Research On Subset Data Aggregation In Smart Grid

As an important application of the Internet of Things(Io T),Smart Grid(SG)needs to collect consumers’ data in real-time through smart meters and other smart devices.However,these data inevitably involve the sensitive information of families and enterprises,and the release and use of data raised a lot of privacy issues.The privacy-preserving subset data aggregation divides users into different groups and aggregates their real-time power consumption data respectively,which collects data while protecting personal privacy,and ensures the smooth operation of smart grid.However,the current privacy-preserving subset data aggregation protocol still has some problems,such as large resource consumption,low data utility,and internal attackers stealing users’ privacy.In this dissertation,privacypreserving subset data aggregation research is carried out.Aiming at the balance between user privacy and data utility,user privacy,and data collection efficiency in subset data aggregation,two privacy-preserving subset data aggregation protocols are proposed.The specific research are as follows:To improve the data utility of current privacy-preserving subset data aggregation result in smart grid,a privacy-preserving adjustable subset data aggregation protocol is proposed.the data bit is extended in this protocol to get the data situation and the number of users of each subset of users,and each data bit is calculated accurately to prevent overflow.At the same time,the blinding factor is used to ensure that the consumers’ individual data cannot be obtained by the data collector in the collection process,and the homomorphism of Paillier encryption is used to ensure that the data collector can get the overall situation of the user data.Compared with traditional data aggregation,data collectors can obtain more detailed data distribution while protecting user privacy.Compared with other subset data aggregation protocols,our protocol has the advantage of the adjustment of subsets,reducing the computational cost and communication overhead.To solve the problem of heavy communication overhead and various privacy attacks in current privacy-preserving subset data aggregation,a lightweight privacy-preserving subset data aggregation protocol is proposed.According to the large number of smart grid users and the characteristics of real-time data transmission,the protocol uses the replacement strategy to aggregate data and changes the data transmission mode of users based on the randomized response(RR)mechanism to prevent external eavesdroppers from obtaining the data transmission of users.Meanwhile,EC-El Gamal encryption and distributed decryption are adopted to protect individual data,effectively preventing internal data collectors from colluding to steal user data.The security analysis shows that the proposed protocol satisfiesε-local differential privacy and can resist differential privacy attacks.This protocol only requires part of the consumer to submit data,greatly reducing the communication overhead and improving aggregation efficiency.Experiments show that the computing and communication overhead of the proposed protocol has a huge advantage over other protocols.