Learning to Detect

TL;DR

This paper introduces deep neural network architectures for MIMO detection, including a novel DetNet that unfolds gradient descent, achieving state-of-the-art accuracy with low complexity and adaptability to various channels.

Contribution

The paper proposes a new neural network architecture, DetNet, specifically designed for MIMO detection, and demonstrates its effectiveness and flexibility over traditional methods.

Findings

DetNet achieves state-of-the-art detection accuracy.

The proposed networks maintain low computational complexity.

Networks can be trained to handle diverse channel distributions.

Abstract

In this paper we consider Multiple-Input-Multiple-Output (MIMO) detection using deep neural networks. We introduce two different deep architectures: a standard fully connected multi-layer network, and a Detection Network (DetNet) which is specifically designed for the task. The structure of DetNet is obtained by unfolding the iterations of a projected gradient descent algorithm into a network. We compare the accuracy and runtime complexity of the purposed approaches and achieve state-of-the-art performance while maintaining low computational requirements. Furthermore, we manage to train a single network to detect over an entire distribution of channels. Finally, we consider detection with soft outputs and show that the networks can easily be modified to produce soft decisions.