Research Of Privacy Preserving Algorithm Based On Slicing Technology In Wireless Sensor Network

Wireless sensor network(WSN)represents a distributed network system intended for data acquisition,transmission and processing.WSNs exchange data wirelessly,and they are widely used in medical,aerospace,military.Due to wireless transmission and random deployment of wireless sensor nodes,the data can be eavesdropped easily,thus,the privacy preserving is very important.This thesis studies WSN privacy preserving based on slicing technology,which mainly includes the following.Firstly,in order to solve the problem of packet loss that privacy preserving algorithm based on slicing technology is incapable to solve,the linear redundancy non-encryption algorithm for privacy preserving is proposed.This algorithm uses hidden data and adds perturbation data to improve the privacy preserving.Nonetheless,it can restore the original data when data are partly lost.According to the simulation via TOSSIM,in the case of partial packet loss,the algorithm can completely restore the original data.Furthermore,we compared accuracy of proposed algorithm,probability of data reduction,data fitting degree,communication overhead,and packet loss rate.Secondly,in order to provide a solution for instability and internal attacks in WSNs,a privacy homomorphism algorithm based on slicing technology is proposed.The algorithm builds the feedback tree between source node and destination node,thus,the slicing strategy can be adjusted dynamically,and the instability caused by packet loss can be overcome.At the same time,this algorithm encrypts hidden data and original data and checks the data of the target node to enhance privacy preserving.The simulations via TOSSIM,wherein dynamic adjustment,reduction probability,and communication cost of proposed algorithm and traditional privacy preserving algorithm were compared,have shown that proposed algorithm has higher privacy preserving performance than traditional privacy preserving algorithm.Finally,the prototype system is designed for both previously mentioned algorithms.The performance of these algorithms in terms of monitoring,traffic analysis,data modification,and replay attack is presented.Namely,the linear redundancy non-encryption algorithm for privacy preserving can effectively cope with monitoring and traffic analysis.Furthermore,it was shown that privacy homomorphism based on slicing technology can be also used to deal with data modification and replay attacks.