基于TLS的改进子空间投影算法

Modified Subspace Projection Algorithm Based on TLS

针对经典MUSIC算法在信源相干、低信噪比、小快拍数等非理想环境下性能失效的问题,提出了一种改进的基于TLS的加权子空间投影算法。首先对阵列接收的数据协方差矩阵进行重构处理,以达到解相干目的;其次充分利用子空间信息,基于总体最小二乘拟合方法对特征值进行拟合修正,基于修正MUSIC算法思想,利用校正后的噪声特征值和信号特征值分别对噪声子空间和信号子空间进行加权处理,得到改进后的噪声子空间和信号子空间,并将两者结合得到新的空间谱函数;最后进行谱峰搜索,完成信号源的波达方向估计。仿真结果表明,改进后的算法既适用于相干信号环境,在低信噪比、小快拍数及信号入射角度间隔较小的情况下,又能有效估计出信源的波达方向。

A modified TLS-based weighted subspace projection algorithm is proposed to address the problem of performance failure of the classical MUSIC algorithm in non-ideal environments such as source coherence,low SNR and small snapshots number.First,the algorithm reconstructs the received data covariance matrix for the purpose of decoherence.Then,it makes full use of the subspace information to fit and modify the eigenvalues based on the total least squares fitting method,and weights the noise subspace and signal subspace based on the idea of the modified MUSIC algorithm using the corrected noise eigenvalues and signal eigenvalues,respectively,to obtain the improved noise and signal subspaces,and combines the two to obtain the new spatial spectral function.Finally,the spectral peaks are searched to complete the DOA(Direction of Arrival)estimation of the source target.Simulation results indicate that the modified algorithm is not only suitable for coherent signal environments,but also effective in estimating the DOA of the source with low SNR,small snapshots and small angle interval.