End-to-End Learning of Communications Systems Without a Channel Model

2018/04/06 by Aoudia, Fayçal Ait, Hoydis, Jakob · 4 citations
#Artificial Intelligence (cs.AI) #FOS: Computer and information sciences #Information Theory (cs.IT) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1804.02276

Abstract

The idea of end-to-end learning of communications systems through neural network -based autoencoders has the shortcoming that it requires a differentiable channel model. We present in this paper a novel learning algorithm which alleviates this problem. The algorithm iterates between supervised training of the receiver and reinforcement learning -based training of the transmitter. We demonstrate that this approach works as well as fully supervised methods on additive white Gaussian noise (AWGN) and Rayleigh block-fading (RBF) channels. Surprisingly, while our method converges slower on AWGN channels than supervised training, it converges faster on RBF channels. Our results are a first step towards learning of communications systems over any type of channel without prior assumptions.

Cited by

• CoNet-Rx: Collaborative Neural Networks for OFDM Receivers
• Encoding Optimization for Low-Complexity Spiking Neural Network Equalizers in IM/DD Systems
• A New Pathway to Integrated Learning and Communication (ILAC): Large AI Model and Hyperdimensional Computing for Communication
• A Multi-Task Foundation Model for Wireless Channel Representation Using Contrastive and Masked Autoencoder Learning

Related

• Model-free Training of End-to-end Communication Systems
• Deep Reinforcement Learning Autoencoder with Noisy Feedback
• Channel model for end-to-end learning of communications systems: A survey
• WGAN-based Autoencoder Training Over-the-air
• End-to-end Learning for OFDM: From Neural Receivers to Pilotless\n Communication