TY  - JOUR
AV  - none
JF  - IEEE Open Journal of the Computer Society
KW  - Aircraft detection; Computer crime; Cyber attacks; Data privacy; Distributed computer systems; Intrusion detection; Learning systems; Network intrusion; Network security; Neural networks; Training aircraft; Unmanned aerial vehicles (UAV); Aerial vehicle; Attention neural network; Cyber security; Cybersecurity threat; Federated learning; Intrusion detection system; Intrusion Detection Systems; Neural-networks; Unmanned aerial vehicle; Vehicle network; Antennas
UR  - https://www.scopus.com/inward/record.uri?eid=2-s2.0-105018834532&doi=10.1109%2fOJCS.2025.3616394&partnerID=40&md5=5ec18d53f2ef37478b92a566b7e2b9f0
ID  - scholars20323
PB  - Institute of Electrical and Electronics Engineers Inc.
TI  - FedWGCA: A Federated Learning Based UAV Intrusion Detection with Gradient Clipping and Attention-Based Neural Networks
SN  - 26441268
N1  - Cited by: 0; All Open Access, Gold Open Access
Y1  - 2025///
A1  - Fahim-Ul-Islam, Md
A1  - Chakrabarty, Amitabha
A1  - Hakimi, Halimaton Saadiah
A1  - Maidin, Siti Sarah
N2  - Unmanned Aerial Vehicles (UAVs) are progressively employed in various applications, including surveillance and logistics. However, their rising usage is coupled by increased cybersecurity threats. Conventional intrusion detection systems (IDS) frequently inadequately address specific problems presented by UAV networks, including dynamic operational environments, varied data distributions, and severe resource constraints. Federated Learning (FL) has emerged as a viable alternative, offering a decentralized approach to collaborative training of intrusion detection models while preserving data privacy. However, FL is not without its shortcomings, including poisoning and backdoor attacks, which can impair the accuracy and reliability of the models, thereby exposing UAVs to advanced cyber threats. This research attempts to design resilient FL-based frameworks that address these drawbacks, boosting the security and resilience of UAV networks in the face of emerging cyber hazards. We present our proposed weighted gradient clipping aggregation (FedWGCA) framework to mitigate the impact of malicious updates during the model training process. Experimental studies show that our FedWGCA outperforms state-of-the-art methods, surpassing FedAvg, FedNova, FedOpt, and FedProx by up to 7.23 in accuracy, 9.99 in precision, and 10.14 in recall. For enhancing resilience in our FL architecture, we further present our robust attention neural network, ANET, which outperforms XGBoost by 0.90 and DNN by 7.10, showcasing its superior precision and reduced false positives in local training. © 2020 IEEE.
ER  -