CTH09-4: A Lattice-Reduction-Aided Soft Detector for Multiple-Input Multiple-Output Channels

Lattice basis reduction is a powerful technique that enables a hard-output detector for a multiple-input multiple-output channel to approach maximum-likelihood performance with low complexity. In this work we propose a soft-output detector that combines lattice- reduction-aided detection and list decoding. The proposed algorithm performs nearly as well as the list-sphere detector but with much lower complexity. Numerical results reveal that the complexity per bit for the proposed algorithm decreases as the size of the QAM alphabet increases.