Decoding With Hypothesis Testing: A Near ML Decoding Scheme for MIMO Systems

Abstract

A near maximum-likelihood (NML) scheme for the decoding of multiple-input-multiple-output (MIMO) systems is addressed by incorporating the technique of hypothesis testing in the searching procedure. The proposed decoding scheme selects the best node based on the node metric, determines one child node of the best node via hypothesis testing, and connects the best node with some sibling nodes of the child node. From simulation results, it is confirmed that the proposed scheme has a lower computational complexity than other NML decoders and that the performance difference between the proposed and maximum-likelihood schemes is negligibly small.