贪婪算法重构信号性能研究  

Study on Performance of Greedy Algorithms for Signal Reconstruction

作　　者：季秀霞[1] 卞晓晓[1] 

机构地区：[1]南京航空航天大学金城学院,江苏南京211156

出　　处：《计算机仿真》2013年第11期255-259,共5页Computer Simulation

基　　金：航空科学基金(2011ZC52034)

摘　　要：在信号重构优化问题的研究中,压缩感知理论是一种充分利用信号稀疏性或者可压缩性的全新的信号采样理论,通过采集少量的信号值就可以实现稀疏性或可压缩信号的精确重构。为了准确重构信号,利用压缩感知理论与一系列用于信号重构的贪婪算法,提出了一种改进的正交匹配追踪算法,通过设置限性下降的迭代阈值挑选原子,采用简单有效的剔除小系数原子更新支撑集,对信号的稀疏度进行自适应估计,从而实现信号重构。针对一维信号和二维图像做了仿真,比较各类贪婪算法的重构精度和重建构速度方面的差异。仿真结果表明,改进算法可以准确有效重构信号,同时运算量也较低,具有一定的优势。Compressive sensing is a novel signal sampling theory under the condition that the signals are sparse or compressible. In this case, the small amount of signal values can be reconstructed accurately when the signals are sparse or compressible. This paper reviewed compressive sensing theory and a series of greedy pursuit algorithms and presented an improved orthogonal matching pursuit algorithm that selects atoms and determines candidate set more quickly with the linear decreasing threshold, removes lower-energy atoms from the candidate set, and estimates spar- sity degree adaptively. The simulation experiments was done for one-dimension signals and two-dimension image sig- nals and the reconstruction accuracy and reconstruction speed of a variety of greedy algorithms were compared. The simulation results show that the improved algorithm can accurately reconstruct signal with low run speed, has certain advantages.

关 键 词：压缩感知 信号重构 贪婪算法 匹配追踪 

分 类 号：TP911.73[自动化与计算机技术]