1: \begin{abstract}

2:   Approximate message passing is an iterative algorithm for compressed sensing and related

3:   applications. A solid theory about the performance and convergence

4:   of the algorithm exists for measurement matrices having iid entries

5:   of zero mean. However, it was observed by several authors that for

6:   more general matrices the algorithm often encounters convergence

7:   problems. In this paper we identify the reason of the

8:   non-convergence for measurement matrices with iid entries and

9:   non-zero mean in the context of Bayes optimal inference. Finally we

10:   demonstrate numerically that when the iterative update is changed

11:   from parallel to sequential the convergence is restored.

12: \end{abstract}

13: