Several Studies On Privacy Protection Issues

In this information times, the rapidly developing internet has been popularized among all industries, and has burst in all aspects of human life. Network is a vast treasure of information, which brings huge benefits to human society. However, network also easily results in information leaking, and thus severly challenges the information security.Cryptography, as a cornerstone of information security, can be used to guarantee the confidentiality, integrality and authentication of information, and therefore to preserve the privacy of private data. Secure multiparty computation (SMC) is presently a research focus in the international cryptographic community and a key technology of privacy preserving and information security in cyberspace. Using SMC, two or more parties can perform cooperative computations that have social, economic, or scientific significance on their private data without leaking them to mine the value of private data.Radio frequency identification (RFID) is a core technology of the Internet of things that can automatically identificate non-contact objects. RFID has advantages over traditional manual identification that it can automatically identify, and has strong capacity of resisting disturbance, wide information processing range, long life etc. However, RFID is also faced with security problems. For the spread applications of RFID, researchers should take the cost into account while designing privacy preserving RFID protocols.This paper studies four practical privacy-preserving problems arising from RFID and SMC. The main contents and contributions of this study are as follows.1. In some RFID applications, it is necessary to prove that a set of tags exist simultaneously, and this is called RFID grouping proof. Efficient and secure grouping proof protocols with low cost are necessary for wide RFID applications. This dissertation proposes a lightweight privacy-preserving grouping proof (LPGP) protocol for RFID systems which uses only a pseudo-random generator with relatively lower computational complexity and a hash algorithm to improve the operating efficiency. Using provable security methodology, we prove that the LPGP protocol can meet the security and privacy requirements of grouping proofs for RFID systems.2. Tag cloning attack is a serious threat to RFID applications. Cloned tags detection is an effective security mechanism to tackle this attack. This study presents a deterministic privacy-preserving cloned tag detection (DCTD) protocol for anonymous RFID systems to detect cloned tags. Using the pseudonym of tags to prevent information revealing and using a tree-based anti-collision algorithm to find irreconcilable collisions, the protocol can quickly detect all the cloned tags within a deterministic time. Experiments show that DCTD protocol, with less detection time and higher accuracy, outperforms the existing detection protocols.3. The millionaires'problem is a basic building block of many other SMC protocols. To the best of our knowledge, almost all protocols used to address the millionaires'problem are applicable to integers, and thus limits their applications. This study proposes two precise and efficient millionaires'protocols for rational numbers based on Paillier's additively homomorphic encryption scheme. We prove that these protocols are secure using simulation paradigm, and have a much lower computational complexity compared with existing protocols. Compared with the existing protocols, ours are more efficient and have wider applications.4. This dissertation investigates a new important SMC problem, privacy-preserving interval computation (PIC). We construct two efficient rational PIC protocols based on the idea of rational millionaires'protocols and Paillier's additively homomorphic encryption scheme. Then we prove their privacy-preserving property using the well accepted simulation paradigm. Finally, we demonstrate some specific applications of these two PIC protocols.