一种基于CKF的无线传感器网络分布式定位算法

为提高无线传感器网络节点定位的精度，降低算法计算复杂性，提出了一种基于容积卡尔曼滤波的无线传感器网络分布式节点定位算法。该算法假定移动锚节点按预定路径在传感区域移动，并周期性广播自身位置信标信息；每个未知位置节点首先收集多个锚节点信标信息及信号强度信息，然后估算出锚节点信标位置与未知节点的距离，最后在未知节点上运用容积卡尔曼滤波算法完成自身位置的分布式定位。仿真结果表明：本文所提算法具有优良的定位性能，定位精度和无迹卡尔曼滤波算法相当，明显优于极大似然估计定位算法，而计算复杂性则低于无迹卡尔曼滤波算法。

A Distributed Localization Algorithm Based on Cubature Kalman Filter in Wireless Sensor Networks

To improve the localization accuracy and decrease the computation complexity,a distributed node local?ization algorithm based on cubature kalman filter in wireless sensor networks is proposed. The algorithm supposes that a mobile beacon moves by predetermined trajectory around a sensor field,and periodically broadcasts its cur?rent location. Each sensor collects the location and RSS of beacons,measures the distance between itself and the beacon,and individually calculates their locations via a Cubature Kalman Filter algorithm. Simulations show that the proposed algorithm has a good localization performance,the localization accuracy is same to the UKF algorithm, better than the MLE localization algorithm,and the computation complexity is smaller than the UKF algorithm.