距离误差加权与多通信半径的蝙蝠优化定位算法

传统DV-Hop定位算法存在明显的定位误差,改进的粒子群优化算法由于易陷入局部最优、局部收敛过慢等问题无法满足节点的定位精度要求.针对于此,通过设置跳数阈值优选锚节点以排除异常锚节点对定位精度的干扰;引入多通信半径广播方法修正最小跳数;采用距离误差和跳数归一化思想修正平均跳距;通过利用立方映射均匀化初始蝙蝠种群,引入Levy飞行特征加强算法跳出局部最优能力,使用Powell局部搜索加快算法收敛等三方面改进蝙蝠算法,并利用改进的蝙蝠算法定位未知节点.仿真结果表明,相比传统DV-Hop、BIDV-Hop、GAPSODV-Hop等3种算法,本文改进的定位算法有效降低了定位误差,提高了定位精度.

Bat-Based Optimal Location Algorithm With Distance-weighted Error And Multiple Communication Radius

Traditional DV-Hop localization algorithm has clear positioning errors, andthe improved particle swarm optimization algorithm cannot satisfy the requirement of node location accuracy because it is easy to trap into local optimal solution and too slow local convergence. In view of this,this thesis attempts to eliminate the interference of abnormal anchor nodes to location accuracy through setting the threshold value of hop number;This thesis modifies minimum hop number by means of multi-communication radius broadcasting method;Meanwhile,the average jump distance is corrected by using distance error and hop number normalization. Bat algorithm is improved through using cubic mapping homogenization initial bat population,introducing into hop local optimum ability of Levy flying char-acteristics enhanced algorithm,and adopting Powell local search enhanced algorithmconvergence,and then the im-proved algorithm is used to locate unknown nodes. The simulation results show that,compared with the traditional DV-Hop,BIDV-Hop,GAPSODV-Hop and other algorithms,the improved localization algorithm in this thesis can ef-fectively reduce the positioning errors and improve the positioning accuracy.