Low-complexity detection for multi-antenna differential unitary space-time modulation systems

As known, conventional sphere detection (SD) obtains lower computational complexity compared to the optimal maximum-likelihood (ML) detection. However, its computational complexity is still high especially for low Signal-to-Noise Ratio (SNR) region. Therefore, in this paper, a new low-complexity sphere detection algorithm is proposed for differential unitary space-time modulation systems with multiple antennas. The main idea of the proposed algorithm is to constraint the searching radius of the sphere by a heuristic SNR-dependent factor. The use of this factor provides a much reduced computational complexity and a near-optimal performance, and it keeps the complexity low over the entire range of SNRs. Furthermore, the benefits are expected to be more and more evident and useful with the increasing number of antennas, i.e., massive multiple-input multiple-output (MIMO) systems. Finally, the simulation results further demonstrate the advantages of the proposed low-complexity detection algorithm.