基于自适应SR-CKF的序贯式WSNs目标跟踪算法

摘要

Based on adaptive square root cubature kalman filter(SR-CKF),a sequential WSNs dynamic target tracking algorithm is proposed in this paper to solve the problem of dynamic target tracking in wireless sensor networks(WSNs),which estimates the states and predicts the target location sensors that acquire from dynamic systems.The computational complexity is reduced by transferring the mean of the target state and the factors to the square root of covariance matrix directly.By allocating the object tracking to each node of the dynamic cluster set,collision and interference in wireless communication are also reduced and node communication and computational burden are also lowered.Based on covariance matching principle,an adaptive SR-CKF is built to improve the robustness of the whole system under circumstance of the adverse observational information.The simulation shows that the sequential WSNs tracking algorithm can effectively improve the tracking accuracy and the tracking stability and reduce the communication energy dissipation of wireless sensor nodes.%针对无线传感器网络(WSNs)动态目标跟踪问题,即通过对传感器获取的动态系统状态进行估计,预测目标的位置.提出一种基于自适应平方根容积卡尔曼(SR-CKF)的序贯式WSNs动态目标跟踪算法.该算法在运算过程中直接传递目标状态均值和协方差矩阵的平方根因子,降低了计算的复杂度.将目标跟踪过程序贯式地分配到动态簇集的每一个节点上,减小了无线通信过程中碰撞和干扰现象的发生,降低了节点通信和计算负担.针对不良观测信息,基于新息协方差匹配原理,建立了自适应SR-CKF,提高了整个系统的鲁棒性.实验仿真结果表明,本文提出的基于自适应SR-CKF的序贯式WSNs目标跟踪算法有效的提高了跟踪的精度和稳定性并且减小了传感器节点间通信的能量损耗.