一种改进的双库MP信号稀疏分解方法.pdfVIP

Two dictionaries Matching Pursuit for sparse decomposition of signals Peng Xu Dezhong Yao Center of Neuroinformatics, School of Life Science Center of Neuroinformatics, School of Life Science and Technology, University of Electronic Science and and Technology, University of Electronic Science and Technology of China, ChengDu, 610054, China Technology of China, ChengDu, 610054, China leisure_xp @ dy ao@ Abstract Signal may be decomposed sparsely and power-focally in an over-complete dictionary with Matching Pursuit (MP). In this paper, proposed is a modified MP method named two dictionaries MP to decompose signal more sparsely. In the iteration procedure of the two dictionaries MP, the over-complete dictionary is classified into two separate dictionaries with the selected and unselected atoms, and in each iteration, the algorithm was designed to have more chances than the original MP to choose the atom in the selected atom dictionary as the optimal atom by a simulate annealing threshold function, thus the algorithm avails for a more sparse decomposition. The decomposition results for a cosine-modulated exponential signal and an actual speech signal showed that the proposed two dictionaries MP could decompose signal more sparsely. Keywords: Sparsity; Over-complete dictionary; Matching pursuit; Simulate annealing threshold function 1. Introduction Hunting for a compact and effective signal expression is one of the research focuses in the signal processing (SP) because the more compact expression the signal has, the higher ratio compression the sig