Approaching Large-Scale MIMO Capacity with Turbo Fixed-Complexity Sphere Detection

11 March 2013

New Image

In this paper we present an MIMO detection algorithm, focusing on large-scale MIMO system. It is well known that the a posteriori Probability (APP) for large-scale MIMO is inapplicable due to exponentially increasing complexity. The classic one-sweep Fixed-complexity Sphere Detection (FSD) allows us to reduce the search space by hard-detecting several reliable streams and detects the rest streams in an APP detection manner. Inspired by the Overlapped Subspace Detection (OSD) concept, we introduce an iterative detection mechanism, referred to as Turbo Fixed-complexity Sphere Detection (Turbo-FSD), to improve the reliability of all detected streams. Numerical results reveal that the large-scale MIMO capacity can be very closely approached. Further, according to the suggested parameter setting, the corresponding complexity is even comparable to that of the linear receivers.