Research On Secure Multi-Party Computation Against Malicious Adversaries

In this era of rapid network development,information big data has gradually penetrated into all walks of life.Traditional industries have also followed the development of the times,merged and reformed,and gradually introduced new digital technology.In our daily life,emerging technologies such as big data,blockchain,cloud computing and Internet of things have brought fast,convenient and efficient to our life,production and travel.However,things are viewed from both positive and negative aspects.While promoting the progress of human society,data and information will also give rise to new problems,which need to be solved urgently.Among them,information security and privacy protection are becoming complex and diverse.Many information leakage problems in life have brought unprecedented challenges to mankind.Cryptography is the core technology of information security and privacy protection.Cryptography plays an important role in ensuring information security.With the development of cryptography,it has become the research foundation in the fields of privacy protection,Cyberspace Security and blockchain.In recent years,cryptography is the core technology of privacy computing,and the research hotspot in cryptography is secure multi-party computing.In the process of data communication,participants who do not trust each other are unwilling to share more personal information.Therefore,secure multi-party computing solves the problem of private data computing,realizes controllable data sharing,participates in multi-party collaborative computing,and then the data information is protected.The set problem and geometric problem of secure computing are the key research problems of secure multi-party computing,and most of the security protocols designed by people are designed under the semi honest model.Although the protocol under the semi honest model is relatively simple and efficient,it can’t resist the attack of malicious enemies,and the security of the protocol is difficult to be better guaranteed.Therefore,the design of secure multi-party computing protocol under malicious model has extremely important research significance and practical value.Aiming at the problems of secure multi-party set computing and secure multi-party geometric computing,this thesis analyzes the semi honest model protocol designed by current researchers.In order to ensure the security of the protocol more effectively,on this basis,this thesis analyzes the possible malicious behavior,and designs a secure multi-party computing protocol under the malicious model.The main research contents are as follows:(1)Judging the relationship between integer points and integer intervals is the most basic problem in the study of secure computing sets.This paper analyzes the protocol of the relationship between integer points and integer intervals under the semi honest model,and lists some malicious behaviors that may exist in the protocol one by one.Using the cryptographic tools such as Paillier encryption scheme,Goldwasser-Micali encryption algorithm,zero knowledge proof and cut-choose method,a confidentiality judgment protocol for the relationship between integer points and integer intervals under the malicious model is designed.Finally,using the real/ideal model paradigm method,it is proved that the protocol is secure under the malicious model.Compared with the existing schemes,it not only maintains better performance,but also can resist the attack of malicious opponents,so as to make the protocol more fair.(2)Privacy computing Manhattan distance is a new problem in the research of security geometry,which has a wide application prospect.However,there is little research on the secure calculation of Manhattan distance,and the only protocol scheme can’t resist the attack of malicious enemies.This paper analyzes the Manhattan distance secure computing protocol that may be attacked by malicious enemies,and designs a protocol with high security performance and can resist malicious behavior.The security of the protocol under the malicious model is proved by using the real/ideal model paradigm method.Finally,the application scenario of the protocol is given.(3)The inclusion of points and convex polygons is also an important secure computational geometry problem,which has great application value in many fields.However,most of the current research schemes for the inclusion of points and convex polygons are designed under the semi honest model,which can not resist the attack of malicious enemies.This paper analyzes the possible malicious attacks in the point and convex polygon inclusion problem determination protocol under the semi honest model,and designs the point and convex polygon inclusion problem determination protocol under the malicious model by using Paillier encryption algorithm and cryptographic tools such as zero knowledge proof and cut-choose method.Using the real/ideal model paradigm,it is proved that the protocol is secure under the malicious model.Compared with the existing schemes,it is not only efficient,but also resistant to malicious enemy attacks and fair.