基于RSSI和改进MCL算法的移动节点定位跟踪设计  

Locating and Tracking for Mobile Sensor Node Based on RSSI and Improved MCL Algorism

作　　者：孔素真[1] 孙雅娟[1] 

机构地区：[1]河南牧业经济学院信息工程系,郑州450011

出　　处：《计算机测量与控制》2014年第5期1637-1639,1643,共4页Computer Measurement &Control

基　　金：河南省重点科技攻关计划项目(122102210487)

摘　　要：由于传统节点定位方法大多针对静止传感器网络,不能适用于网络结构和节点位置动态变化的移动传感器网络,提出了一种基于RSSI测距和改进的MCL(Monte Carlo Localization)算法的移动传感器节点定位跟踪方法;首先描述了经典MCL算法和接收信号强度RSSI测距方法,然后设计了一种改进的MCL算法,将传统的MCL方法预测粒子位置的过程即预测和滤波两个阶段,更新为锚节点TTL受控泛洪方式广播自身位置、采用拉格朗日插值法预测节点下一时刻的位置和速度、求取锚盒采样区域、k跳锚节点粒子滤波和根据预测下一时刻的节点位置和速度与当前时刻的位置信息确定各粒子权重的5个阶段;采用仿真器MCL-Simulator进行仿真,结果证明:文中方法能有效实现移动节点的定位,与其它方法相比,具有较小的平均定位误差,具有很强的可行性。Aiming at the traditional localization method usually aiming at the static wireless sensor network, can not suit the mobile sen- sor network with dynamic changing network structure and node distance, a locating and tracking method for mobile sensor node was proposed based on RSSI and improved MCL. Firstly, the RSSI ranging distance method and classic MCL was introduced, then an improved MCL algo- rithm was proposed by renewing the process of classic MCL such as predicating and filtering to five stages as anchor node broadcast its posi- tion using limit TTL flooding protocol, using Lagrange interpolation to predict the position and speed in the next time, obtaining anchor sam- pling area, particle filter and according the predicting position and speed in the next time and the current position to weight particle. The ex periment was simulated in the MCL--Simulator shows the method in this paper can realize the node moving, compared with the other meth- ods, it has less average locating error, so it has big feasibility.

关 键 词：移动节点 定位跟踪 蒙特卡罗 粒子 

分 类 号：TP212.9[自动化与计算机技术—检测技术与自动化装置] TN929.5[自动化与计算机技术—控制科学与工程]