A Novel Hybrid Particle Swarm Optimization-Simulated Annealing Algorithm for Maximizing Coverage in IoT-Enabled Wireless Sensor Networks

Abstract

<title>Abstract</title> This paper presents a novel hybrid Particle Swarm Optimization-Simulated Annealing (PSO-SA)algorithm for optimizing coverage in IoT-enabled Wireless Sensor Networks (WSNs). We addressthe critical challenge of maximizing network coverage while considering computationalcomplexity. Our approach combines the global search capabilities of PSO with the local optimizationstrengths of SA to achieve superior performance across various network scales. Weconducted extensive simulations for networks ranging from 10 to 50 nodes. The proposed algorithmsignificantly improved network coverage, with improvements of up to 27.8% for smallernetworks (N=10) and 9.16% for larger networks (N=50). The optimized configurations consistentlyachieved coverage ratios above 95% for networks up to 20 nodes, and maintained coverageabove 75% for larger networks of 30-50 nodes. The algorithm demonstrated rapid convergence,typically achieving near-optimal solutions within 200-300 iterations. However, we observed anexponential increase in execution time as network size grew, from approximately 200ms for 10nodes to over 6000ms for 50 nodes. This highlights the trade-off between optimization qualityand computational cost for larger networks. Our results also show that the hybrid PSO-SAapproach maintains higher solution diversity for larger networks, allowing for more thoroughexploration of the solution space. The algorithm achieved a balance between coverage improvementand computational efficiency, with the number of function evaluations remaining relativelyconstant (around 1.8 ∗ 10^5) across all network sizes.