三维空间中基于改进的粒子群算法的节点定位研究

Research on Node Localization in Three-Dimensional Space Based on Improved Particle Swarm Algorithm

三维空间下的节点定位一直都是无线传感中的主要研究方向,粒子群算法作为常用的计算节点定位的方法存在一定的误差,本文首先在该算法的基础上引入了凸函数的惯性权值和信任度系数,前者使得粒子能够在局部范围内获得最优解,避免陷入局部最优,后者使得不同性质的粒子能够在全局范围中获得最优解,其次,分析目标定位误差函数,使用泰勒级数对其进行优化,使得定位精度得到提高。仿真实验说明本文算法在锚节点数量,节点通信半径和总节点个数等方面提高了节点定位精度,具有很好的应用前景。

Node localization in3D space has always been the main research direction in wireless sensor,and as a common method to calculate node localization,particle swarm algorithm has certain errors.Based on this algorithm,this paper first introduces the inertia weight and confidence coefficient of the convex function.The former makes the particle obtain the optimal solution within local range and avoids it from falling into local optimum.The latter makes particles of different properties obtain the optimal solution within the global range.Secondly,this paper analyzes the target localization error function and optimizes it with the Taylor series so as to improve the localization accuracy.Simulation experiment shows that algorithm in this paper has improved the node localization accuracy in terms of the amount ofanchor nodes,node communication radius and total amount of nodes,so it has good application prospect.