Classical OMP Algorithm for Signal Sparse Decomposition

This program demonstrates several distinctive features and advantages in signal processing applications. Signal sparse decomposition has become a prominent research area in modern signal processing, and this implementation focuses on the classical Orthogonal Matching Pursuit (OMP) algorithm - an effective method for sparse signal representation. The code incorporates advanced optimization techniques including parallel computing and GPU acceleration, significantly improving algorithm execution efficiency and runtime performance. The implementation follows a step-by-step approach where at each iteration, the algorithm selects the dictionary atom most correlated with the current residual, then solves a least-squares problem to update the sparse coefficients. Future developments will focus on enhancing the program's capabilities to better meet user requirements and expanding its applications to broader domains such as image processing and speech signal analysis, with planned improvements including adaptive dictionary learning and real-time processing features.