Trust-Aware Multimodal Multi-Agent Coordination for Secure UAV Swarms in 6G Networks
IEEE Open Journal of the Communications Society · 2026 · 7 · 14232-14248
DOI: 10.1109/ojcoms.2026.3748516 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Open Journal of the Communications Society, 2026. Verfasst von Hafiz Muhammad Attaullah und 3 weiteren Autor:innen. Die Arbeit nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deanship of Scientific Research, Islamic University of Madinah, Saudi Arabia.
Abstract (Originalsprache)
Unmanned Aerial Vehicle (UAV) swarms are expected to play a critical role in future sixth-generation (6G) networks by enabling intelligent sensing, autonomous coordination, and real-time mission execution across smart city, disaster response, industrial Internet-of-Things (IoT), and surveillance applications. However, the distributed and autonomous nature of UAV swarms makes them highly vulnerable to cyber-physical threats, including spoofing attacks, false data injection, communication manipulation, model poisoning, and compromised UAV behavior. Existing security solutions primarily focus on isolated aspects such as authentication, intrusion detection, federated learning, or trust management, while lacking a unified mechanism for multimodal trust assessment and resilient swarm-level coordination. To address these challenges, this paper proposes a Trust-Aware Multimodal Multi-Agent Coordination framework for secure UAV swarm operations in adversarial 6G environments. The proposed framework integrates multimodal trust fusion using vision-emulated (identity and spatial consistency), radio-frequency (RF), and motion-based observations, cross-modal consistency verification, trust-aware federated learning, semantic communication, and adaptive swarm coordination. A dynamic trust evaluation mechanism is developed to identify trusted, suspicious, and malicious UAVs, while a trust-weighted federated aggregation strategy mitigates the impact of poisoned model updates and compromised agents. Furthermore, semantic communication is employed to reduce communication overhead and improve coordination efficiency. The proposed framework is evaluated using the T-ITS cyber-physical UAV dataset and NS-3-based swarm simulations under various attack scenarios. Experimental results demonstrate that TMMAC achieves a trust classification accuracy of 99.27%, an attack detection rate of 97.82%, a communication latency of 69 ms, a mission success rate of 94.18%, and an energy consumption of 232 Wh, consistently outperforming recently proposed trust management, federated learning, blockchain-assisted, and intrusion detection baselines. The results confirm that the proposed framework significantly enhances security, resilience, communication efficiency, and operational reliability for next-generation 6G-enabled UAV swarm networks.
Förderung
- Deanship of Scientific Research, Islamic University of Madinah, Saudi Arabia
Zitieren
APA
Attaullah, H. M., Ehsan, M., Siddiqui, M. S., & Rafi, M. (2026). Trust-Aware Multimodal Multi-Agent Coordination for Secure UAV Swarms in 6G Networks. IEEE Open Journal of the Communications Society, 7, 14232-14248. https://doi.org/10.1109/ojcoms.2026.3748516
Vancouver
Attaullah HM, Ehsan M, Siddiqui MS, Rafi M. Trust-Aware Multimodal Multi-Agent Coordination for Secure UAV Swarms in 6G Networks. IEEE Open Journal of the Communications Society. 2026;7:14232-14248. doi:10.1109/ojcoms.2026.3748516
BibTeX
@article{Attaullah2026,
author = {Attaullah, Hafiz Muhammad and Ehsan, Muhammad and Siddiqui, Muhammad Shoaib and Rafi, Muhammad},
title = {Trust-Aware Multimodal Multi-Agent Coordination for Secure UAV Swarms in 6G Networks},
journal = {IEEE Open Journal of the Communications Society},
year = {2026},
volume = {7},
pages = {14232--14248},
doi = {10.1109/ojcoms.2026.3748516}
}
EUVIMED