MATLAB Implementation of 2D DOA Estimation for L-shaped Arrays Using Augmented Matrix Pencil Method

This article presents a MATLAB implementation of 2D Direction of Arrival (DOA) estimation using the augmented matrix pencil method for L-shaped arrays. We begin by introducing the fundamental concepts and background of 2D DOA estimation. The implementation involves creating array manifolds using MATLAB's sensor array toolbox and constructing covariance matrices from received signals. We then delve into the core principles of L-shaped array geometry and the augmented matrix pencil method, explaining how these techniques enhance DOA estimation accuracy through improved spatial sampling and eigenvalue decomposition.

The MATLAB implementation section details key programming aspects including: matrix construction for signal subspace estimation using SVD (svd() function), eigenvalue decomposition (eig() function) for parameter extraction, and peak search algorithms for angle estimation. The code incorporates proper array calibration and includes performance testing modules to evaluate estimation accuracy under various SNR conditions using Monte Carlo simulations.

Finally, we summarize the project's main findings regarding resolution improvement and computational efficiency, and discuss potential research directions such as extension to higher-dimensional arrays and integration with machine learning techniques for adaptive parameter selection.

This comprehensive tutorial provides detailed insights into 2D DOA estimation using augmented matrix pencil methods with L-shaped arrays, enabling researchers to effectively apply these concepts and techniques in their own work through practical MATLAB implementations.