Research On Secure Multiparty Computation In The Application Of Internet Of Things

With the rapid development of Internet, cooperative computing between network users become more and more common. If there are security and privacy issues in the process of cooperative computing, while the participants do not trust each other, it is impossible to cooperate. Secure multiparty computation was proposed to deal with the cooperative computing and privacy protection simultaneously. It enables a set of untrusting parties to compute any function of their private inputs while revealing nothing but the result of the function. As of today, secure multiparty computation has been a hot area of research for decades in cryptology, it has extended to many areas, and has also been extensively studied. There are already many scholars study secure multiparty computation protocols on specific scenarios. However there is little research on IOT (Internet of Things) scenarios. Considering the application of Internet of things will be more and more popular, and many applications contain cooperative computing, so it is very important to solve the cooperative computing securely.This paper first introduces some basic knowledge of secure multi-party computation, including the concepts of the participant, the attacker, the computing model and the definition of security. Next, we briefly describe some cryptography tools and basic protocols which are often used to construct secure multiparty computation protocols, such as:homomorphic encryption, millionaires’protocol, oblivious transfer protocol, scalar product protocol and so on. Protocols proposed later in this paper are all built based on these cryptography tools.Then we introduce the basic concept and architecture of Internet of things, and analyze the privacy issue such as data integration, user data protection, users’ location information, we also give the corresponding secure multiparty computation solution.Secondly, under the background of IOT, some cooperation between different organizations is often modeled as solving linear algebra problems, such as linear system of equations or linear inequalities. So we focus on the problem of privacy-preserving cooperative scientific computations and then we design a secure two-party protocol and a secure multi-party protocol based on homomorphic encryption scheme for solving distributed linear system of equations. At last we give the correctness, security and complexity analysis.Finally, considering there are more and more wireless sensor networks deployed in the environment of people’s life. These wireless sensor networks may violate people’s privacy when they are collecting information silently. Then we focus on the problem of privacy preserving computational geometry, and we propose a privacy-preserving point-polygon relation determination protocol, this protocol is used to test the user’s privacy state, determining whether the user is in the monitoring area of wireless sensor network. At last we analyze the correctness, security of the protocols.