edge network to offload resource-consuming EDM computation to edge servers. Automatic resource provisioning is achieved by harnessing the computational capabilities of smart meters in the dynamic edge network. To enforce DRL-assisted policies in dense 6G networks, the state information from multiple edge servers is required. However, adversaries can "poison" such information through false state injection (FSI) attacks, exhausting SG edge computing resources. Toward addressing this issue, we investigate the impact of such FSI attacks with respect to abusive utilization of edge resources, and develop a lightweight FSI detection mechanism based on supervised classifiers. Simulation results demonstrate the efficacy of DRL in dynamic resource allocation, the impact of the FSI attacks, and the effectiveness of the detection technique.

cs.NI↗

arXiv2022

Adversarial Analysis of the Differentially-Private Federated Learning in Cyber-Physical Critical Infrastructures

Federated Learning (FL) has become increasingly popular to perform data-driven analysis in cyber-physical critical infrastructures. Since the FL process may involve the client's confidential information, Differential Privacy (DP) has been proposed lately to secure it from adversarial inference. However, we find that while DP greatly alleviates the privacy concerns, the additional DP-noise opens a new threat for model poisoning in FL. Nonetheless, very little effort has been made in the literature to investigate this adversarial exploitation of the DP-noise. To overcome this gap, in this paper, we present a novel adaptive model poisoning technique {\alpha}-MPELM} through which an attacker can exploit the additional DP-noise to evade the state-of-the-art anomaly detection techniques and prevent optimal convergence of the FL model. We evaluate our proposed attack on the state-of-the-art anomaly detection approaches in terms of detection accuracy and validation loss. The main significance of our proposed {\alpha}-MPELM attack is that it reduces the state-of-the-art anomaly detection accuracy by 6.8% for norm detection, 12.6% for accuracy detection, and 13.8% for mix detection. Furthermore, we propose a Reinforcement Learning-based DP level selection process to defend {\alpha}-MPELM attack. The experimental results confirm that our defense mechanism converges to an optimal privacy policy without human maneuver.

cs.CR↗

arXiv2021

DeSMP: Differential Privacy-exploited Stealthy Model Poisoning Attacks in Federated Learning

Federated learning (FL) has become an emerging machine learning technique lately due to its efficacy in safeguarding the client's confidential information. Nevertheless, despite the inherent and additional privacy-preserving mechanisms (e.g., differential privacy, secure multi-party computation, etc.), the FL models are still vulnerable to various privacy-violating and security-compromising attacks (e.g., data or model poisoning) due to their numerous attack vectors which in turn, make the models either ineffective or sub-optimal. Existing adversarial models focusing on untargeted model poisoning attacks are not enough stealthy and persistent at the same time because of their conflicting nature (large scale attacks are easier to detect and vice versa) and thus, remain an unsolved research problem in this adversarial learning paradigm. Considering this, in this paper, we analyze this adversarial learning process in an FL setting and show that a stealthy and persistent model poisoning attack can be conducted exploiting the differential noise. More specifically, we develop an unprecedented DP-exploited stealthy model poisoning (DeSMP) attack for FL models. Our empirical analysis on both the classification and regression tasks using two popular datasets reflects the effectiveness of the proposed DeSMP attack. Moreover, we develop a novel reinforcement learning (RL)-based defense strategy against such model poisoning attacks which can intelligently and dynamically select the privacy level of the FL models to minimize the DeSMP attack surface and facilitate the attack detection.

cs.CR↗