Near maximum likelihood detection schemes for wireless MIMO systems

In this letter, we propose two near maximum-likelihood detection (MLD) schemes for multiple input-multiple output antenna systems over flat fading channels. The schemes are based on utilizing an initial estimate of the received symbol and improving it by selectively updating various bits within the symbol. Simulation results show that the performance of the proposed schemes can approach that of MLD but with a significant reduction in complexity.

I. Introduction

Recently, the application of multiple input–multiple output (MIMO) antenna systems and space-time signal processing techniques to wireless communication systems has become an important issue [1], [2]. By employing multiple antennas, one can increase data rates significantly as well as improve performance. Possible detection techniques include zero-forcing (ZF), minimum mean-square error (MMSE) detection, maximum-likelihood detection (MLD) [3] and the Vertical Bell Laboratories Layered Space Time (V-BLAST) detection. Among these schemes, MLD is optimal but its complexity is exponential in the number of transmit antennas which makes it unsuitable when the number of transmit antennas is large. As a result, some effort has gone into developing suboptimum MLD schemes that provide similar performance but with less complexity.