一种改进的快速Mini-Norm算法

Ｋｕｍａｒｅｓａｎ和Ｔｕｆｔｓ提出的Ｍｉｎｉ－Ｎｏｒｍ算法通过对空间协方差矩阵进行特征分解，随后构造噪声子空间向量来求解方位。

A modified fast algorithm for Mini-Norm direction-of-arrival estimation

The original direction-of-arrival estimator, proposed by Kumaresan and Tufts (KT method), employs the noise-subspace projection matrix, calculated by the eigendecomposition of the spatial covariance matrix. A fast algorithm of this estimator is presented by V. T. Ermolaer and A. B. Gershman(VA method) which calculates the required mini-norm function using the special power basis instead of the eigenvector basis. In our paper, the VA method is further modified so as to simplify some parameter settings.After modification, the new method can efficiently implement the selection of parameters,confirm the convergence of the iteration and provide a substantial saving in the computational load as compared with those of KT method and VA method. Some computer simulation and experimental result are presented to verify the high performance and accuracy of the modified algorithm.