A High Resolution Algorithm for Nonplanar Array with Arbitrary Geometry without Source Number Estimation

The performances of most of the high resolution methods always depend on the estimation of the source number. In real application, when the estimated number of signals is not correct, the orthogonality between signal subspace and noise subspace can not be maintained any more. And the performance of DOA estimation algorithm will deteriorate severely. In this paper, a high resolution algorithm called m-MVM without source number estimation and eigendecomposition for Direction-Of-Arrival (DOA) estimation is proposed, which is an improvement of the Minimum Variance Method (MVM). Furthermore, it is suitable to nonplanar arrays of arbitrary geometries. Some representative computer simulations are presented to illustrate the performance comparison between different algorithms and different arrays.