Node localization method for massive sensor networks based on clustering particle swarm optimization in cloud computing environment
Li Pan · International Journal of Wavelets Multiresolution and Information Processing · 2019
In order to reduce the positioning error of wireless sensor network nodes and deal with the positioning of a large number of sensor network nodes, a mass sensor node positioning method based on particle swarm optimization (SNPSO) is proposed. The node location error of node distance correction value is corrected by SNPSO algorithm, and the sensing data is encoded by the index of cloud computing resources. The dynamic target strategy (DTS) algorithm is used to solve the strict deadline constraints. The algorithm focuses on optimizing execution time to meet deadline constraints, and once feasible solutions are obtained, it focuses on optimizing execution costs within deadline constraints. The performance of the algorithm is simulated and analyzed on the platform of MATLAB 2016. Compared with the sensor positioning method, SNPSO improves the positioning accuracy of mass sensor nodes. The simulation results verify the validity of SNPSO. Compared with the improved quantum genetic algorithm (IQGA) under different scale data scheduling and different deadlines constraints, the proposed algorithm can improve the positioning accuracy of mass sensor nodes. The proposed algorithm can find the optimal solution of cloud computing resource scheduling with lower execution cost under strict deadline constraints, and can more easily meet the needs of massive sensor network node location data processing.