Iterative SAGE Channel Tracking Algorithm for Coded MIMO Time-Selective Fading Channels

Multiple input multiple output (MIMO) systems have been proven to increase capacity hence allowing faster data rates. To achieve this capacity, perfect channel knowledge must be known at the receiver. When mobility is involved, the channel varies with time and has to be estimated precisely. Conventional expectation-maximization (EM) algorithm can be used for tracking the channel parameters but as the number of antennas increase, the EM algorithm has to simultaneously track many parameters and this results in slower convergence. We propose a space-alternating generalized expectation-maximization (SAGE) algorithm for tracking channel changes for a time-selective fading channel which has faster convergence than the EM algorithm. We show the bit error rate (BER) performance of the SAGE algorithm when concatenated with a low complexity multi-dimensional parity check (MDPC) code. We further propose an iterative joint "turbo" SAGE-MDPC which further improves the BER performance of the system.