Extended SPICE Algorithm Based on the Group Sparse Model
Yue Xiao · 2021 CIE International Conference on Radar (Radar) · 2021
In this work, an improvement of the SPICE algorithm based on the group sparse regression problem is proposed. By applying the group sparse array model, an improved covariance fitting criterion is obtained, thereby avoiding cumbersome model order estimation. The algorithm can be iteratively decomposed into a group of convex optimization problems, and each problem can be solved in a closed form. In addition, through simulation experiments, we have obtained the relationship between the algorithm grouping and the number of grids and verified that the angle of arrival of the incident signal is in the same group and in different groups. The numerical results show that the algorithm has a good estimation performance.