基于模拟退火的信号稀疏分解研究  

Research of Signal Sparse Decomposition Based on Simulated Annealing

作　　者：徐帅[1,2] 赵云龙[1,2] 张江泉[2] 

机构地区：[1]西南交通大学信息科学与技术学院,四川成都610031 [2]西南交通大学峨眉校区,四川峨眉614202

出　　处：《通信技术》2008年第11期188-190,共3页Communications Technology

摘　　要：信号稀疏分解广泛应用于图像和信号处理领域,特别是在数据压缩和数据存储、特征提取领域应用广泛。但信号稀疏分解本身是典型的NP困难问题,要成功的进行信号稀疏分解是十分困难的。文中利用模拟退火算法来快速寻找Matching Pursuit(MP)过程每一步的最优原子,提出了一种基于模拟退火的信号稀疏分解算法。仿真结果表明该算法能有效和快速地进行信号稀疏分解。Sparse decomposition of signal is extensively applied to image and signal processing fields, especially to data compressing, data storing, and feature extracting. However, sparse decomposition of the signal is a NP difficult problem. It is very difficult to make successful sparse decomposition of the signal. In this paper, the simulated annealing algorithm is used to find quickly the optimal atom in the process of Matching Pursuit （MP）, and then an algorithm of sparse decomposition of the signal based on simulated annealing is proposed. Simulation results show that this algorithm can effectively and quickly make sparse decomposition of the signal.

关 键 词：稀疏分解 过完备原子库 模拟退火 信号分析 

分 类 号：TP911.72[自动化与计算机技术]