A hybrid approach for cluster head determination of unmanned aerial vehicle in flying ad-hoc networks

Abstract

In recent years, flying ad hoc networks (FANETs) have witnessed a notable increase in its applications after the onset of the collaborations of the small unmanned aerial vehicles (UAVs). Because of its inherent characteristics, FANETs are used in diverse application ranging from the military to civil domain. Conversely, there are certain issues pertaining to the communication among the UAVs in view of the high mobility and limited battery resources available in the UAVs, resulting in their short lifetime. The paper is an attempt to address these issues plaguing to the short lifespan of the UAVs. In this paper, we propose a hybrid bio-inspired algorithm HGSOFA for optimizing cluster head (CH) selection in a FANETs. HGSOFA utilizes the hybrid implementation of glowworm swarm optimization (GSO) and firefly algorithm (FA). In this paper, we explain the step-by-step working of the HGSOFA and then performance is evaluated through rigorous simulations. Two separate network areas with varying node density is considered for conducting all the simulations. A robust experimental environment is developed using Taguchi and orthogonal methods. HGSOFA’s performance is tested against the conventional GSO and FA algorithms in respect of cluster building time, energy consumption and first node death. Comparable results have showcased the advantages of the HGSOFA as compared to other algorithms.