Research On The Private Set Intersection In Cloud Outsourcing Environment

Private Set Intersection(PSI)is an important component of the secure multi-party computing,which can be described as many participants who want to use their own private data to complete a specific calculation and in the process we must ensure that participants cannot learn any other private data except final results.At present,PSI is becoming a hot research issue,and it certainly has important theoretical and realistic value.With the rapid development of cloud computing and mobile Internet,the private set intersection protocol based on cloud outsourcing computing has become a hotspot in the field of information security.However,users need to submit the private data to cloud servers in the process of calculation or storage,so the key problem of cloud outsourcing private set intersection protocols becomes how to ensure the data security.However,existing cloud outsourcing private set intersection protocols have lower data storage rate and consume more resources,and they often need more frequent data interaction between the parties involved,meanwhile,researches on verifiable computing are absent to some extent.To solve above problems,this thesis researches on the cloud outsourcing private set intersection protocol by introducing the bloom filter,the proxy oblivious transmission and the exchange method of the secret key.The main contribution of this thesis are as follows:(1)In the process of calculation of the private set intersection between two parties,combined with the cloud outsourcing environment,the efficient garbled bloom filter is implemented to achieve safe and efficient calculation.In this thesis,the limitation of the protocol proposed by Dong which only one party can learn final results is breached,meanwhile,the problem of the protocol cannot be integrated with the cloud outsourcing environment is solved.Because the cloud server acts as proxies for the calculation,each cloud tenant does not have to interact with each other,thereby this method greatly reduces the operational load.In addition,cloud tenants do not need keep online but just transfer their privacy data to the server at the initial stage,and then the server does the corresponding data calculation,thus cloud tenants can timely and easily get the final results.(2)On the basis of the above-mentioned improved private set intersection protocol between two parties,aiming at problems that portable mobile terminals have lower hardware level and limited resource computing ability,a multi-party private set intersection scheme is proposed in this thesis.In this scheme,the efficient and correct bloom filter and its variants are used to realize the data storage and query,and outsource the complex computing tasks to the cloud server.Meanwhile,the parties use their own public and private secret keys to avoid the data security issues brought by use of the same key.The protocol can be effectively applied to mobile terminals with weak computing capabilities(such as smart phones,PDAs,etc.).(3)In the actual scene,some cloud service providers may arbitrarily return the information of intersection results so that the customer cannot get the correct results of the intersection,leading to users not only leaked their own privacy data but also can not get the data information they need.To prevent this,a corresponding information monitoring mechanism is required to ensure the completeness and correctness of their data.In this thesis,the participants copy elements k times,and the finally results have to be detected whether contain all k copies so as to verify the integrity and correctness of the data.This method can effectively prevent cloud servers from tampering maliciously with the data of participants.