Scholarlink

<p>The 6G network will drive wireless networks toward the Internet of Everything, but they also face the challenge of massive energy consumption. This paper considers reconfigurable intelligent surfaces (RIS)-assisted multi-user multiple-input single-output (MU-MISO) system, aiming to maximize the system’s energy efficiency by jointly optimizing beamforming vectors and RIS phase shifts while satisfying base station quality-of-service requirements and power budget constraints. The resulting optimization problem is inherently non-convex. To address this challenge, we propose an improved Twin Delayed Deep Deterministic Policy Gradient with Prioritized Experience Replay (TD3per) algorithm, enabling the agent to distinguish the importance of experience samples, thereby improving sampling efficiency and reducing training time. Simulation results demonstrate that our TD3per-based scheme achieves higher energy efficiency compared to existing baseline methods.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Energy efficiency optimization in RIS-assisted MU-MISO systems with modified TD3per algorithm

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月14日

查看全文
  • Jinsha Wei,
  • Zhi Lin,
  • Ruiqian Ma,
  • Mengzhao Guo,
  • Zimo Feng,
  • Yong Wang,
  • Lei Wang,
  • Qingsong Zhao

摘要

The 6G network will drive wireless networks toward the Internet of Everything, but they also face the challenge of massive energy consumption. This paper considers reconfigurable intelligent surfaces (RIS)-assisted multi-user multiple-input single-output (MU-MISO) system, aiming to maximize the system’s energy efficiency by jointly optimizing beamforming vectors and RIS phase shifts while satisfying base station quality-of-service requirements and power budget constraints. The resulting optimization problem is inherently non-convex. To address this challenge, we propose an improved Twin Delayed Deep Deterministic Policy Gradient with Prioritized Experience Replay (TD3per) algorithm, enabling the agent to distinguish the importance of experience samples, thereby improving sampling efficiency and reducing training time. Simulation results demonstrate that our TD3per-based scheme achieves higher energy efficiency compared to existing baseline methods.