An improved detection algorithm based on Lattice reduction for MIMO system

Lattice reduction (LR) is a powerful technique for improving the performance of linear MIMO detection methods. The efficient LR algorithms can largely improve the performance of the linear detectors (LDs). However, there are two problems involved in the LR-aided detection algorithms. One is that, the reduced basis is still not orthogonal, and most existing LR-aided algorithms devote to find a near orthogonal matrix. The other one is that the original constellation is no longer the feasible set of the detected symbol, and the elements of the new symbol vector are correlated. In this paper, an improved LR-aided algorithm (ILRA) considering the feasible set of the new symbol has been proposed. We verify that the feasible set is the integer set in a polyhedron. Eliminating the integer constraint, we simplify the detection to a convex optimization problem. Finally, we get the new detected symbol by solving the convex optimization problem. The simulation results show that our algorithm can achieve significant performance gain over the previous LR-aided LDs, especially in the low-order QAM system.