Fuzzy Logic-Based Adaptive Clustering for Enhanced Stability and Energy Efficiency in FANETs

Flying Ad Hoc Networks (FANETs), formed by small drones operating in an ad hoc configuration, face significant hurdles in designing efficient routing protocols due to their three-dimensional movement, rapid topological shifts, high mobility, resource constraints, and sparse node distribution. This study models the clustering and cluster head (CH) selection challenge in FANETs as a constrained combinatorial optimization problem, recognized as NP-hard, requiring innovative and resilient solutions. We introduce ESOFCluster, a pioneering CH selection strategy that employs a fuzzy logic framework to manage the dynamic characteristics of FANETs. The framework utilizes four normalized parameters—residual energy, safe average distance, relative velocity, and link connectivity duration—to facilitate energy-efficient, stable, and flexible cluster formations. Through simulations in NS-3 with a realistic mobility model, the proposed approach named ESOFCluster is evaluated against LEACH, EMASS, and SOFCluster using FANET-specific performance indicators such as role change frequency, packet delivery success, cluster durability, and energy usage. The findings reveal that ESOFCluster outperforms its counterparts by improving network robustness, enhancing data transmission reliability, extending cluster persistence, and optimizing energy efficiency. These advancements establish ESOFCluster as a versatile and scalable clustering solution for FANETs, offering substantial potential for use in UAV swarms, disaster management, and critical missions where dependable and energy-conscious communication is vital.