Expectation Maximization - Matrix Pencil Method for Direction of Arrival estimation

Abstract

In this paper, we present a Direction of Arrival estimation algorithm by combining Expectation Maximization (EM) and Matrix Pencil Method (MPM). It is possible that, when using a long uniform array, a few elements may fail, or the half-wavelength spacing between the elements may not be satisfied due to the spatial constraint. Under such situation, the array is known as sublattice array, which is a NonUniform Linear Array (NULA). Well known Matrix Pencil Method (MPM) cannot be directly applied for sublattice array, because the MPM technique exploits the uniform array geometry. The proposed method is a combination of the EM and the MPM technique. The EM algorithm estimates the output of the omitted or missing elements to form an equivalent output of the uniform linear array, and then, the application of MPM is possible. This technique improves the performance significantly and also has the computational advantage, when the sublattice array is considered.