DOA estimation using Matrix Pencil and ESPRIT methods using single and multiple snapshots

In this presentation, we compare the performance of the Matrix Pencil (MP) and ESPRIT (Estimation of Signal Parameters using Rotational In-variance Technique) methods for direction of arrival estimation using single and multiple snapshots. The Matrix Pencil method is a technique that directly deals with the data to estimate the poles corresponding to the directions of arrival, and works well under a correlated signal case. This is in contrast to ESPRIT, a statistical subspace based estimation technique, which will work for the correlated case after applying some additional spatial smoothing techniques to the covariance matrix of the data. Simulation results are provided to compare the performance between these two methods. The results clearly demonstrates that when a small number of snapshots, including a single one, is available to process the data, the MP method always performs better than the ESPRIT method.