Detection based on Markov Chain Monte Carlo simulation for MIMO systems

Abstract

In this paper, we present a soft-input soft-output(SISO) detection scheme based on Markov Chain Monte Carlo(MCMC) simulation technique for Multiple-Input-Multiple-Output(MIMO) systems. The MCMC technique uses Metropolis-Hasting algorithm to obtain Bayesian estimates of the transmitted symbols from the received signals and is also suited for coded MIMO systems. We compare this MCMC detector with the optimal MIMO detector using the maximum likelihood(ML) decoding. Numerical results show that the proposed detector can offer a good trade-off between achievable performance and algorithmic complexity.