Adaptive Multi-Objective Evolutionary Design for Sustainable and Efficient Wireless Sensor Networks

Recently, the demand of Wireless Sensor Network (WSN) increased rapidly due to the vast applications in environmental monitoring, health care, Smart cities, Industrial automation, agriculture and Transports, etc. However, limited battery backup in sensor nodes which affects the lifetime of the WSN. In this paper, a novel multi objective optimization problem is formulated to maximize the life time by considering the distance distribution of sensor nodes with their Cluster Heads (CHs) and distribution of CH's life time. The recent multi objective algorithms like Decomposition based Multi Objective Evolutionary (MOEAD), Multi Objective Grasshopper Algorithm (MOGOA), Multi Objective Particle Swarm Optimization (MOPSO), Multi Objective Grey Wolf optimization (MGWO) algorithm and Improved Multi Objective Differential Evolution (IMODE) algorithms are considered for the designing the multi-objective WSN and their performances are compared based on the scalability, lifetime, energy distribution and cluster distribution of sensor nodes. The simulation results demonstrated that MOEAD based WSN achieves 24.2% more life time than second best algorithms. Further, it achieves 97% (in average) accuracy in cluster quality. Moreover, MOEAD outperformed than other algorithms in Pareto optimal solution in terms of convergence, diversity and solution quality.