Research On Privacy Protection Of Data Aggregation In Wireless Sensor Networks

In recent years,with the continuous development of low-power embedded technology and wireless communication technology,Wireless Sensor Networks(WSNs)have been widely used in various fields,and the security of its sensing data has attracted much attention.However,due to the uncertainty of the network deployment environment and the inherent security problems of wireless networks,the privacy of sensing data cannot be guaranteed,so privacy-preserving problem of data has always been concerned in WSNs.In addition,based on the requirement of comprehensive grasp of sensing environment,sensor nodes need to collect multiple types of data,such as fire monitors need to monitor temperature,humidity,smoke concentration and other data at the same time.But,restricted by the energy problem of WSNs,the network can not directly transmit a large number of original data,which requires data aggregation technology to reduce redundant data.So,it is necessary to use multi-dimensional data aggregation privacy-preserving protocol to balance the transmission energy consumption and security issues in WSNs.In this paper,privacy-preserving of data aggregation in WSNs is taken as the main research content.According to the research of its basic theory,the paper study the encrypted privacy-preserving algorithm and the un-encrypted privacy-preserving algorithm to reduce the energy consumption in the network,ensure the data security and aggregation accuracy.Firstly,the basic theories of WSNs are researched and analyzed.Focus on data aggregation for privacy-preserving algorithms of WSNs,And analysis four technologies of network topology,privacy protection,data aggregation and message authentication,which lays the foundation for designing privacy-preserving algorithm of multi-dimensional data aggregation.Secondly,the paper proposes a data aggregation for privacy-preserving algorithm based on homomorphic encryption(PEPA).The LEACH(low energy adaptive clustering hierarchy)algorithm is improved to build cluster dynamic network topology to balance the energy consumption of nodes.Next,according to Elliptic Curve Cryptography(ECC),a key management algorithm with high confidentiality is designed,which combined with homomorphic encryption algorithm.The aggregation node can directly aggregate ciphertext,which avoid the risk of data leakage in process of data transmission.Message authentication algorithm is designed,which uses the Message Authentication Code(MAC)of multi-dimensional data to rotation generate data packets,which makes MAC more difficult to crack.In addition,the scheme uses privacy homomorphic encryption scheme and hop-by-hop message authentication to ensure confidentiality and integrity of data.Finally,the paper proposes a multi-dimensional data aggregation for privacy-preservingalgorithm based on Gaussian distribution perturbation(MDGD),which is a un-encryption data aggregation privacy-preserving scheme.According to the basic characteristics of Gaussian distribution,a data perturbation algorithm is designed.The sensor nodes use distribution function eigenvalues of perturbed data to replace the true data values for transmission,eliminating redundant data and protecting data privacy.In addition,multi-dimensional data aggregation algorithm is proposed,which can safely aggregate various types of data,so that the size of the aggregated data packet is similar than that of the common node.This was not done in the previous multi-dimensional data aggregation.Simulation results and data analysis show that the scheme greatly ensuring the confidentiality for data aggregation process while reduce the network communication overhead.