Direction of Arrival Estimation Using Compressive Sensing

This paper presents a method for direction of arrival estimation employing compressive sensing techniques. The approach involves dimensionality reduction of received signals using singular value decomposition, followed by sparse signal recovery through L1-norm minimization. In practical implementation, the algorithm typically begins with constructing a sensing matrix that represents the array manifold, then applies SVD to compress the covariance matrix of received signals. The core estimation phase involves solving an optimization problem using algorithms like basis pursuit or LASSO, which minimize the L1-norm while maintaining data fidelity constraints. This methodology provides enhanced insights into signal transmission and reception processes and demonstrates broad application potential in practical scenarios. Furthermore, we explore optimization strategies to improve estimation accuracy and computational efficiency, including parameter tuning of regularization terms and adaptive threshold selection. The method's extensibility to other domains such as radar systems, wireless communications, and acoustic signal processing is also discussed, highlighting its potential to advance scientific and technological development.