Abstract: In compressed sensing，many signals have the feature of block-sparse. As to this feature，it’s an urgent demand to explore an efficient recovery algorithm.We proposed a regularized adaptive matching pursuit algorithm after research and summarized the existing greedy algoritms based on block-sparse signal.This algorithm mainly in the ligt of regularized method under a condiction that a block-sparse degree is unknown，so that the signal support set can be determined more accurately by the algorithm，then we can reconstruct a signal precisely.Firstly，the algorithm initializes a sparsity degree and step size of a block signal，it maximize the correlation between residual and measurement matrix and realize the subset of the signal support can be selected. Then the algorithm update the selected subset in the second time.Finally，the exact support set is acquired through iteration.The experimental results prove that the proposed algorithm can get better reconstruction performances than other existing greedy algorithms based on block signal and it has less iterations and iteration time than the other adaptive algorithm based on block signal.

Key words: Block signal, Adaptive, Regularized, greedy algorithm