Research On Coverage Control Of Wireless Sensor Network Based On Improved Particle Swarm Algorithm

Wireless Sensor Networks(WSN)are composed of many sensor nodes with wireless communication capabilities.The sensor node is responsible for collecting and processing sensing signals from the target area,and transmitting these signals to the data center.With the rapid development of communication technology and embedded technology,sensor nodes are developing towards multi-function,miniaturization and intelligence.At present,WSN has been widely used in environmental protection,industrial and agricultural production and daily life.Due to the uncertainty of the deployment environment,the deployment of sensors is mostly random,but random deployment will have uneven node distribution and a large number of redundant nodes,resulting in poor coverage of the target area.This article mainly studies how to ensure network service quality and maximize network coverage while minimizing energy consumption to extend the life cycle effect of the network.The main research of this article includes:(1)In order to maximize the coverage ratio of wireless sensor networks(WSNs)while minimizing the energy consumption,and extend the network life,in this paper,a particle swarm optimization(PSO)based on co-evolution of resampling technology and beetle antennae search(BAS)is proposed to optimize the coverage control problem of WSNs.The resampling technology balances the global search capability and convergence speed of PSO,increases the overall diversity of the particle swarm,prevents early convergence,and strengthens the particles ability of jumping out of low quality bottom;while the BAS can search its neighborhood by relying on two tentacles of an individual,which enhances the search ability of each single particle.Based on the weighting objective function of coverage ratio and sensor sleep rate,the proposed algorithm not only strengthens the search ability of individual particles,but also ensures the diversity and activity of particle swarm to improve WSN network coverage performance.Experimental results show that RBASPSO's ability to handle complex multi-peak benchmark functions is improved by 1-8 orders of magnitude compared with the three peer algorithms.In WSN coverage optimization,the improvement of coverage of RBASPSO algorithm is better than other peer BASPSO,RPSO and PSO algorithms.With the increase of the number of nodes in the network,the redundancy rate is also constantly improved.When the number of nodes is120,the redundancy rate is enhanced by 4.42% compared to parallel algorithms.(2)Aiming at the problem of insufficient convergence of sensor coverage model and particle swarm algorithm,this paper studies the node cooperative sensing model and reinforcement learning mechanism,and designs a resampled particle swarm algorithm based on reinforcement learning enabled.The experimental results prove that the use of node cooperative sensing model in coverage control is significantly better than the disk model,and the network coverage rate is increased by more than4.83% compared with BASPSO,RPSO and PSO algorithms.And the resampling particle swarm algorithm with reinforcement learning mechanism is better than similar algorithms in convergence and algorithm time complexity.