Scholarlink

<p>With the explosive growth of mobile data traffic, the traditional macrocell-based mobile communication network architecture is difficult to meet the demand, and Ultra-dense Edge Computing (UDEC), which is a deep fusion of Mobile Edge Computing (MEC) and Ultra-dense Network (UDN), has become one of the innovative changes to address the challenge. Edge Computing (UDEC) and become one of the innovative changes to solve the challenge. The limited computing power of traditional edge cloud can no longer meet the real-time changing computing needs of users, so the problem of Power Allocation (PA) and Joint Request Offloading and Resource Scheduling (JRORS) for mobile users are proposed. problem. In the PA problem, energy consumption is minimized by employing convex techniques, and in the JRORS problem, energy efficiency is maximized by establishing a mathematical model of mixed integer nonlinear programming problem with the objective of minimizing the request response delay. On this basis, an improved mayfly algorithm (ataMA) is proposed to solve the edge computing task offloading and scheduling optimization problem. ataMA firstly improves the original mayfly algorithm’s flight approach to improve the search accuracy by introducing a finer local search mechanism, and enhances the global search capability by increasing the random flight range. Then the variation rate of the original mayfly algorithm is improved, and the appropriate variation rate adjustment strategy accelerates the convergence speed of the algorithm while keeping the algorithm alive during the search process, which improves the robustness of the algorithm. Simulation experiments are conducted to compare the ataMA with the classical MVO, GA, ALO, AOA, and the latest RIME, SOA,ASMO, SO, CDO and SHO scheduling methods for nine different groups of mobile users, respectively, with energy consumption and energy efficiency as the objective functions, and the experimental results show that the proposed ataMA has a significant advantage in terms of reducing the energy consumption and increasing energy efficiency, which always maintains a high performance state in dynamic UDEC networks.</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

Edge computing task unloading and scheduling optimization by mayfly algorithm with improved flight damping ratio and mutation rate

  • 期刊论文

  • 发表日期：

    2026年6月17日

查看全文
  • Xiao-Fei Sui,
  • Jie-Sheng Wang,
  • Si-Wen Zhang,
  • Yun-Hao Zhang,
  • Shi-Hui Zhang,
  • Xue-Lian Bai

摘要

With the explosive growth of mobile data traffic, the traditional macrocell-based mobile communication network architecture is difficult to meet the demand, and Ultra-dense Edge Computing (UDEC), which is a deep fusion of Mobile Edge Computing (MEC) and Ultra-dense Network (UDN), has become one of the innovative changes to address the challenge. Edge Computing (UDEC) and become one of the innovative changes to solve the challenge. The limited computing power of traditional edge cloud can no longer meet the real-time changing computing needs of users, so the problem of Power Allocation (PA) and Joint Request Offloading and Resource Scheduling (JRORS) for mobile users are proposed. problem. In the PA problem, energy consumption is minimized by employing convex techniques, and in the JRORS problem, energy efficiency is maximized by establishing a mathematical model of mixed integer nonlinear programming problem with the objective of minimizing the request response delay. On this basis, an improved mayfly algorithm (ataMA) is proposed to solve the edge computing task offloading and scheduling optimization problem. ataMA firstly improves the original mayfly algorithm’s flight approach to improve the search accuracy by introducing a finer local search mechanism, and enhances the global search capability by increasing the random flight range. Then the variation rate of the original mayfly algorithm is improved, and the appropriate variation rate adjustment strategy accelerates the convergence speed of the algorithm while keeping the algorithm alive during the search process, which improves the robustness of the algorithm. Simulation experiments are conducted to compare the ataMA with the classical MVO, GA, ALO, AOA, and the latest RIME, SOA,ASMO, SO, CDO and SHO scheduling methods for nine different groups of mobile users, respectively, with energy consumption and energy efficiency as the objective functions, and the experimental results show that the proposed ataMA has a significant advantage in terms of reducing the energy consumption and increasing energy efficiency, which always maintains a high performance state in dynamic UDEC networks.