Scholarlink

With the rapid growth of multimedia applications, the computational demands of mobile computing systems have increased, making it challenging for traditional offloading methods to handle the complexity of such services. This paper proposes an enhanced Deep Deterministic Policy Gradient (DDPG)-based algorithm, E-DDPG, for UAV-assisted task offloading to improve Quality of Service (QoS) and resource utilization. By incorporating state normalization and prioritized experience replay, the proposed method achieves greater stability and training efficiency. Experimental results demonstrate that the E-DDPG algorithm outperforms baseline algorithms like DDPG in terms of model convergence speed and task processing delay.

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

E-DDPG: An Enhanced DDPG-Based Approach for UAV Task Offloading Optimization

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ziqi Zhou,
  • Qing-An Zeng,
  • Zhenjiang Zhang

摘要

With the rapid growth of multimedia applications, the computational demands of mobile computing systems have increased, making it challenging for traditional offloading methods to handle the complexity of such services. This paper proposes an enhanced Deep Deterministic Policy Gradient (DDPG)-based algorithm, E-DDPG, for UAV-assisted task offloading to improve Quality of Service (QoS) and resource utilization. By incorporating state normalization and prioritized experience replay, the proposed method achieves greater stability and training efficiency. Experimental results demonstrate that the E-DDPG algorithm outperforms baseline algorithms like DDPG in terms of model convergence speed and task processing delay.