Adaptive Stochastic Gradient Equalizer Design for Multiuser MIMO System

Summary

The demand for a very high data rate required for various upcoming applications in future 5G generation, such as machine‐to‐machine communications, video streaming, and Internet of Things (IoT), etc., is not possible to meet with the conventional multiple‐input multiple‐output (MIMO) system. To overcome these challenges, promising candidates for the 5G are Multiuser MIMO (MU‐MIMO) and Massive MIMO (M‐MIMO) as they promise to provide both higher throughput and broader coverage. However, these systems require efficient design of precoder and equalizer to improve the overall system performance. Typically, precoders and equalizers are designed via supervised techniques by estimation of the Channel State Information (CSI) at the base station using pilot sequences. However, this increases the overhead of transmission and hence degrades the spectral efficiency. To deal with these issues, we propose in this work to design a semi‐blind adaptive equalizer. This will be based on minimizing the bit error rate (BER) expression via a stochastic gradient approach. For this, we first derive a closed‐form expression for the BER in the MU‐MIMO systems. This is then utilized to develop an iterative adaptive stochastic gradient equalizer. Simulation results are presented to investigate the performance of the proposed equalizer.