EUVIMEDEUVIMED entdecken
Crossref

Edge-QSFL: Quantum-Enhanced Split Federated Learning With Multi-Head Temporal Networks for Edge-Based Intelligent Transportation Security

Yue Zhao, Farhan Ullah, Nazeeruddin Mohammad, Umar Raza, Jawad Ahmad

IEEE Transactions on Consumer Electronics · 2026-05 · 72 (2) · 3929-3939

DOI: 10.1109/tce.2026.3665233 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Consumer Electronics, 2026-05. Verfasst von Yue Zhao und 4 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 32 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The fast development of Intelligent Transportation Systems (ITS) and the Internet of Vehicles (IoV) has shifted the consumer mobility experience. However, the rapid expansion of such technologies presents serious cybersecurity risks, demanding the implementation of a privacy-preserving, adaptive, and powerful Intrusion Detection System (IDS). Currently, Federated Learning (FL) paradigms in ITS are vulnerable to data poisoning and model-based challenges in diverse applications due to a lack of particular architectures capable of processing temporal data streams. This paper introduces Edge-QSFL, a novel framework that combines quantum-enhanced Split Federated Learning (SFL) and multi-head temporal networks. Our approach extracts features from IoV streaming data using lightweight multi-head Temporal Convolutional Networks (TCN) on edge clients. Subsequently, a quantum-based adaptive aggregation mechanism is implemented by a global server. This mechanism employs dynamic weighting to guarantee that the learning process is resilient to data heterogeneity and poisoning. The architecture also incorporates quantum-enhanced security packages to protect the integrity of models. Extensive analyses of three sample datasets, CICIoV2024, N-BaIoT, and UAVAttackData, indicate almost perfect accuracy scores of 99.42%, 94.82%, and 98.0%, respectively, and show to be more resilient to cyber threats. The results reveal that Edge-QSFL presents a scalable, secure, and high-performance solution, making it suitable for the next generation of ITS networks.

Zitieren

APA

Zhao, Y., Ullah, F., Mohammad, N., Raza, U., & Ahmad, J. (2026). Edge-QSFL: Quantum-Enhanced Split Federated Learning With Multi-Head Temporal Networks for Edge-Based Intelligent Transportation Security. IEEE Transactions on Consumer Electronics, 72(2), 3929-3939. https://doi.org/10.1109/tce.2026.3665233

Vancouver

Zhao Y, Ullah F, Mohammad N, Raza U, Ahmad J. Edge-QSFL: Quantum-Enhanced Split Federated Learning With Multi-Head Temporal Networks for Edge-Based Intelligent Transportation Security. IEEE Transactions on Consumer Electronics. 2026;72(2):3929-3939. doi:10.1109/tce.2026.3665233

BibTeX
@article{Zhao2026,
  author = {Zhao, Yue and Ullah, Farhan and Mohammad, Nazeeruddin and Raza, Umar and Ahmad, Jawad},
  title = {Edge-QSFL: Quantum-Enhanced Split Federated Learning With Multi-Head Temporal Networks for Edge-Based Intelligent Transportation Security},
  journal = {IEEE Transactions on Consumer Electronics},
  year = {2026},
  volume = {72},
  number = {2},
  pages = {3929--3939},
  doi = {10.1109/tce.2026.3665233}
}
Kostenlos starten →