An improved lattice reduction aided detection scheme in MIMO systems

Abstract

Multiple input multiple output(MIMO) system has been widely researched in recent years. To match high-speed data stream, detector in receiver becomes especially important. Exiting detectors, such as zero-forcing(ZF), minimum mean square error(MMSE) and maximum likelihood(ML) can not meet the demand. So lattice reduction aided detection(LRAD) has been introduced as a new type of MIMO detector. The key point of LRAD is to transform the system model into an equivalent one with better-conditioned channel matrix. As a result, LRAD can significantly improve bit error rate(BER) performance with small additional complexity in MIMO systems. But at the same time, it brings a lot of problems, such as detection signal constellation distortion and quantization error diffusion. In particular, quantization operation in LRAD is suboptimal, because the quantization operation will cause quantization error. In order to meet this challenge, this paper proposes an improved LRAD algorithm based on quantization error correction using noise prediction. The proposed algorithm correct quantization error by generating a list of candidate values from the original LRAD detected estimated values. Then the final better estimates are searched by ML criterion. Simulation results show that improved performance can be obtained, whereas additional complexity is reasonable small. The performance of the improved LRAD algorithm is shown to improve at least 0.8dB compared with conventional algorithms.