Abstract: In response to the problem of mobile node position for WSN, a hybrid location algorithm based on MCL for mobile unknown node is proposed, which is improved in the prediction and filtering stage. First, according to the number of anchor nodes information at time t received by unknown node, it selects the most suitable method to reduce the sampled area from four cases. Second, it decides whether it needs to filter or not according to four different cases. The simulation results show that the improved algorithm can reduce the sampled area significantly, at the same time, it improves the success rate of sampling and positioning accuracy, reduces the amount of sampled frequency calculation and the energy consumption, and prolongs the network life cycle.

Key words: Wireless Sensor Network（WSN）, Monte Carlo Localization（MCL）, anchor nodes, mobile nodes

摘要： 针对无线传感器网络中移动节点定位问题，提出一种移动未知节点的混合定位算法。该算法在预测和过滤阶段均作了改进。在预测阶段，根据未知节点在t时刻接收到的锚节点信息个数区分四种情况进行讨论，选择最适合的方法来缩小采样区域;在过滤阶段，针对预测阶段四种不同的可能情况判断是否需要过滤。仿真结果表明，该算法在显著地缩小了采样区域的同时，提高了采样成功率和定位精度;减少了采样次数和计算量，降低了能耗，延长了网络的生存周期。

关键词: 无线传感器网络, 蒙特卡罗定位算法, 锚节点, 移动节点