Compressed sensing of superimposed chirps with adaptive dictionary refinement

作     者：HU Lei ZHOU JianXiong SHI ZhiGuang FU Qiang 

作者机构：ATR Key Laboratory National University of Defense Technology 

出 版 物：《Science China(Information Sciences)》 (中国科学:信息科学(英文版))

年 卷 期：2013年第56卷第12期

页      面：171-185页

核心收录：

学科分类：0711[理学-系统科学] 07[理学] 08[工学] 080401[工学-精密仪器及机械] 0804[工学-仪器科学与技术] 080402[工学-测试计量技术及仪器] 

基　　金：supported by National Natural Science Foundation of China(Grant Nos.60972113 61101179) 

主　　题：compressed sensing(CS) superimposed chirps dictionary refnement variational Bayesian approximation expectation-maximization(EM) algorithm 

摘      要：The compressed sensing(CS)theory shows that accurate signal reconstruction depends on presetting an appropriate signal sparsifying *** CS of superimposed chirps,this dictionary is typically taken to be a waveform-matched dictionary formed by blindly discretizing the frequency-chirp rate ***,since practical target parameters do not lie exactly on gridding points of the assumed dictionary,there is always mismatch between the assumed and the actual sparsifying dictionaries,which will cause the performance of conventional CS reconstruction methods to degrade *** address this,we model the waveformmatched sparsifying dictionary as a parameterized one by treating its sampled frequency-chirp rate grid points as the underlying *** a consequence,the sparsifying dictionary becomes refnable and its refnement can be achieved by optimizing the underlying *** on this,we develop a novel reconstruction algorithm for CS of superimposed chirps by utilizing the variational expectation-maximization(EM)*** alternating between steps of sparse coefcients estimation and dictionary parameters optimization,the algorithm integrates the process for dictionary refnement into that of signal reconstruction,and thus can achieve sparse reconstruction and dictionary optimization *** results demonstrate that the algorithm efectively deals with the performance degradation incurred by dictionary mismatch,and also outperforms the state-of-the-art CS reconstruction methods both in compressing the signal measurements and in suppressing the measurement noise.