Reasearch On Secure Gene Matching Oriented Fully Homomorphic Encryption Over Integer

With the development of technology and the progress of society, the amount ofdaily data increases rapidly and we are also associated with an explosion of information.The cloud storage is unprecedented attentioned. The main purpose of cloud storagetechnology is that the cloud providers provide centralized management and service tothe files which are uploaded by users or some companies. It is not only convenient forusers to use files through a variety of accesses, reducing cross-platform, cross-device’stedious, but also can saves the customer hardware overhead. Although the cloud storagetechnology, the main thechnology of cloud computing, makes people’s life and workeasier, it also brings some problems which can not be ignored. Especially, the datasecurity issue is the most important. Usually if users want to use the cloud services, theyhave to send their data and information into the cloud, such that they will lose thecontrol of their personal data. According to a survey, only20%of people are willing toput their personal data on the cloud storage, leaving the vast majority of the most worryabout whether their personal privacy will be well compromised.Based on the situation above, the research of fully homomorphic encryption hasbecome hot in recent years. Through this method, the users can send encrypted data tothe cloud storage. The cloud service provider can do effectively operation on theencrypted data, and then it returns information to the users who need it. What’s more, inthe whole process the plaintext data is secured to the cloud service provider.This thesis presents a secure gene matching system, which is based on the fullyhomomorphic encryption over integer. It focuses on the contradiction between theappeal of personal information security and the fuzziness of jurisdiction. The presentstudy shows that Homomorphic encryption is a form of encryption which allowsspecific types of computations to be carried out on ciphertext and obtain an encryptedresult which decrypted matches the result of operations performed on the plaintext.Getting the above two parts together is fully homomorphic encryption. And thisalgorithm also needs to ensure that the depth of the arithmetic circuit unrestricted andcan do operations arbitrary times. Getting the above two parts together is fullyhomomorphic encryption. In addition, it has performed an encoding processing on thedata to make sure that the whole algorithm can be effective applicated. The main workof this thesis is as follows.In the perspective of the development process of scientific, this thesis has reviewedof the theoretical system of this study. It began from the public key encryption systemproposing. Then some the semi-homomorphic encryption schemes came out. After the proposing of full homomorphic encryption, and the fully homomorphic encryption overinteger scheme fanally came out.In public key cryptography, the procedure of algorithm has consisted of threeparts, which is called key generation (KeyGen), encryption (Encrypt) and decryption(Decrypt) algorithm. But in the fully homomorphic encryption in order to solve thenoise interference and ensure the encryption circuit can do computation on arbitrarydepth, we need to import a fourth part of the algorithm-evaluate (Eval). These fourparts have been combined to complete the encryption algorithm, which can solve theproblem of fully homomorphic encryption.Secure gene matching problem oriented, this thesis has designed a system thatallows gene sequences matching without revealing plaintext sequences. At the end ofthis thesis, it has made a general introduction to the system. And then it has improvedand optimized the algorithm based on the problem of coupled single-chain in reality.