Summary of Multi-agent Deep Reinforcement Learning For Energy Efficient Multi-hop Star-ris-assisted Transmissions, by Pei-hsiang Liao et al.
Multi-Agent Deep Reinforcement Learning for Energy Efficient Multi-Hop STAR-RIS-Assisted Transmissions
by Pei-Hsiang Liao, Li-Hsiang Shen, Po-Chen Wu, Kai-Ten Feng
First submitted to arxiv on: 26 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Signal Processing (eess.SP)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel architecture for wireless communications called Multi-Hop Simultaneously Transmitting and Reflecting Reconfigurable Intelligent Surface (STAR-RIS) to achieve wider coverage. Building upon existing research on single STAR-RIS, the authors integrate multi-hop transmissions to overcome limitations. The proposed architecture combines active beamforming of base stations with passive beamforming of STAR-RISs to maximize energy efficiency constrained by hardware limitations. Additionally, the impact of STAR-RIS element on-off states on energy efficiency is investigated. To tackle this complex problem, a Multi-Agent Global and Local deep Reinforcement learning (MAGAR) algorithm is designed, which elevates collaboration among local agents. Numerical results show significant improvement compared to benchmarks like Q-learning, multi-agent deep Q-network with global reward, and multi-agent DQN with local rewards. Moreover, the proposed architecture achieves higher energy efficiency than mode switching based STAR-RISs, conventional RISs, and deployment without RISs or STAR-RISs. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research paper is about improving wireless communication coverage by using a new type of technology called Reconfigurable Intelligent Surfaces (RIS). Instead of just one RIS, the authors suggest using multiple ones to cover a wider area. They also propose a way to make these RISs work more efficiently together. To solve this complex problem, they developed an algorithm that helps different parts of the system communicate with each other effectively. The results show that their approach works better than previous methods and can provide longer-lasting coverage. |
Keywords
» Artificial intelligence » Reinforcement learning
