Soft Log Likelihood Ratio Replacement for Low-Complexity Maximum-Likelihood Detection

For multiple-input multiple-output systems, spatial multiplexing techniques can provide significant throughput gain. However, this promised gain can be attained by using maximum-likelihood detection (MLD) which requires huge computational complexity, and thus several low-complexity schemes are proposed to reduce the search space of the MLD. In this letter, we consider the log likelihood ratio (LLR) computation in low complexity MLD for soft channel decoding. We design a weighted combining method for a reliable LLR computation, compensating for the performance loss incurred by insufficient candidates of low-complexity MLD. The proposed LLR generation algorithm effectively reduces the computational complexity while maintaining the system performance of spatial multiplexing systems.

I. Introduction

The multiple-input multiple-output (MIMO) system is now widely adopted as a key element for broadband wireless applications because of its potential for providing increased channel capacity and robustness against multipath fading channels [1]–[4]. For instance, various MIMO schemes are supported by current standards such as 3GPP Long Term Evolution Advanced (LTE-A) [5], [6] and IEEE 802.16m [7] to meet the requirements of next generation communications. Among MIMO systems, spatial multiplexing is of interest as it can theoretically provide significant throughput gain that is proportional to the number of antennas. Currently, spatial multiplexing techniques are evolving into multiuser and multicell MIMO systems [8].