Research On Privacy-preserving Raw Data Aggregation Scheme For Edge Computing

With the rapid development of data analysis technology and the widespread application of smart terminal devices,data aggregation has become an important technology in various fields.In data aggregation for edge computing,the edge server collects and aggregates sensing data from terminal devices to provide services such as intelligent computing and real-time control,then uploads the aggregation results to the cloud server for performing global data analysis.Considering that sensing data may involve the privacy information of terminal users,privacypreserving data aggregation schemes for edge computing have been successively proposed.Privacy-preserving raw data aggregation supports the server to collect raw sensing data for calculating arbitrary aggregation functions,which has a wide range of application scenarios.This thesis studies the privacy-preserving raw data aggregation scheme for edge computing.The main works are as follows:1.This thesis proposes a communication-efficient privacy-preserving raw data aggregation scheme for edge computing.The scheme uses the DC-net to achieve raw data aggregation,and utilizes an efficient sequence number distribution algorithm to generate confidential information for terminal devices,thereby canceling the trusted authority.The edge server decrypts by its keys to obtain raw data,preventing the adversary from eavesdropping on all ciphertexts to learn the data.The scheme can achieve privacy and anonymity under the semihonest model.Compared with the existing privacy-preserving raw data aggregation schemes,the scheme has a significant advantage in communication overhead.2.A fault-tolerant and privacy-preserving raw data aggregation scheme for edge computing is proposed.The scheme collects and aggregates the raw data by DC-net,adopts key exchange protocol and sequence number distribution algorithm to distribute keys and confidential data for terminal devices,and introduces the Shamir's secret sharing algorithm and double-masking technology to achieve fault-tolerance for terminal device dropping and network delay.The security analysis proves the privacy and anonymity of the scheme under the semi-honest model.The experimental evaluation shows the practicability of the scheme.