Researches Into Optinmal Coverage Problem In Wireless Sensor Networks Based On Binary Particle Swarm Optimization

Wireless sensor networks (WSN) is one of the most significant technologies in the21stcentury, and has drawn lots of attentions all over the world. The network nodes optimalcoverage problem (OCP) involves the sensor quality and the network lifetime, and thereforehas become one of the hottest researches topics in the WSN area. However, most of thecurrent algorithms for the OCP encounter the disadvantages of being too complexity forimplementation and being easy to be trapped into local optima. The OCP in the WSN is stillbeing actively studied. Therefore, research into developing a simpler and efficient algorithmto optimize the problem has become a significant and challenging topic.The intelligent computation algorithms (ICA) have the advantages of natural paralleled,strong robustness, and global optimization, and have obvious advantages in dealing with thecomplex problems. Therefore, the ICA will be very suitable to solve the OCP in the WSN.This paper makes the investigations on the ICA in solving the OCP in the WSN, and proposescoverage mechanism based on a discrete binary particle swarm optimization (BPSO)algorithm to optimize the problem. The paper gives a review on the WSN and its OCP.Moreover, the implementation details for the BPSO solving the OCP are described, includingthe particle code representation, fitness function construction, and the complete algorithmflowchart. Simulations have been conducted and the results are compared with the onesobtained by genetic algorithm (GA). The experimental results and comparisons demonstratethe effectiveness and efficiency of our proposed algorithm.The paper is highlighted with the contributions to the theoretical researches and practicalapplications of the PSO and the WSN as follows. Firstly, the PSO is successfully applied toprovide a new solution to the OCP in the WSN. Secondly, the proposed algorithm is simperthan the traditional algorithms, and with a higher computational effectiveness. Thirdly, theproposed algorithm has a better performance that can obtain higher coverage rate with lesssensor nodes. Fourthly, this paper has implemented a simulation system with using the VisualC++6.0Development Tool, making it convenient for further investigations and researches.