Trainable Communication Systems: Concepts and Prototype

TL;DR

This paper introduces a neural network-based trainable communication system that optimizes transmission and decoding processes, demonstrating significant performance improvements through joint training, code design, and real-world SDR implementation.

Contribution

It presents a fully differentiable neural iterative decoding structure and demonstrates end-to-end training on actual wireless channels, enabling adaptable and improved communication performance.

Findings

Significant gains over conventional methods on AWGN channels.

Effective integration with practical bit-metric decoding receivers.

Successful implementation and training on software-defined radios.

Abstract

We consider a trainable point-to-point communication system, where both transmitter and receiver are implemented as neural networks (NNs), and demonstrate that training on the bit-wise mutual information (BMI) allows seamless integration with practical bit-metric decoding (BMD) receivers, as well as joint optimization of constellation shaping and labeling. Moreover, we present a fully differentiable neural iterative demapping and decoding (IDD) structure which achieves significant gains on additive white Gaussian noise (AWGN) channels using a standard 802.11n low-density parity-check (LDPC) code. The strength of this approach is that it can be applied to arbitrary channels without any modifications. Going one step further, we show that careful code design can lead to further performance improvements. Lastly, we show the viability of the proposed system through implementation on software-defined radios (SDRs) and training of the end-to-end system on the actual wireless channel. Experimental results reveal that the proposed method enables significant gains compared to conventional techniques.