基于边缘计算的节能智能共享优化方法研究  

Research on Energy-efficient Intelligent Sharing Scheme Based on Edge Computing

作　　者：慕欣航 王召巴[1] 谢俊峰 逯丰亮 MU Xin-hang;WANG Zhao-ba;XIE Jun-feng;LU Feng-liang(School of Information and Communication Engineering,North University of China,Taiyuan 030051,China)

机构地区：[1]中北大学信息与通信工程学院,山西太原030051

出　　处：《计算机技术与发展》2025年第3期84-90,共7页Computer Technology and Development

基　　金：山西省自然科学基金(202203021212151)。

摘　　要：随着人工智能(AI)和边缘计算(MEC)技术的发展,使远程智能能够迁移到距用户更近的网络边缘,为用户提供更快捷、更优质的服务。为了提高边缘云服务AI模型的准确性以及边缘智能的训练效率和服务质量,智能网络支持的边缘计算(INEC)正逐渐成为一种有效的解决方案,它使每个边缘节点在共享与学习其他边缘节点的智能下快速地改进自身的AI模型。因此,如何有效地在边缘节点之间进行智能共享是实现边缘计算的关键。考虑到绿色通信的要求,该文研究了一种基于边缘计算的节能智能共享优化方法,该方案在考虑共享模型的性能、智能共享的总能耗与总时延的情况下将其建模为一种非线性整数规划问题。为了解决这一优化问题,采用了一种将遗传算法和粒子群算法相结合的混合算法来选择最优智能共享决策。之后,通过大量的仿真实验,评估了该方案与算法的收敛性和可行性。With the advancement of Artificial Intelligence(AI)and Mobile Edge Computing(MEC),remote intelligence can migrate closer to the network edge,providing users with faster and higher-quality services.To enhance the accuracy of AI models in edge cloud services,as well as the efficiency and quality of edge intelligence training and services,Intelligent Network-supported Edge Computing(INEC)is gradually becoming an effective solution.It enables each edge node to rapidly improve its AI models by sharing and learning intelligence from other edge nodes.Therefore,efficient intelligent sharing among edge nodes is crucial for realizing INEEC.Considering the requirements of green communication,we study an energy saving intelligent sharing optimization method based on edge computing.The scheme models the problem as a nonlinear integer programming problem,taking into account the performance of shared models,total energy consumption,and total delay of intelligent sharing.To address this optimization problem,a hybrid algorithm combining Genetic Algorithm(GA)and Particle Swarm Optimization(PSO)is proposed to select the optimal intelligent sharing decisions.Subsequently,we conducted extensive simulation experiments to evaluate the convergence and superiority of the proposed scheme.

关 键 词：边缘计算 智能共享 边缘智能 绿色通信 遗传算法-粒子群算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]