L-Shape 2-D DOA Estimation with Second-Order Statistics for Coherently Distributed Source

A computationally efficient method for 2D direction-of-arrival (DOA) estimation of coherently distributed source is presented. The key idea is to construct second-order statistics calculated analytically based on the Schur-Hadamard product steering vector. So a closed form solution to 2D DOA estimation can be derived, which does not involve any eigendecomposition or singular value decomposition as in common subspace techniques. The proposed algorithm is also a blind estimator without any prior knowledge about angular signal intensity. The method is illustrated by some numerical examples.