Flying Ad Hoc Networks facilitate UAV-to-UAV communication for exchanging data to support many real-time applications such as surveillance, intelligent transportation, disaster response, and other critical services. Clustering is commonly utilized to support efficient communication for information collection and processing as the network expands. However, malicious UAVs may deliberately gain access to clusters and disrupt network operations by dropping packets, spreading false information, or by forwarding attacks, thereby severely degrading network reliability. To address these challenges, this paper proposes a trust-aware and secure cluster head selection mechanism, named TTSCH, designed to enhance overall network security by early identifying and removing malicious node from the network. The proposed scheme evaluates UAV behavior using multiple trust metrics, including packet forwarding ratio, communication reliability, energy consistency, and mobility stability. These trust values are integrated with traditional clustering parameters to securely elect reliable cluster heads while excluding malicious UAVs from the election process. Furthermore, a dynamic malicious node isolation mechanism is incorporated to handle on–off and insider attacks. When compared to state-of-the-art clustering systems in the presence of variable malicious UAVs, simulation results show that the proposed TTSCH significantly improves packet delivery ratio, throughput, and network stability while lowering delay.
