水声传感器网络信号到达时间差目标定位的最小二乘法估计性能

A weighted total least squares method based on double error model is proposed for the target localization in the simultaneous presence of measurement error and sensor node position error in underwater acoustic sensor networks. A TDOA target localization model with double errors is established. The weighted total least squares method is used to solve the proposed model, and the expression of the weights is derived. The proposed algorithm model is to use the prior information of measurement error variance and sensor node position error variance to obtain the final target position estimate by iteration method. The simulation and water tank experimental verification of the proposed method were made by simulating the simultaneous existence of the measurement error and the sensor node positions error. The results show that the proposed target localization method based on weighted total least squares has higher localization performance.