Element-based Lattice Reduction aided K-Best detector for large-scale MIMO systems

Abstract

Recently, large-scale Multiple-Input Multiple-Output (MIMO) systems have caught great attention for increasing the system throughput as well as improving the system performance. The main challenge in the design of these MIMO systems is the detection techniques used at the receiver. Lattice Reduction (LR) techniques have shown good potential in MIMO decoding due to their good performance and low complexity compared to Maximum Likelihood (ML) detector. The Lenstra, Lanstra, and Lovasz (LLL) LR algorithm has been employed for decoding while combined with linear detectors such as ZF as well as with K-Best detection. However, the LLL-aided detectors have shown limited performance, when increasing the number of antennas at the transmitter and receiver. Therefore, in this paper we propose to use the so-called Element-based Lattice Reduction (ELR) combined with K-Best detector for the sake of attaining a better BER performance and lower complexity than the LLL-aided detection. Explicitly, the ELR-aided detectors are capable of attaining a $2$~dB performance improvement at BER of $10^{-5}$ compared to the LLL-aided detectors when considering a MIMO system with $200$ transmit and receive antennas. Furthermore, for the same MIMO configuration, the ELR basis update requires nearly an order of magnitude reduction in the number of arithmetic operations compared to the LLL algorithm.