Sparse Bayesian Learning-Aided Joint Sparse Channel Estimation and ML Sequence Detection in Space-Time Trellis Coded MIMO-OFDM Systems

Abstract

Sparse Bayesian learning (SBL)-based approximately sparse channel estimation schemes are conceived for space-time trellis coded (STTC) multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) systems relying on trellis-based encoding and decoding over the data subcarriers. First, a pilot-aided channel estimation scheme is developed employing the multiple response extension of SBL (MSBL) framework. Subsequently, a novel data-aided joint channel estimation and data decoding framework relying on optimal maximum likelihood sequence detection (MLSD) is intrinsically amalgamated with our powerful EM-based MSBL algorithm. Explicitly, an MSBL-based MIMO channel estimate is gleaned in the E-step followed by a novel modified path-metric-based Viterbi decoder in the M-step. Our theoretical analysis characterizes the performance of the proposed schemes in terms of the associated frame error rate (FER) upper bounds by explicitly considering the effect of estimation errors along with evaluating the product measure of the STTC under consideration. Finally, our simulation results are complemented by the Bayesian Cramer-Rao bound (BCRB), the associated complexity analysis and the performance of the proposed schemes for validating the theoretical bounds.