Evaluation of reduced search iterative MIMO detection schemes

Abstract

This paper presents a complexity reduction method for iterative MIMO detection scheme. We first investigate the statistical characteristics of the log-likelihood ratios (LLRs) across the MIMO detector and decoder, and set a threshold value for the reduced search MIMO detector. Based on this threshold value, the decision is made whether re-estimation of MIMO detection using the soft information provided by the decoder is necessary. This is to eliminate unnecessary repetitive calculation for symbols with high reliability. As iteration goes by, complexity is largely reduced. We compare the performance of the proposed scheme by using maximum likelihood (ML) or maximum a posteriori (MAP). Simulation results show that the proposed complexity reduction method reduces the complexity by about 30% with negligible bit error rate (BER) performance degradation.