A new lattice decoder for multiple antenna systems

Sphere decoding for multiple input multiple output (MIMO) channel systems has been shown to achieve near-maximum likelihood (ML) performance with low complexity. In this paper, we present an algorithm which is intended to explore the lattice sphere packing representation of a multi-antenna system and the space-time codes. We apply a new sphere decoding algorithm to the resulting lattice code. For the uncoded system sphere decoding yields, with a huge decrease in complexity compared to the optimum maximum likelihood sequence estimator (MLSE) detection algorithm, a small decrease in the bit error rate (BER) performance. The fact that the sphere decoding algorithm does not depend on the constellation size gives rise to systems with very high spectral efficiency, maximum likelihood performance and low decoding complexity. In our simulations, we use both the 16-QAM and 64-QAM constellations. This work intends to improve the multi-antenna system performance and was developed for the simulation of enhanced UMTS access and core networks (SEACORN) project.