LLL Lattice reduction inversion in wireless MIMO receivers

摘要

In a wireless Multiple Input Multiple Output (MIMO) receiver of Space Time Block Coded (STBC) data via several antennas 27, a Lenstra Lenstra Lovasz (LLL) lattice reduction algorithm in a decoder 30 is improved by calculating a lattice reduction inverse T-1 at each step after an initial condition via combination or swapping of rows and columns. This avoids the computational burden of leaving matrix inversion until the end. Distortion of the QAM constellation (fig. 1), caused by the conversion of the original channel matrix H (estimated from pilot signals) into a more quasi-orthogonal matrix H having shorter base vectors, is corrected by perturbing the quantized detected values ^z by 0, +a or -a to build a table of candidate signals c, and then selecting the best candidate signal x.