基于递推总体最小二乘的机载单站无源定位算法

对于机载单站无源定位中观测向量和系数矩阵均存在噪声的情况,采用经典的最小二乘(LS)算法会导致估计结果有偏,而一般的批处理算法运算量大,不满足定位的实时性要求.为此提出一种递推总体最小二乘定位(RTLS)算法.在建立机载无源测向定位模型的基础上,引入由系数矩阵和观测向量构成的增广矩阵,利用矩阵分解的性质建立了增广自相关逆矩阵和最右奇异向量的递推方程,从而推导出RTLS定位算法.仿真结果表明,RTLS算法的收敛速度和定位精度优于RLS算法和TWDRLS算法,并且实现了TLS算法对目标位置的实时估计,能够定位需求.

Recursive Total Least Squares-Based Airborne Single-Observer Passive Location Algorithm

In the application of airborne single-observer passive wireless radio frequency injection location, both the observation vector and coefficient matrix are noisy, the application of the typical least square (LS) algorithm will make the location estimation biased. Furthermore, such general batch processing algorithm is rather complicated and needs large amount of computation which fails to comply with the real-time requirement of passive location. Therefore, a recursive total least squares (RTLS) algorithm is proposed. At first, airborne passive bearing location model is build.Based on the above an augmented matrix made up of coefficient matrix and observation vector is introduced, then the recursive equations of the augmented inverse correlation matrix and the most right singular vector are obtained according to the properties of matrix decomposition. Simulation results show that convergence rate and location accuracy of RTLS algorithm are better than those of RLS algorithm and TWDRLS algorithm, and the real time estimation of TLS algorithm is achieved, which can meet the requirement of location.