The research of Aitken-EM accelerate algorithm in Non-Gauss noise

The EM iterative algorithm is commonly used in recent years for missing data, which has the character of easy and popular applicability. But the EM algorithm has a weakness that is slow convergence speed. A new acceleration algorithm is proposed which unifys the EM iterative algorithm and the Aitken acceleration method in order to solve the problem. Using the Aitken-EM acceleration algorithm to analysis the multiuser detection under the Non-Gaussian noise simulation, the results show that the proposed algorithm has a good performance, and compare with EM algorithm, the convergence speed is better than standard EM algorithm.