一种指数渐消记忆加权滤波的内交点定位算法

Internal intersection location algorithm with exponential fading memory weighted filtering

针对无线室内定位存在的非视距误差及信号跳变的问题,提出了一种基于指数渐消记忆加权测距滤波的内交点定位算法。为了减小非视距误差对无线系统定位精度的影响,提出了一种内交点定位算法,它无需识别基站是否处于视距环境,也无需任何先验的信息,仅基于当前时刻的各基站测距信息实现目标位置的解算。针对无线信号易跳变的问题,采用指数渐消记忆加权算法对测距信息进行滤波处理,通过选取合适的渐消记忆因子有效调节测距信号跳变的幅度。实测数据结果表明:利用所提出的方法定位精度优于2m,显著优于传统的加权最小二乘定位算法,且定位结果的跳变现象明显减弱。

In order to solve the problem of non-line-of-sight （NLOS） errors and signal hopping in indoor wireless location, an interior intersection location algorithm based on exponential fading memory weighted ranging filtering is proposed to reduce the NLOS＇s effect on the positioning accuracy. This method does not need to identify whether the base station is in the line-of-sight environment, nor does it need any prior information. It only needs to calculate the location of the target based on the ranging information of the base stations at the current time. For the problem of signal hopping, an exponential fading memory weighted algorithm is applied to filter the ranging information of base stations, and the appropriate fading memory factor is selected to weaken the jump of wireless signals. Experimental results show that the precision of the proposed method is better than 2 m, which is significantly superior to that of the traditional weighted least squares algorithm, and the jumping phenomenon of the positioning results is obviously weakened.