Least square approach for subspace intersection method of bearing estimation in shallow water

To address the challenge of bearing estimation in shallow water, a least square approach for subspace intersection method is proposed. The proposed method includes a constrained least square subspace intersection and total least square subspace intersection. It is shown that MUSIC (multiple signal characterization) is a special case of constrained least square subspace intersection when the received signal only contains one normal mode. Simulations show that the performance of the proposed least square subspace intersection method is better than that of the original subspace intersection (SI).