A Fast Convergence Phase Estimation Method Based on Expectation-Maximization Algorithm
Wang, Ge, Hongyi Yu · ASME Press eBooks · 2011
Expectation-Maximization (EM) Algorithm is an approach to iterative computation of maximum-Likelihood estimates when the observation can be viewed as incomplete data. This paper aimed to the convergence speed problem, presents a fast convergence method of phase estimation based on EM, which modify the priori probability with posteriori probability in the EM algorithm iterative. For the QPSK carrier phase estimation, compared with the traditional method that assumed the transmitted signals is independent and no prior information. Analysis and simulation results show that without affect the estimation performance, the algorithm convergence speed is faster. Meanwhile, the algorithm realizes the joint maximum-likelihood (ML) estimation of phase estimation and symbol detection.