Empowering wireless communications and sensing with deep learning technology

Abstract

In recent years, deep learning (DL) technologies have witnessed dramatic progress due to their nonlinearity. Deep learning has brought many breakthroughs in various fields, such as computer vision, natural language processing and speech recognition, which motivate researchers from other fields to explore the possibility of adopting deep learning techniques. Many efforts have been made and much progress has been witnessed in bioinformatics, medicine, material science, civil engineering, etc. The computer network and communications field as well. Both physical layers like coding and modulation schemes and upper layers like communication network deployment report remarkable progress. Since it is in the early stage, there are still many issues to be solved and there remains huge potential. Specifically, this thesis explores the feasibility of using deep learning techniques to enhance next-generation communication efficiency and broaden the ubiquitous radio frequency (RF) sensing boundary.