CFFS: An Improved Cluster-Driven Firefly Strategy for FANET Security

Flying Ad hoc Networks are becoming highly appealing due to their versatile applications in domains like surveillance, search and rescue, environmental monitoring, and communication relays, offering significant societal benefits. However, the use of these networks in dynamic environments leaves them open to a range of security risks, including denial-of-service, jamming, and routing attacks, in addition to manipulation and information dropping. To address these challenges, we introduce CFFS, a security approach employs the capabilities of the nature-inspired firefly algorithm. Our framework detects anomalous behavior by dynamically computing weighted parameters for firefly nodes using the Gradient Descent Method. Furthermore, we enhance the efficiency of CFFS by employing a cluster-oriented network system. Additionally, we employ a supervised support vector machine learning method to achieve our objective of threat identification. Through comprehensive experiments, we observe a sufficient enhancement in metrics, including 23.2% reduction in total delay, along with average improvements of 31.6% and 29.3% in threat identification and throughput, respectively over state-of-the-art approaches.