基于信道模糊关联识别的NLOS测距误差补偿算法

针对NLOS测距存在误差的问题，提出一种基于信道模糊关联识别的NLOS测距误差补偿算法.根据先验信道特征参数分布信息构造信道特征参数模糊隶属矩阵，并利用灰色关联分析方法计算归一化权值矩阵，进而获得模糊综合评价矩阵来对信道环境进行识别.在此基础上，根据信道识别结果构建Huber代价函数，通过Huber线性回归方法对原始测距结果进行迭代重构，将重构结果作为Kalman滤波的测量值进行滤波.仿真结果表明，所提算法可以有效提高NLOS环境下的测距精度，在信噪比大于-2 dB时可以达到厘米级测距精度.

NLOS Ranging Error Compensation Algorithm Based on Fuzzy Association Channel Identification

Aiming at the problem of ranging errors in non line of sight (NLOS) ranging, an NLOS ranging error compensation algorithm is proposed based on fuzzy association channel identification. The algorithm constructs fuzzy membership matrix of channel feature parameters based on the prior channel feature parameter distribution information, and uses gray correlation analysis method to calculate the normalized weight matrix, so that to obtain the fuzzy comprehensive evaluation matrix to identify the channel environment. On this basis, the Huber residual cost function is constructed according to the channel identification result, and the original ranging result is iteratively reconstructed by Huber linear regression method, which is filtered as the measured value of Kalman filter. Simulation show that the proposed algorithm can improve the ranging accuracy under NLOS effectively, and the range accuracy of this algorithm can reach centimeter level when signal-to-noise ratio is -2 dB.