Spectrally Efficient CSI Acquisition Approach for Large-Scale MIMO Systems

Abstract

Large-scale multiple input multiple output (MIMO) has a high spectral and energy efficiency and is considered as the promising technique for future 5G wireless communications. In practice, the accurate knowledge of channel state information (CSI) is of great importance to guarantee the performance of MIMO systems. However, the conventional training sequence (TS) based approaches usually suffer from the spectral efficiency degradation problems in large-scale MIMO systems. In this paper, based on the theory of structured compressive sensing (SCS), we propose a spectrally efficient CSI acquisition approach from the small dimensional inter block interference free (IBI-free) region within the received TS. By exploiting the spatial-temporal correlations of the sparse MIMO channels, the proposed spatially-temporally joint compressive sampling matching pursuit (STJ-CoSAMP) algorithm can have better recovery performance with less observation than the classical structured CoSAMP algorithm. Furthermore, a $32\times32$ MIMO system is considered to evaluate the proposed scheme, and simulation results show that it has better performance and higher spectral efficiency than the conventional MIMO schemes.