IEEE International Symposium on Personal, Indoor and Mobile Radio Communications
13–16 September 2021 // Virtual Conference // By 6G Flagship

Deep Learning for Wireless Communications


Dr. Geoffrey Ye Li
Imperial College London, UK

Dr. Zhijin Qin
Queen Mary University of London, UK


It has been demonstrated recently that deep learning (DL) has great potentials to break the bottleneck of the conventional communication systems. In the tutorial, we will provide a comprehensive overview on DL for wireless communications, including physical layer processing, resource allocation and semantic communications.

DL can improve the performance of each individual (traditional) block in the conventional communication systems or jointly optimize the whole transmitter or receiver. Therefore, we can categorize DL enabled physical layer communications into with and without block processing structures. For DL based communication systems with block structures, we introduce joint channel estimation and signal detection based on a fully connected deep neural network, model-drive DL, and some experimental results. For those without block structures, we provide recent endeavors in developing end-to-end learning communication systems.

The traditional wisdom is to explicitly formulate resource allocation as anoptimization problem. However, as wireless networks become increasingly diverse and complex, such as high-mobility vehicular networks, the current design methodologies face significant. Meanwhile, DL represents a promising alternative due to its remarkable power to leverage data for problem solving. We will first introduce how to use DL to solve optimization problems for resource allocation, and then discuss deep reinforcement learning for resource allocation in vehicular networks.

In the past decades, communications primarily focus on how to accurately and effectively transmit symbols. The achieved transmission rate has been improved tens of thousands of times than before and the system capacity is gradually approaching to the Shannon limit. Semantic communications have been regarded as a promising direction, which aims to exchange semantic information rather than receiving the transmitted bit sequences or symbols accurately. We will first introduce the concept of the semantic communication and highlight its key difference from typical communications. Then we will present the initial work on semantic communications.

Structure and content

An Outline of the Tutorial Content, including Its Tentative Schedule

  • Introduction: Why Deep Learning? (5 minutes)
  • DL for Physical Layer Communications (60 minutes)
    • General Introduction
    • Model-Driven DL for Physical Layer Communications
    • DL for End-to-End Wireless Systems
    • Remarks and Questions-and-Answers
  • DL for Wireless Resource Allocation (60 minutes)
    • Introduction
    • Deep Learning assisted Optimization
    • Deep Reinforcement Learning based Approached
    • Remarks and Questions-and-Answers
  • Semantic Communications (45 minutes)
    • Introduction
    • Semantic Communications Principles and Metrics
    • Deep Learning enabled Semantic Communications
    • Remarks and Questions-and-Answers
  • Conclusions (10 minutes)