1. Source Enumeration in Array Processing Using a Two-Step Test.
- Author
-
Lu, Zhihua and Zoubir, Abdelhak M.
- Subjects
- *
SIGNAL processing , *RANDOM matrices , *EIGENVALUES , *APPROXIMATION theory , *SIMULATION methods & models - Abstract
We investigate the problem of enumerating source signals impinging on an array of sensors given noisy, limited observations. The assumption of a large sample size is indispensable for most of the existing approaches so that they suffer significant performance degradation when only a small number of samples is available. Random matrix theory can capture the asymptotic distributions of the eigenvalues accurately and provide good approximations even for a finite sample size. In this paper, we take into account the relationship and distributions of the noise and signal eigenvalues based on random matrix theory and derive a simple two-step test. Numerical simulations demonstrate that it is capable of correctly determining the number of sources in the case of small sample sizes. [ABSTRACT FROM PUBLISHER]
- Published
- 2015
- Full Text
- View/download PDF