High-resolution methods without eigendecomposition for locating the acoustic sources
Abstract
This paper deals with the estimation of the direction of arrival (DOA) of radiating sources. High-resolution methods are used with an approximation of the noise subspace projection matrix. The developed algorithm improves the performance and reduces the complexity of the Ermolaev and Gershman method (IEEE Transactions on Signal Processing, 1994, 42(9)). The proposed algorithm uses the smallest eigenvalue of the spectral matrix as the threshold eigenvalue, which guarantees that the obtained projection matrix is an estimate of the noise subspace. We present some numerical results for verifying the performance of the theoretical results obtained.