A low complexity ZF-based lattice reduction detection using curtailment parameter in MIMO systems

Abstract

As the large-scale multiple-input multiple-output (MIMO) systems have been advanced in wireless communication, the low complexity and high performance receiver technique have been required. Recently, lattice reduction technique have attracted attention in MIMO detection. Especially, lattice reduction aided linear detection like zero-forcing (ZF) can largely improve the performance with low complexity. Among all the lattice reduction algorithms, the Lenstra-Lenstra-Lovasz (LLL) algorithm is well-known and widely used algorithm. LLL algorithm reduces the basis of channel matrix by using two conditions and it adopts a parameter in order to get nearly orthogonal basis. Although the parameter is defined range, a constant value is commonly used. However it can't be said that the constant value is the best value for all MIMO system. In this paper, we propose an appropriate parameter to reduce computational complexity in LLL algorithm for ZF detection and we also propose an architecture to adapt our proposed parameter in MIMO detection. The proposed parameter is showed simply and corresponds to the number of antennas. Simulation results show that the proposed parameter provides LLL lattice reduction aided ZF detection achieving the same performance as the widely used constant value with lower complexity and also correspond to every MIMO system including large-scale MIMO system.