基于压缩感知信号重建的自适应正交多匹配追踪算法*

近年来出现的压缩感知理论为信号处理的发展开辟了一条新的道路，不同于传统的奈奎斯特采样定理,它指出只要信号具有稀疏性或可压缩性,就可以通过少量随机采样点来恢复原始信号。在研究和总结传统匹配算法的基础上，提出了一种新的自适应正交多匹配追踪算法(adaptive orthogonal multi matching pursuit,AOMMP)用于稀疏信号的重建。该算法在选择原子匹配迭代时分两个阶段，引入自适应和多匹配的原则，加快了原子的匹配速度，提高了匹配的准确性，实现了原始信号的精确重建。最后与传统OMP算法

Adaptive orthogonal multi matching pursuit algorithm for signal reconstruction based on compressive sensing

The newly emerging compressive sensing theory in recent years has opened up a new path for the development of signal processing, which describes that it can reconstruct the original signal from a small amount of random sampling as long as the signal is sparse or compressible, which disobeys with the traditional Nyquist sampling theorem. Based on the study and summarize of the traditional matching algorithm, this paper presented a new adaptive orthogonal matching pursuit algorithm (AOMMP) for the reconstruction of the sparse signal. The algorithm divided each iteration into two stages for the choice of matching atoms, which accelerated the matching speed of the atom and improved the accuracy of the matching, ultimately led to exact reconstruction of the original signal. Finally, compared the AOMMP algorithm with the traditional OMP algorithm under the software simulation. Experimental results show that the AOMMP reconstruction algorithm is superior to traditional OMP algorithm on the reconstruction quality and the speed of the algorithm.