DeepRx MIMO: Convolutional MIMO Detection with Learned Multiplicative Transformations

TL;DR

This paper introduces DeepRx, a deep learning-based MIMO receiver architecture that uses convolutional neural networks and novel transformation layers, significantly improving performance over traditional methods especially with sparse pilot data.

Contribution

It presents a novel deep learning MIMO receiver with transformation layers, demonstrating high performance with sparse pilot configurations, a first in the field.

Findings

DeepRx outperforms conventional receivers.

Transformation layers improve performance with sparse pilots.

Fully learned transformations achieve high accuracy.

Abstract

Recently, deep learning has been proposed as a potential technique for improving the physical layer performance of radio receivers. Despite the large amount of encouraging results, most works have not considered spatial multiplexing in the context of multiple-input and multiple-output (MIMO) receivers. In this paper, we present a deep learning-based MIMO receiver architecture that consists of a ResNet-based convolutional neural network, also known as DeepRx, combined with a so-called transformation layer, all trained together. We propose two novel alternatives for the transformation layer: a maximal ratio combining-based transformation, or a fully learned transformation. The former relies more on expert knowledge, while the latter utilizes learned multiplicative layers. Both proposed transformation layers are shown to clearly outperform the conventional baseline receiver, especially with sparse pilot configurations. To the best of our knowledge, these are some of the first results showing such high performance for a fully learned MIMO receiver.