Near-Capacity Iteratively Decoded Markov-Chain Monte-Carlo Aided BLAST System

摘要

In this treatise, we propose an iteratively decoded Bell-labs LAyered Space-Time (BLAST) scheme, which serially concatenates an IRregular Convolutional Code (IRCC), a UnityRate Code (URC) and a BLAST transmitter. The proposed scheme is capable of achieving a near capacity performance with the aid of our Extrinsic Information Transfer (EXIT) chart assisted design procedure. Furthermore, a Markov Chain Monte Carlo (MCMC) based BLAST scheme is employed, which is capable of significantly reducing the complexity imposed. For the sake of approaching the maximum achievable rate, iterative decoding is invoked to attain decoding convergence by exchanging extrinsic information among the three serial component decoders. Our simulation results show that the proposed MCMC-based iteratively detected IRCC-URC-BLAST scheme is capable of approaching the system capacity.