Long-Term CSI-based Design for RIS-Aided Multiuser MISO Systems Exploiting Deep Reinforcement Learning

Hong Ren, Cunhua Pan, Liang Wang, Wang Liu, Zhoubin Kou, Kezhi Wang

Research output: Contribution to journalArticlepeer-review

13 Citations (Scopus)
39 Downloads (Pure)

Abstract

In this paper, we study the transmission design for reconfigurable intelligent surface (RIS)-aided multiuser communication networks. Different from most of the existing contributions, we consider long-term CSI-based transmission design, where both the beamforming vectors at the base station (BS) and the phase shifts at the RIS are designed based on long-term CSI, which can significantly reduce the channel estimation overhead. Due to the lack of explicit ergodic data rate expression, we propose a novel deep deterministic policy gradient (DDPG) based algorithm to solve the optimization problem, which was trained by using the channel vectors generated in an offline manner. Simulation results demonstrate that the achievable net throughput is higher than that achieved by the conventional instantaneous-CSI based scheme when taking the channel estimation overhead into account.

Original languageEnglish
Pages (from-to)567-571
Number of pages5
JournalIEEE Communications Letters
Volume26
Issue number3
Early online date4 Jan 2022
DOIs
Publication statusPublished - 10 Mar 2022

Keywords

  • Array signal processing
  • Channel estimation
  • Coherence time
  • deep reinforcement learning
  • intelligent reflecting surface (IRS)
  • Interference
  • Precoding
  • Reconfigurable intelligent surface (RIS)
  • Rician channels
  • Signal to noise ratio

Fingerprint

Dive into the research topics of 'Long-Term CSI-based Design for RIS-Aided Multiuser MISO Systems Exploiting Deep Reinforcement Learning'. Together they form a unique fingerprint.

Cite this