On Optimization Algorithm of SAMP Step Size Based on Compressed Sensing

The Sparsity Adaptive Matching Pursuit (SAMP) algorithm needs not to know a priori information of sparsity, which makes it has unique advantages compared to other greedy algorithms. In the reconstruction process, the step size can be changed with stage, which increases its reconstruction accuracy. Aiming at the step size of SAMP, it is unreasonable, will only increase and it is not decrease properly. If the result of one iteration is poor, it will continue on the larger step. This reduces the reconstruction quality of the algorithm. This paper designs a step which can be appropriately reduced, and proposes SAMP Optimized Variable Step reconstruction algorithm based on compressed sensing. The simulation results show that the improved algorithm can improve the reconstruction precision compared with the original algorithm, and has high accurate reconstruction probability, and reduces the reconstruction time by 20%.