Reduced-rank DOA estimation based on joint iterative subspace recursive optimization and grid search

In this paper, we propose a reduced-rank direction of arrival (DOA) estimation algorithm based on joint and iterative subspace optimization (JISO) with grid search . The reduced-rank scheme includes a rank reduction matrix and an auxiliary reduced-rank parameter vector. They are jointly and iteratively optimized with a recursive least squares algorithm (RLS) to calculate the output power spectrum. The proposed JISO-RLS DOA estimation algorithm provides an efficient way to iteratively estimate the rank reduction matrix and the auxiliary reduced-rank vector. It is suitable for DOA estimation with large arrays and can be extended to arbitrary array geometries. It exhibits an advantage over MUSIC and ESPRIT when many sources exist in the system. A spatial smoothing (SS) technique is employed for dealing with highly correlated sources. Simulation results show that the JISO-RLS has a better performance than existing Capon and subspace-based DOA estimation methods.