采用距离无偏估计的加权最小二乘定位算法

随着WiFi网络的广泛覆盖，基于接收信号强度的定位技术成为研究热点。针对已有基于接收信号强度定位算法定位性能不高的实际问题，提出一种基于距离无偏估计的加权最小二乘定位算法。该方法首先利用接收信号强度观测模型计算得到信号源与传感器之间距离的无偏估计量，然后根据距离计算公式建立方程组；接着把距离的无偏估计量代入方程组得到关于信号源位置的线性最小二乘模型，同时计算线性最小二乘模型中的噪声协方差矩阵；最后运用加权最小二乘方法计算得到信号源位置的估计量。该文对所提算法进行了充分的计算机仿真，仿真结果表明:在不同的定位环境下，所提算法的定位性能均优于传统加权最小二乘算法和最佳线性无偏估计算法。 

Weighted Least Square Localization Algorithm Using the Unbiased Estimate of the Distance

Received signal strength based positioning technology has become a hot topic with the extensive coverage of the WiFi network. Due to the bad positioning performance of the existing received signal strength based localization algorithm, a weighted least square localization algorithm based on the unbiased estimate of the distance is proposed. In the proposed algorithm, firstly the unbiased estimate of the distance between the target and the sensors were calculated according to the received signal strength observation model. Then linear equations were established based on the distance calculation formula. After that, the unbiased estimate of the distance was substituted into linear equations to establish the linear least squares model on the source position, simultaneously calculated the noise covariance matrix of the linear least squares mode. Finally the estimation of the signal source position was obtained by weighted linear least square method. In the paper, full computer simulations had been done. Simulations results show that the positioning performance of the proposed algorithm is better than both traditional weighted least squares algorithm and best linear unbiased estimation algorithm under different positioning environment.