Physical Layer Communications System Design Over-the-Air Using\n Adversarial Networks

2018/03/08 by Timothy J. O’Shea, Tamoghna Roy, O'Shea, Timothy J. +5 · 1 citation
Computer Science · Engineering · #Advanced SAR Imaging Techniques #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Radar Systems and Signal Processing #Signal Processing (eess.SP) #Wireless Signal Modulation Classification #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.1803.03145

openalex publication_date 2018/03/08 · openalex created_date 2022/10/03 · openalex updated_date 2026/07/28

Abstract

This paper presents a novel method for synthesizing new physical layer\nmodulation and coding schemes for communications systems using a learning-based\napproach which does not require an analytic model of the impairments in the\nchannel. It extends prior work published on the channel autoencoder to consider\nthe case where the channel response is not known or can not be easily modeled\nin a closed form analytic expression. By adopting an adversarial approach for\nchannel response approximation and information encoding, we can jointly learn a\ngood solution to both tasks over a wide range of channel environments. We\ndescribe the operation of the proposed adversarial system, share results for\nits training and validation over-the-air, and discuss implications and future\nwork in the area.\n

Cited by

• A Comprehensive Survey on Network Traffic Synthesis: From Statistical Models to Deep Learning

Related

• Deep Learning Models for Physical Layer Communications
• WGAN-based Autoencoder Training Over-the-air
• Learning to Communicate: Channel Auto-encoders, Domain Specific\n Regularizers, and Attention
• An Introduction to Deep Learning for the Physical Layer
• Deep Learning for Wireless Physical Layer: Opportunities and Challenges