Xuesong Wang

About

I am a faculty member at Taiyuan University of Technology. Previously, I was a Postdoctoral Researcher at The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen). My work brings together machine learning and communication system design, with the goal of making wireless transmission more efficient, reliable, and secure under practical constraints. My research spans semantic communication for images and video, receiver-driven protocols, cross-layer design, and experimental wireless systems.

• 2026 – PresentFaculty Member · Taiyuan University of Technology
• May 2024 – 2026Postdoctoral Researcher · CUHK-Shenzhen
• 2017 – 2023Ph.D., Computer Applied Technology · University of Chinese Academy of Sciences
• 2013 – 2017B.Eng., Communication Engineering (Elite Program) · Beijing Jiaotong University

Research Interests

My research interests span the following three areas:

• Semantic Communications

  Task-oriented image and video transmission with generative models, adaptive requests, and receiver-driven semantic HARQ.

• Green Communications

  Energy-efficient wireless communication, resource-efficient transmission, and sustainable communication system design.

• Deep Learning for Wireless Systems

  DNN-based physical-layer design and practical implementation, including learning-based transmission and USRP-based prototyping.

Additional interests: KV caching and efficient inference; communication security and privacy; visible-light communications; mobile network analysis and optimization.

News

• Sept. 2026Presented our work on diffusion-aided semantic communication with adaptive requests at IEEE PIMRC 2026.
• Dec. 2025Presented our work on task-oriented semantic communication and model inversion attacks at IEEE GLOBECOM 2025.
• Dec. 2023Presented our work on DNN-based physical-layer network coding at ICCC 2023.
• Jan. 2017Joined the Institute of Network Coding, The Chinese University of Hong Kong, as a Research Assistant.

Journal Articles

1. Diffusion-aided Task-oriented Semantic Communications with Model Inversion Attack
   Xuesong Wang, Mo Li, Xingyan Shi, Zhaoqian Liu, Shenghao Yang*
   IEEE Transactions on Cognitive Communications and Networking, 2026.DOI ↗
2. Implementation of DNN-based Physical-Layer Network Coding
   Xuesong Wang, Lu Lu*
   IEEE Transactions on Vehicular Technology, vol. 72, no. 6, pp. 7380–7394, 2023.DOI ↗
3. DNN-Based Physical-Layer Network Coding for Visible Light Communications
   Xuesong Wang, Runxin Zhang, Xinyan Xie, Lu Lu*
   Photonics, vol. 10, no. 1, p. 23, 2022.DOI ↗PDF ↗

Conference & Proceedings Papers

4. Tube-Structured Incremental Semantic HARQ for Generative Video Receivers
   Xuesong Wang*, Xinyan Xie, Runxin Zhang
   IEEE GLOBECOM, 2026.AcceptedarXiv ↗
5. Diffusion-Aided Bandwidth-Efficient Semantic Communication with Adaptive Requests
   Xuesong Wang*, Xinyan Xie, Mo Li, Zhaoqian Liu
   IEEE PIMRC, 2026.DOI ↗
6. Diffusion-Driven Task-Oriented Semantic Communication Amid Model Inversion Attacks
   Mo Li, Xuesong Wang*, Xingyan Shi, Zhaoqian Liu
   IEEE GLOBECOM, 2025.Co-first authorDOI ↗
7. Adaptive Lighting Control in Visible Light Systems: An Integrated Sensing, Communication, and Illumination Framework
   Xinyan Xie, Xuesong Wang, Xin Lai, Yongheng Wen, Fengrui Yang, Haoyang He, Lai Zhang, Dong Zhao*
   ICCC, 2025.DOI ↗
8. DNN-Based Physical-Layer Network Coding with Low-Complexity Codec
   Xuesong Wang, Wenhao Wang, Xinyan Xie, Qi Wenchao, He Yuna, Lu Lu*
   ICCC, pp. 713–718, 2023.DOI ↗

Patents

• CN111314325B ↗GrantedWang, X.; Zhan, Y.; Zhang, C.; Sun, Y.
• CN111935053B ↗GrantedStudent first inventorZhan, Y.; Wang, X.; Wu, S.; He, J.
• CN115499098B ↗GrantedWang, X.; Qi, W.; He, Y.; Lu, L.

Academic Service

• Technical Program Committee: IEEE ICC 2026; IEEE ICC 2027.
• Journal reviewer: IEEE Transactions on Vehicular Technology (TVT); IEEE Transactions on Cognitive Communications and Networking (TCCN); IEEE Communications Letters; IEEE Wireless Communications Magazine; The Journal of Supercomputing.
• Conference reviewer: IEEE ICC, IEEE GLOBECOM, IEEE WCNC, IEEE PIMRC, IEEE ICCC.