无线传感网络覆盖算法及仿真研究

Research on of Wireless Sensor Networks Coverage Simulation

研究无线传感器覆盖算法,针对标准粒子群算法的网络覆盖存在收敛速度慢、易于陷入局部最优值的问题,为满足动态节点选择实时性的要求,提出一种多粒子群的无线传感网络覆盖算法。以无线传感器最大覆盖率为目标函数,通过多个粒子群彼此独立地搜索解空间,加大粒子的搜索范围,减小陷入局部最优的可能性。采用进化粒子,使粒子覆盖更有效率,提高了算法的寻优能力,有效地避免了标准粒子群算法容易出现的"早熟"问题,提高了算法的稳定性。仿真实验表明,与标准粒子群算法、传统遗传算法和新量子遗传算法的优化效果相比较,其覆盖率分别提高了8.39%、3.07%和0.75%;收敛速度提高了25.3%、23.8%和23.8%,证明粒子进化的多粒子群方法有效地优化无线传感网络,实现节点选择的实时性要求。

To maximize the network coverage and extend the life of the network,a Wireless Sensor Networks （WSNs） coverage optimal strategy is proposed based on the evolution of Multi-particle Particle Swarm Optimization （MPSO）. By using the method of Multi-groups parallel searching,the particles,which fall into the best part area according to the theory of evolution,can be chosen rapidly. The strategy also avoids a phenomenon of premature which often occurs when using the method of elementary Particle Swarm Optimization （PSO）,and improves the stability of the algorithm. In the paper,the influence about perceived radius of the nodes on the coverage performance is analyzed through the simulation experiment. The coverage rate and convergence rate increase as the radius of perception speeds up gradually. Experimental results indicate that the MPSO strategy is better than PSO,the Conventional Genetic Algorithms （CGA）,and the New Quantum Genetic Algorithm （NQGA） in coverage optimization.