MATLAB Implementation of the MUSIC Algorithm for High-Resolution DOA Estimation

The MUSIC (Multiple Signal Classification) algorithm is a classic Direction of Arrival (DOA) estimation technique widely used in radar systems, wireless communications, and acoustic signal processing. This algorithm leverages the orthogonal characteristics between signal subspace and noise subspace to achieve high-resolution estimation of signal source angles. The core principle of the MUSIC algorithm involves performing eigenvalue decomposition on the covariance matrix of received signals to separate the signal space into signal subspace and noise subspace. Since the direction vectors of signal sources are orthogonal to the noise subspace, the MUSIC spectrum exhibits peaks at the true signal source directions during spatial scanning. MATLAB implementation of the MUSIC algorithm typically follows these key steps: First, construct the covariance matrix from array received signals using matrix operations like R = X*X'/N where X represents the signal matrix and N is the number of snapshots. Second, perform eigenvalue decomposition using [V,D] = eig(R) to obtain eigenvectors corresponding to signal and noise subspaces. Third, construct the MUSIC spectrum using P = 1./sum(abs(V_noise'*a(theta)).^2) where a(theta) is the steering vector and V_noise contains noise subspace eigenvectors. Finally, identify signal source angles through peak search in the MUSIC spectrum using findpeaks() or similar functions. The main advantage of the MUSIC algorithm lies in its high-resolution capability, particularly under high Signal-to-Noise Ratio (SNR) conditions where it can resolve closely spaced signal sources that conventional beamforming methods cannot distinguish. However, the algorithm is sensitive to array calibration and signal model assumptions, requiring careful consideration of array errors and coherent signal effects in practical applications. For beginners, it's recommended to start with Uniform Linear Array (ULA) configurations to understand the MUSIC algorithm fundamentals before progressing to more complex array structures. By adjusting array parameters and signal scenarios through code modifications, users can thoroughly explore the algorithm's performance boundaries and application techniques. Key MATLAB functions involved include eig() for eigenvalue decomposition, svd() for singular value decomposition, and array manifold generation functions for different antenna configurations.