Abstract We consider the problem of securely evaluating the Greater Than (GT) predicate and its extension – transferring one of two secrets, depending on the result of comparison. We generalize our solutions and show how to securely decide membership in the union of a set of intervals. We then consider the related problem of comparing two encrypted numbers. We show how to efficiently apply our solutions to practical settings, such as auctions with the semi-honest auctioneer, proxy selling, etc. All of our protocols are one round. We propose new primitives, Strong Conditional Oblivious Transfer (SCOT) and Conditional Encrypted Mapping (CEM), which capture common security properties of one round protocols in a variety of settings, which may be of independent interest.
Abstract We consider the problem of securely evaluating the Greater Than (GT) predicate and its extension – transferring one of two secrets, depending on the result of comparison. We generalize our solutions and show how to securely decide membership in the union of a set of intervals. We then consider the related problem of comparing two encrypted numbers. We show how to efficiently apply our solutions to practical settings, such as auctions with the semi-honest auctioneer, proxy selling, etc. All of our protocols are one round. We propose new primitives, Strong Conditional Oblivious Transfer (SCOT) and Conditional Encrypted Mapping (CEM), which capture common security properties of one round protocols in a variety of settings, which may be of independent interest.
Abstract. We consider the problem of securely computing the Greater Than (GT) predicate and its generalization – securely determining membership in a union of intervals. We approach these problems from the point of view of Q-Conditional Oblivious Transfer (Q-COT), introduced by Di Crescenzo, Ostrovsky and Rajagopalan [4]. Q-COT is an oblivious transfer that occurs iff predicate Q evaluates to true on the parties’ inputs. We are working in the semi-honest model with computationally unbounded receiver. In this paper, we propose: (i) a stronger, simple and intuitive definition of COT, which we call strong COT, or Q-SCOT. (ii) A simpler and more efficient one-round protocol for securely computing GT and GT-SCOT. (iii) A simple and efficient modular construction reducing SCOT based on membership in a union of intervals (UI-SCOT) to GT-SCOT, producing an efficient one-round UI-SCOT. 1
In this dissertation, we address several issues that arise in protecting communication between parties, as well as in the area of secure function evaluation. Intuitively, the notion of secure function evaluation is clear and natural: several parties wish to compute some function of their inputs without revealing any information about the inputs, other than what is implied by the value of the function. Research included in this dissertation follows three main directions, briefly described below. The first direction (Chapters 3 and 4) is the design of efficient protocols for concrete functions of interest. Specifically, we present new, more efficient protocols for securely computing the Greater Than (GT) function on the inputs of two parties. Secure evaluation of GT is frequently needed in financial transactions. We introduce new primitives, which are convenient building blocks for more complex tasks, and generalize our GT solutions to satisfy them. Based on this, we construct secure auction protocols, protocols for determining whether an integer lies on an interval, and others. The second direction (Chapter 5) is a fundamental approach to secure evaluation of any function, given as a boolean circuit. We present a very efficient information-theoretic (IT) reduction from the problem of secure evaluation of a polysize formula (or, equivalently, a log-depth boolean circuit) to Oblivious Transfer (a fundamental well-researched cryptographic primitive). Our cost of evaluating each gate of the formula is quadratic in its depth, while in previous reductions it was exponential. Our constructions imply efficient one-round protocols for evaluation of polysize formulas on the players' inputs. We extend our solutions to evaluation of polysize circuits, at the cost of having only computational security. The third direction (Chapter 6) is research on key exchange (KE). In contrast with the previous two directions, here the goal is for two parties to protect their communication against eavesdropping and active interference of an external attacker. KE is a basic procedure, frequently used to establish secure channels between parties. It is a prerequisite to a large number of protocols, including those of the above two directions. We demonstrate a subtle flaw in a previous family of KE protocols and give new KE definitions for the following practical "bank" setting. Here, a server wishes to exchange a key with a client. They have a shared password, and the client carries a "bank card", capable of storing several cryptographic keys. Finally, we present new, more efficient KE protocols for this setting, and prove their security.
We present unconditionally perfectly secure protocols in the semi-honest setting for several functionalities: (1) private elementwise equality; (2) private bitwise integer comparison; and (3) bit-decomposition. These protocols are built upon a new concept called Shared Oblivious Transfer (Shared OT). Shared OT extends the one-out-of-N String OT by replacing strings with integers modulo M and allowing additive secret-sharing of all inputs and outputs. These extensions can be implemented by simple local computations without incurring additional OT invocations. We believe our Shared OT may be of independent interest. Our protocols demonstrate the best round, communication, and computational complexities compared to all other protocols secure in a similar setting. Moreover, all of our protocols involve either 2 or 3 rounds.
We propose a solution to the secure integer comparison problem by reducing it to the secure set membership problem on a set where the elements occur in consecutive order. Our protocol provides flexibility due to this reduction such that it can easily be modified to support more complex boolean functions other than the function evaluating the 'Less Than' predicate (LT). In the paper we show one such boolean function, 'IsInInterval', which checks if an integer is in an interval (3I). We also show that our protocols can be easily extended to support shared outputs so that each party would hold a share of the result. Our generic solution can be applied to many e-commerce and data-mining applications requiring one or more such boolean functions to be computed securely. Our protocol not only is adaptable to different boolean functions but also is experimentally more efficient than the state-of-the-art integer comparison protocols supporting shared outputs.
At present, research on secure multi-party computation is of great interest in modern cryptography. It should be acknowledged that if any function can be computed securely, then it results in a very powerful tool. In fact, all natural protocols are, or can be rephrased to be, special cases of the multi-party computation problems. Design and analysis of the special multi-party computation protocols is meaningful and has attracted much interest in this field. Based on the combination of a public-key cryptosystem of the homomorphic encryption and on the theoretic construction relying on the F-hiding assumption, a protocol for comparing information of equality is proposed. The protocol needs only a single round of interaction and ensures fairness, efficiency and security. The protocol is fair, which means that one party knows the sound result of the comparison if and only if the other one knows the result. The protocol is efficient with the help of an oblivious third party for calculating. However, the third party cannot learn any information about the participant抯 private inputs and even about the comparison result, and cannot collude with any participant. The protocol is secure for the two participants, that is, any information about their secret input will not leak except the final computation result. A precise proof of security of the protocol is presented. Applications of this protocol may include private bidding and auctions, secret ballot elections, commercial business, identification in a number of scenarios and so on. It is believed that the protocol may be of practical significance for electronic transaction.
Private integer comparison has been an essential computation function for many applications, including online auction, credential identification, data mining, and joint bidding. In the setting of two-party computation, two parties with private inputs (x and y) want to jointly compare them without revealing the value of those inputs to others (also known as the Millionaires’ problem) while the output should ensure correctness and preserve data privacy. The private inputs only can be revealed if they are equal, i.e., x = y. Many related works have been proposed to solve the integer comparison problem in various settings, focusing on different properties such as round and computation complexity. Most solutions decompose integers into bitwise representation and then securely evaluate the function in a Boolean circuit on encrypted bits. However, this type of solution is costly (especially for large integers) as each bit requires encryption and decryption. In this paper, we transform the private integer comparison into a block comparison problem. In particular, we employ a block vectorization mechanism to encode the private inputs into blocks. We show the security of our two-party protocol in the semi-honest model. Also, we implement the protocol to demonstrate its efficiency using block vectorization mechanism and homomorphic encryption. The experimental result proves that our proposed solution achieves high efficiency, particularly for large integer comparisons.