A Message Passing Based Receiver for Extra-Large Scale MIMO

TL;DR

This paper introduces a low-complexity, scalable message passing receiver for extra-large scale MIMO systems that adapts to channel non-stationarities and outperforms traditional zero-forcing methods.

Contribution

It proposes a Variational Message Passing based receiver for XL-MIMO that scales efficiently and handles channel non-stationarities, improving performance and reducing complexity.

Findings

Significant performance gains over zero-forcing receiver

Complexity scales almost linearly with antennas and users

Effective adaptation to channel non-stationarities

Abstract

We consider a massive MIMO system where the array at the access point reaches a dimension that is much larger than the array in current systems. Transitioning to an extremely large dimension and hence large number of antennas implies a need to scale up the multi-antenna processing while maintaining a reasonable computational complexity. In this paper, we study the receiver of such an extra-large scale MIMO (XL-MIMO) system. We propose to base the reception on Variational Message Passing (VMP). The motivation is that the complexity of VMP scales (almost) linearly with the number of antennas and number of users, hence enabling low-complexity reception in crowd scenarios. Furthermore, VMP adapts to the non-stationarities of the MIMO channel that appear due to the large dimension of the array. Through numerical results, we show significant performance improvement and computational complexity reduction compared to a zero-forcing receiver.