Signal Detection for MIMO SC-FDMA Systems Exploiting Block Circulant Channel Structure

In this paper, we propose a novel receiver scheme for multiple-input-multiple-output (MIMO) single-carrier frequency-division multiple access (SC-FDMA) systems. Various optimal and near-optimal MIMO detection techniques developed for flat-fading channels are not readily applied to MIMO SC-FDMA systems due to the large dimensions of the effective channels. Recently, the system equation of single-input-single-output (SISO) SC-FDMA systems was divided into a number of disjoint subsystems with moderate dimensions, and a previous MIMO detection technique was applied to the subsystems. In this paper, instead of naively extending the previous scheme for SISO SC-FDMA to MIMO SC-FDMA systems, we first express the large system equation of MIMO SC-FDMA, such that the effective channels are block circulant for an arbitrary number of transmit and receive antennas. The block circulant channel structure is then exploited to lower the computational complexity of projections to construct nonoverlapping subsystems. The proposed circulant projection also takes advantage of the sparse channel structure, offering higher postprojection signal-to-noise ratios (SNRs) of the resulting subsystems than those of a naive projection that does not exploit the channel structure. The simulations confirm the desirable performance of the proposed scheme when a relatively small number of subcarriers are used. The proposed technique is also compared with the previous iterative block decision feedback equalization (IB-DFE) and minimum-mean-square-error-prewhitened-maximum-likelihood (MMSE-prewhitened-ML) detection techniques.