A modified gravitational search algorithm and its application in lifetime maximization of wireless sensor networks

Abstract

Recently, academic communities and industrial sectors have been affected by significant advancements inwireless sensor networks (WSNs). Employing clustering methods is the dominant method to maximize the WSN’slifetime, which is considered to be a major issue. Metaheuristic algorithms have attracted wide attention in the researcharea of clustering. In this paper, first a novel nature-inspired optimization algorithm based on the gravitational searchalgorithm (GSA) is defined. To control the exploitation and exploration capabilities of this algorithm, along withcalculating the masses value, the tournament selection method is employed. Tournament size, the parameter of thismethod, is computed automatically using a function during the computational process of the algorithm. The abilitiesof the algorithm are balanced using this problem-independent parameter. Therefore, the performance of the proposedalgorithm is improved in this paper. Moreover, a modified GSA is applied to an energy-efficient clustering protocol forWSNs to minimize the objective function defining the compact clusters that have cluster heads with high energy. Theproposed search algorithm is evaluated in terms of some standard test functions. The results suggest that this methodhas better performance than other state-of-the-art optimization algorithms. In addition, simulation results indicate thatthe proposed method for the clustering problem in WSNs has better performance on network lifetime and delivery datapackets in BS than other popular clustering methods.