Hybrid Fuzzy C-Means and Ant Colony Optimized (ACO) Clustering Approach for Efficient Routing in Flying Ad-Hoc Networks (FANETs)

With the rapid growth of Flying Ad-Hoc Networks (FANETs), the need for efficient routing algorithms is critical to ensure optimal and reliable communication among Unmanned Aerial Vehicles (UAVs). To address this challenge, this paper presents a new approach named FGA-FANET (Fuzzy-Genetic- Ant FANET), a clustering-based routing algorithm for Flying Ad- Hoc Networks. The proposed algorithm leverages the strengths of the Fuzzy C-Means (FCM) and Ant Colony Optimization (ACO) algorithms to improve routing performance. FGA-FANET operates in two main stages. First, the FCM algorithm is applied, offering a soft clustering method that allows UAVs to belong to multiple clusters with varying degrees of membership, enhancing the network's adaptability to dynamic conditions. In the second stage, the ACO algorithm is introduced to optimize node assignment within clusters. ACO excels at dynamic pathfinding, using pheromone-based routing to improve cluster allocation and boost the network's ability to adapt to real-time changes in topology. Furthermore, integrating Genetic Algorithms (GA) for optimizing the parameters of FCM and ACO could enhance overall performance by fine-tuning key metrics like energy efficiency, network stability, and routing overhead. Extensive simulations were carried out to assess the performance of FGA- FANET, comparing it to three existing methods: Q-FANET, CACONET, and GWOCNET. The results show that FGA-FANET offers notable improvements across all evaluated metrics. It surpasses other approaches in areas such as network lifetime, routing overhead, cluster head lifetime, number of clusters, and clustering efficiency.