摘要
针对当前隐私保护数据聚合研究方案在计算成本、通信开销、数据完整性验证以及灵活性等方面存在的不足,本文提出了一种雾辅助的轻量级隐私保护数据多级聚合方案(Fog-assisted Lightweight Privacy-preserving Data Multilevel Aggregation,F-LPDM A).该方案利用云雾协作的多级聚合模型使中间层次的雾节点能够定期从连接的智能电表处收集数据,并导出细粒度的雾级聚合结果,该细粒度聚合可有效节省通信开销,提高聚合方案的灵活性.同时,为了提高多级聚合的效率,利用模数的性质对Paillier加密算法进行优化使计算成本降低,而多级聚合的结果呈现为一元多项式系数即为细粒度电量的消耗量,随后利用霍纳法则对该聚合结果进行高速解析,以此提高智能电网的效率.再者,借助散列函数的单向性在网络边缘和云端实现轻量级认证;通过分析发现F-LPDMA方案可以实现数据的机密性和隐私保护,确保雾设备和云中心在整个聚合过程中不能获知用户的私人信息.最后,本文在计算成本和通信开销方面对F-LPDMA方案进行了评估,并通过与现有聚合方案的比较分析表明F-LPDM A方案具有更低的计算和通信开销.
According to shortcomings of current privacy-preserving data aggregation schemes in the computation cost,communication overhead,data integrity verification and flexibility,this paper presents a Fog-assisted Lightweight Privacy-preserving Data Multilevel Aggregation Scheme for Smart Grid named F-LPDMA.This scheme utilizes the multi-level aggregation model of cloud-fog collaboration to enable intermediate-level fog nodes to collect data from the connected smart meters periodically and derive fine-grained fog-level aggregation results to save communication overhead and improve the flexibility of smart grid.Meanwhile,to improve the efficiency of multi-level aggregation,F-LPDMA explores properties of the modulus to optimize Paillier encryption algorithm for reducing the computation cost.The result of multi-level aggregation is an unary polynomial that the coefficient of the polynomial is the fine-grained power consumption,and then using Horner’s rule to achieve high-speed parsing of aggregated plaintext that obtain fine-grained aggregation results to improve the efficiency of smart grid.Moreover,it realizes the lightweight authentication at the edge of the network and the cloud by means of the unidirectionality of the hash function.The analysis show s that the proposed F-LPDMA scheme can achieve data confidentiality and privacy preservation,ensures the fog device and cloud center are insensible to the private information during the entire aggregation process.Finally,this paper is evaluated in terms of computation cost and communication overhead,the comparative analyses of our proposed scheme with existing aggregation schemes indicate that the F-LPDMA scheme has less computation and communication overheads.
作者
杨丽
陈思光
YANG Li;CHEN Si-guang(Jiangsu Engineering Research Center of Communications and Network Technology,Nanjing University of Posts and Telecommunications,Nanjing 210003,China;Jiangsu Key Lab of Broadband Wireless Communication and Internet of Things,Nanjing University of Posts and Telecommunications,Nanjing 210003,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2020年第6期1224-1230,共7页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61971235,61771258)资助
江苏省“六大人才高峰”高层次人才项目(XYDXXJS-044)资助
江苏省“333高层次人才培养工程”资助
南京邮电大学'1311'人才计划资助
中国博士后科学基金(面上一等)项目(2018M630590)资助
南京邮电大学国家自然科学基金孵化项目(NY217057,NY218058)资助
江苏省通信与网络技术工程研究中心开放课题重点项目(JSGCZX17011)资助.
关键词
数据聚合
智能电网
雾计算
哈希链
同态加密
data aggregation
smart grid
fog computing
hash chain
homomorphic encryption
