无线传感器网络节点自身定位算法  

Self-Localization Algorithm for Wireless Sensor Network

作　　者：廖先林[1] 耿娜[1] 石凯[1] 赵林亮[1] 

机构地区：[1]东北大学信息科学与工程学院,辽宁沈阳110004

出　　处：《东北大学学报（自然科学版）》2007年第6期801-804,共4页Journal of Northeastern University(Natural Science)

基　　金：国家高技术研究发展计划项目(2003AA712032)

摘　　要：无线传感器网络中,节点的自身定位至关重要,在军事和民用领域中有着广泛的应用前景.目前的定位算法主要分为两种类型,即基于距离的定位算法和距离无关的定位算法.这两种类型的算法各有优势和不足.考虑了两种算法的优缺点,提出了一种廉价实用的自身定位算法,该方法通过RSSI测距技术测量节点间点到点的距离,并在多跳网络中对累加距离进行广播,最后用几何方法和最大似然估计法进行定位.仿真显示该算法在平均测距误差为10%,锚节点比为10%时,平均定位误差约为节点射频通信距离的20%.In wireless sensor networks, the node self-localization is vital and promising to a wide scope of both military and civilian applications. In this respect there are two types of algorithms, i.e. range-based and range-irrelevant, either has its own advantages and disadvantages. Taking account of those of the two algorithms, a low-cost and practical algorithm is proposed, using RSSI （received signal strength indicator） for range t between nodes and propagating the cumulative range in meter in a multijetter network, then localizing nodes geometrically via maximum likelihood estimation. Simulation results showed that by the algorithm the average error is 10 %, while the position error is about 20 % of nodes＇ RF transmission range if the ratio of anchor nodes is 10%.

关 键 词：传感器网络 自身定位 最大似然估计 锚节点 射频通信距离 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]