Slab-sphere decoding: efficient maximum-likelihood detection for asymmetric MIMO antenna systems

Maximum-likelihood (ML) detection is of the most importance to yield the minimum erroneous detection for both multiuser detection and space-time decoding. For multiple-input multiple-output (MIMO) antenna systems where the number of receive antennas is at least the number of spatially multiplexed signals, ML detection can be obtained efficiently using spherical detection or commonly referred to as sphere decoding. However, sphere decoding (and its variants) is inherently not applicable for asymmetric (fat) MIMO antenna systems where there is less number of receive antennas than transmitted signals. In this paper, our aim is to extend the idea of sphere decoding to cope with asymmetric MIMO for efficient exact-ML detection. The proposed method is based on the geometric understanding that for asymmetric MIMO systems, the bottom layer of the decoding sphere will be deformed to a decoding slab. By devising an efficient decoding technique to identify all the lattice points that fall inside the slab for a given width, we propose a so-called slab-sphere decoding (which can be viewed as the generalization of sphere decoding for asymmetric MIMO) that guarantees to achieve exact-ML detection while reducing a great deal of complexity.