14.Joint symbol detection and channel estimation based on variational Bayesian in MIMO relay system

Based on the variational Bayesian inference and tensor decomposition, a joint symbol detection and channel estimation of MIMO relay system under Rayleigh fading channel is proposed. In this algorithm, the symbol matrix and the channel matrix are modeled as unknown matrices by introducing the implicit parameter variables, and recursive formula of the unknown matrix is deduced by the principle of maximizing the model evidence. Compared with the traditional estimation model based on tensor model, this algorithm introduces the a priori information of the channel into symbol estimation and channel estimation process to improve the estimation performance. The simulation results show that the proposed algorithm has high estimation accuracy compared with the existing joint estimation algorithm based on ALS (Alternating Least Square) and PLS (Partial Least Squares).