Research On IoT Data Processing Mechanism Based On Edge Computing

As the large-scale use of 5G communication technology,it has also driven the development of Internet of Things.The number of IoT devices connected to the Internet has increased linearly,and the massive IoT data generated has brought a great burden to the cloud computing center.Data aggregation technology can merge redundant data,reduce communication overhead and energy consumption,and thus become an important data processing technology in the IoT.However,aggregating data based on traditional cloud computing requires uploading IoT devices’ data to the cloud center,which not only increases the load of transmission bandwidth and increases network latency,but also increases the risk of data leakage.Therefore,according to the advantages of edge computing,such as low latency,high security,and location awareness,this thesis studies the IoT data aggregation mechanism based on edge computing.The main work of this thesis is as follows:(1)For data processing problems in IoT with edge computing,this thesis proposes a lightweight privacy-preserving selective data aggregation mechanism.This mechanism introduces the concepts of boolean response and numerical response,constructs a selective data aggregation method,and sets filter conditions according to the attributes of the data source,which can effectively reduce the consumption of computing and storage resources.Secondly,the improved Pailliar homomorphic encryption algorithm is used to ensure the privacy and confidentiality of the data,and the efficient online/offline signature technology is also used to ensure the integrity of the data.Moreover,the mechanism is fault-tolerant,even if there are malfunctioning IoT devices in the system,the system can complete the aggregation process and recover the aggregated data to support the intelligent decisions.In particular,the mechanism supports dynamic member management,which increases the flexibility of the mechanism.Finally,the theoretical analysis proves the security of the mechanism,and the simulation experiments show that the mechanism is lightweight and has less computation cost and communication overhead.(2)Although the introduction of edge computing solves the delay and efficiency problems faced by traditional cloud computing,the data stored in cloud servers and edge servers may be subject to tampering attacks initiated by attackers,and the cloud server still faces the potential single point of failure.Taking advantage of the decentralized,traceable,and immutable characteristics of blockchain,this thesis proposes a secure and anonymous data aggregation mechanism based on the above research work.The mechanism introduces blockchain technology at the edge layer,constructs a three-layer data aggregation architecture,realizes distributed management of the edge layer,designs an identity authentication mechanism based on the optimized Merkle tree and non-interactive zero-knowledge proof.In addition,the mechanism can realize fine-grained data aggregation and support dynamic member management,which is more practical.Finally,Monte Carlo simulation results prove that the mechanism has stronger security compared with other mechanisms,and simulation experiments illustrate the superiority of the mechanism in terms of computation cost and communication overhead.