锚节点稀疏环境下蒙特--卡罗盒定位算法

摘要

Propose a sparse distributed anchor node Monte-Carlo boxed( SDANMCB ) localization algorithm, which transfer node with high positioning precision to virtual anchor node to assist other nodes for localization;according to area of sampling box and neighbour anchor node amounts to adjust sample numbers needed for positioning;after filtering,adjust weight of samples according to posterior distribution of sample. Simulation results show this algorithm has obvious improvement in localization precision,sampling efficiency,and in low anchor node density,localization effect has great improvement.%提出一种适用于锚节点稀疏环境下的蒙特-卡罗盒定位（ SDANMCB）算法。算法在定位过程中将定位精度高的节点转换为虚拟锚节点来辅助其他待定位节点进行定位；同时根据采样箱的面积和附近锚节点数量调整定位所需要的样本数；滤波后根据样本的后验分布调整样本权重。仿真结果表明：算法在定位精度、采样效率上都有明显提升，并且在锚节点密度较低时定位效果有较大改善。