An Improved Sparsity Adaptive Matching Tracking Algorithm Based on SAMP

Considering the shortcomings of the sparsity adaptive matching pursuit (SAMP) algorithm such as long reconstruction time and fixed step size, a weighted regularization variable step sparsity adaptive matching tracking (WRVS-SAMP) algorithm is proposed. The algorithm estimates the initial sparsity from the perspective of an atom matching test, Li weighted regularization is used to improve the accuracy of candidate atoms, and PCA is used to reduce the dimension of the candidate atom set matrix to reduce the interference of redundant information, the residual energy value of adjacent stages of signal reconstruction is compared with the pre-set threshold, and the step size and the termination iteration of the algorithm are adjusted adaptively to realize the accurate signal reconstruction. The simulation results show that the reconstruction accuracy of the improved algorithm is better than that of the SAMP algorithm, and the reconstruction rate is significantly improved.