Error-Rate Improvement via SVD-Based Reduced-Stream Transmission for MIMO and Large-Scale MIMO Systems

In this paper, we propose and discuss a new spatial-multiplexing scheme designed to better exploit the potential of multiple-input multiple-output (MIMO) channels for improved error- rate performance. Specifically, we utilize singular value decomposition (SVD) to select only the “good“ parts of the channel for transmission. The modulation order is increased to maintain the same transmission rate and ensure a fair compari-son. Our findings are then extended to large-scale MIMO systems, where both the precoder and detector are constructed using the hybrid analog/digital architecture to reduce the implementation complexity. Simulation results demonstrate that, given a specific transmission rate, our proposed methods generally outperform conventional spatial-multiplexing approaches regarding the over-all error-rate performance.