摘要
无线传感器网络中现有的大多数定位算法都是针对静态节点的,不能应用于节点移动的情况.本文提出了MCBE(Monte Carlo Localization Boxed Using Estimation)移动定位算法,该算法利用锚盒子(Anchor box),即包含待定位节点可能位置且平行于坐标轴的矩形以及非锚节点位置来帮助定位.锚盒子大小和定位误差存在正相关关系,算法根据待定位节点的一跳和二跳锚节点位置计算出锚盒子大小,当锚盒子大于一定值时,利用锚盒子小的非锚节点位置缩小锚盒子,最后在锚盒子范围内对位置采样,用蒙特卡罗MCL方法定位.理论分析和仿真结果表明,MCBE能有效地缩小采样区域,提高定位精度.
Most of localization methods in wireless sensor networks are aimed at static nodes and cannot be applied to mobile nodes. This paper proposed an algorithms called MCBE(Monte Carlo Localization Boxed Using Estimation) which utilizes an anchor box to help localization including the node' s possible positions and being parallel to axes. The size of anchor box is close related to the localization error i.e. the larger the size of anchor box is, the bigger the localization error is. The position estimations of nodes which have small anchor boxes are used to reduce the neighbors' anchor boxes that are larger than some certain value. The position samples are then obtained from the anchor box randomly. The analysis and simulations show that MCBE can limit the sample area and improve the position estimation.
出处
《小型微型计算机系统》
CSCD
北大核心
2008年第11期2051-2054,共4页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(60673061)资助
湖南省自然科学基金项目(06JJ50111)资助
关键词
无线传感网络
定位
移动
蒙特卡罗
wireless sensor networks
localization
mobile
Monte Carlo