具有隐私保护的边缘计算高效数据卸载方法  

Efficient data offloading method for edge computing with privacy protection

作　　者：董骏[1] 冯锋 Dong Jun;Feng Feng(Dept.of Information&Intelligent Engineering,Ningxia Vocational College of Finance&Economics,Yinchuan 750021,China;School of Information Engineering,Ningxia University,Yinchuan 750021,China)

机构地区：[1]宁夏财经职业技术学院信息与智能工程系,银川750021 [2]宁夏大学信息工程学院,银川750021

出　　处：《计算机应用研究》2021年第7期2072-2076,共5页Application Research of Computers

基　　金：国家自然科学基金资助项目(71561023);宁夏回族自治区教育厅质量工程—移动应用开发。

摘　　要：针对用户终端数据卸载过程中难以同时实现高隐私安全和低时间消耗的目标,提出了一种具有隐私保护的边缘计算高效数据卸载方法。首先,利用时间计算模型和隐私熵值分别将用户终端时间消耗和数据隐私安全程度进行量化,并建立一个多目标优化问题模型;其次,利用改进强度帕累托进化算法对时间消耗和隐私熵值进行联合优化;最后,利用基于熵权法的多属性决策方法选取最优的时间消耗和隐私熵组合策略。在多终端用户多计算任务的边缘计算下展开实验研究和对比分析,结果表明,该方法在降低传输时间的基础上还增强了数据卸载传输的安全性。In order to achieve the goal of high privacy security and low time consumption in the process of user terminal data offloading,this paper proposed an efficient data offloading method with privacy protection for edge computing.Firstly,it quantified the user terminal time consumption and data privacy security by time calculation model and privacy entropy respectively,and established a multi-objective optimization problem model.Secondly,it jointly optimized the time consumption and privacy entropy by improved strength Pareto evolutionary algorithm.Finally,it selected the combination strategy of optimal time consumption and privacy entropy by multi attribute decision making method based on entropy weight method.Experimental research and comparative analysis were carried out under the edge computing of multi terminal users and multi computing tasks.The results show that the proposed method not only reduces the transmission time,but also enhances the security of data offloading transmission.

关 键 词：边缘计算 数据卸载 隐私保护 时间消耗 隐私熵 多目标优化算法 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]