Target Angle Estimation - Radar Direction of Arrival (DOA) Algorithms

Direction of Arrival (DOA) estimation algorithms for radar targets represent a critical research direction in signal processing, primarily used to determine the wave arrival direction of signal sources. Among these, the Multiple Signal Classification (MUSIC) algorithm stands as a classical high-resolution DOA estimation method with extensive applications in radar, sonar, and wireless communication systems.

The core concept of the MUSIC algorithm leverages the orthogonality between signal subspace and noise subspace to achieve super-resolution angle estimation. Its processing flow comprises several key steps implemented in typical code implementations: first constructing a covariance matrix from array reception data using matrix operations, then performing eigenvalue decomposition to separate signal and noise subspaces through mathematical libraries like numpy.linalg.eig(), and finally conducting spectral peak search to identify directions orthogonal to the noise subspace - where the peaks correspond to estimated target angles through peak detection algorithms.

Compared with conventional beamforming methods, the MUSIC algorithm demonstrates superior angular resolution capability, enabling distinction between two targets with angular separation smaller than traditional method limits. However, the algorithm presents certain limitations including requiring precise knowledge of signal source numbers (often estimated via information criteria like AIC/MDL), high dependence on array calibration accuracy, and performance degradation under low signal-to-noise ratio conditions. To address these constraints, subsequent developments have produced improved MUSIC variants such as Root-MUSIC (implementing polynomial root-solving) and Weighted-MUSIC (applying optimal weighting matrices) to enhance algorithm robustness and practical applicability.