一种稀疏度自适应正交多匹配追踪重构算法

A Sparsity Adaptive Orthogonal Multi Matching Pursuit Algorithm

压缩感知理论是一种利用信号稀疏性或可压缩性对信号进行采样同时压缩的新颖的信号采样理论。针对稀疏度未知信号重构问题,提出了一种稀疏度自适应正交多匹配追踪重构算法。该算法在广义正交匹配算法(generalized orthogonal multi matching pursuit,GOMP)基础上结合稀疏自适应思想。根据相邻阶段信号能量差自适应调整当前步长大小选取支撑集的原子个数,先大步接近,后小步逼近信号真实稀疏度,从而实现对信号精确重构。实验仿真结果表明,该算法能有效精确重构信号。具有良好的重构性能和较高的重构效率。

Compressive sensing is a novel signal sampling theory under the condition that signal is sparse or compressible. In this case, the small amount of signal values can be reconstructed accurately when the signal is sparse or compressible. A sparsity adaptive orthogonal multi matching pursuit algorithm is proposed for reconstruc- tion without prior information of the sparsity. It realizes the close approach of signal sparse step by step based on the frame of Generalized Orthogonal Multi Matching Pursuit（GOMP）. In the beginning, it uses high value of step size to approach the true sparsity of the signal rapidly. Then it switchers to small value of step size to achieve the precise approach of signal. Finally, it realizes the precise reconstruction of sparse signal. At last, the experimental results show that the proposed algorithm can get better reconstruction performances and speed than other algorithms.