Scholarlink

<p>The rapid growth of the Internet of Things (IoT) has increased the need for efficient, adaptive communication networks to handle dynamic data traffic. This paper presents a Quantum Multi-Agent Reinforcement Learning (QMARL) framework for optimizing resource allocation and trajectory planning in Multi-UAV-assisted Mobile Edge Computing (MEC) systems. By combining quantum computing principles with Actor-Critic neural networks, the framework allows UAVs to learn decentralized, adaptive strategies for resource management. Simulations on the Torchquantum platform demonstrate that the QMARL approach outperforms traditional methods in resource utilization, latency, and energy efficiency, achieving optimality with fewer resources. Leveraging quantum advantages like parallelism and entanglement, the framework enables faster convergence and solution exploration.</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

Quantum multi-agent reinforcement learning for efficient resource allocation and trajectory optimization in multi-UAV IoT networks

  • 期刊论文

  • 发表日期：

    2025年12月5日

查看全文
  • R. Kiruthiga,
  • B. Nithya

摘要

The rapid growth of the Internet of Things (IoT) has increased the need for efficient, adaptive communication networks to handle dynamic data traffic. This paper presents a Quantum Multi-Agent Reinforcement Learning (QMARL) framework for optimizing resource allocation and trajectory planning in Multi-UAV-assisted Mobile Edge Computing (MEC) systems. By combining quantum computing principles with Actor-Critic neural networks, the framework allows UAVs to learn decentralized, adaptive strategies for resource management. Simulations on the Torchquantum platform demonstrate that the QMARL approach outperforms traditional methods in resource utilization, latency, and energy efficiency, achieving optimality with fewer resources. Leveraging quantum advantages like parallelism and entanglement, the framework enables faster convergence and solution exploration.