Efficient lattice-reduction-aided conditional detection for MIMO systems

Two near-optimal, low-complexity latticereduction- aided (LRA) conditional detectors are proposed for multiple-input multiple-output (MIMO) systems. The reduction is performed only on a selection of columns of the channel matrix and conditional optimization is performed on the remaining ones. In the proposed schemes, the best submatrix for conditional detection is selected by considering all possible submatrices and choosing the one that gives the best metric. For quasi-static channels, where the cost of lattice reduction can be negligible over the whole packet over which the channel is constant, the complexity of the proposed schemes is only linear in the size of the QAM modulation used. The near-optimal error performances of the proposed schemes are verified by analysis and simulations.