EM-Based Channel Estimation Algorithms for MIMO Turbo Receivers

Multiple-input multiple-output (MIMO) systems combined with orthogonal frequency division multiplexing (OFDM), can improve communication quality and capacity. In this paper, we consider channel estimation issue in a MIMO-OFDM system using low-density parity-check (LDPC) codes at transmitter and iterative detection based on a maximum a posteriori (MAP) demodulator and a soft LDPC decoder at receiver. We first consider a previously proposed solution based on least-square estimator (USE). Since matrix inversion is required in LSE algorithms, their computational cost for implementation is high. Then, we resort to the expectation-maximization (EM)-type algorithms, which significantly reduce the complexity of the channel estimation, and propose a modified turbo receiver for LDPC-coded MIMO-OFDM systems which performs the channel estimation by space alternating generalized EM (SAGE) algorithm. Based on simulation results, the performance of the proposed receiver is quite satisfying.