基于粒子滤波的分布式目标跟踪算法

根据无线传感器网络分布式目标跟踪的特性，提出一种改进粒子滤波算法。将簇内各节点最新的观测数据用极大似然估计法得到目标的状态信息，该信息作为参考分布更换粒子滤波部分粒子，引入模糊推理的数据融合方法为各个节点滤波结果分配不同权值，通过加权平均法得到目标的状态信息。仿真实验表明该算法能有效提高目标跟踪的精度。

Distributed Target Tracking Algorithm Based on Particle Filtering

For the features of distributed target tracking in Wireless Sensor Network（WSN）, this paper proposes a new algorithm to improve the particle filtering. The latest data are got from the sensor nodes to estimate the target state（its instantaneous location and velocity） by the maximum likelihood estimation method. Then part of particles is replaced with the ones distributed referred to the state. After that, the fuzzy reasoning method is introduced for data fusion. The weighted average method is used to get the final target state. Simulation results indicate that the algorithm can improve the tracking accuracy effectively.