Institute of Information Science, Academia Sinica



Press Ctrl+P to print from browser

TIGP (SNHCC) -- Deep Learning-based Precoding Design for RIS-aided MIMO Systems with Practical Reflectivity Consideration


TIGP (SNHCC) -- Deep Learning-based Precoding Design for RIS-aided MIMO Systems with Practical Reflectivity Consideration

  • LecturerDr. Po-Heng Chou (Research Center for Information Technology Innovation, Academia Sinica)
    Host: TIGP (SNHCC)
  • Time2023-03-06 (Mon.) 14:00 – 16:00
  • LocationAuditorium 106 at IIS new Building
This talk is about deep learning-based precoding design for the Reflecting Intelligent Surface (RIS) assisted Multiple-Input Multiple-Output (MIMO) system that is potential B5G/6G technology. We consider (1) Single RIS and Single User and (2) Multi-RIS and Multi-User two system models to maximize the spectral efficiency of the transmitter precoding matrix (active beamforming) and RIS phase offset matrix (passive beamforming). The major difference between our work and previous IRS system optimization works is that we consider the phase shift interval of the reflecting element and its corresponding reflectivity under practical scenarios. However, practical considerations increase the difficulty of system optimization. For the above two models, (a) Supervised Learning Deep Neural Networks (DNNs) and (b) Reinforcement Learning Deep Deterministic Policy Gradients (DDPGs) are proposed as deep learning solutions. The computational complexity is much lower than the traditional optimal algorithm, and the performance is very close to the optimal solution.
Dr. Po-Heng Chou was born in Tainan, Taiwan, in 1987. He received the B.S. degree in electronic engineering from National Formosa University, Yunlin, Taiwan, in 2009, the M.S. degree in communications engineering from National Sun Yat-sen University, Kaohsiung, Taiwan, in 2011, and the Ph.D. degree directed by Prof. Char-Dir Chung (IEEE Fellow), from the Graduate Institute of Communication Engineering (GICE) of National Taiwan University (NTU), Taipei, Taiwan, in 2020. He is currently a Postdoctoral Fellow hosted by Prof. Yu Tsao at the Research Center for Information Technology Innovation (CITI), Academia Sinica, Taipei, Taiwan. His research interests include Deep Learning-based Signal Processing, Wireless Communications and Networks.
Dr. Chou was awarded the Outstanding University Youth Award and Phi Tau Phi Honorary Membership by NTU in 2019 to honor his impressive academic achievement. Dr. Chou has been elected as the Distinguished Regular Postdoctoral Scholar of CITI by Academia Sinica during 2022-2023.