In the research area of card-based secure computation, one of the long-standing open problems is a problem proposed by Crépeau and Kilian at CRYPTO 1993. This is to develop an efficient protocol using a deck of physical cards that generates uniformly at random a permutation with no fixed points (called a derangement), where the resulting permutation must be secret against the parties in the protocol. All the existing protocols for the problem have a common issue of lacking a guarantee to halt within a finite number of steps. In this paper, we investigate feasibility and infeasibility for the problem where both a uniformly random output and a finite runtime is required. First, we propose a way of reducing the original problem, which is to sample a uniform distribution over an inefficiently large set of the derangements, to another problem of sampling a non-uniform distribution but with a significantly smaller underlying set. This result will be a base of a new approach to the problem. On the other hand, we also give (assuming the abc conjecture), under a certain formal model, an asymptotic lower bound of the number of cards for protocols solving the problem using uniform shuffles only. This result would give a supporting evidence for the necessity of dealing with non-uniform distributions such as in the aforementioned first part of our result.
In the research area of card-based secure computation, one of the long-standing open problems is a problem proposed by Crépeau and Kilian at CRYPTO 1993. This is to develop an efficient protocol using a deck of physical cards that generates uniformly at random a permutation with no fixed points (called a derangement), where the resulting permutation must be secret against the parties in the protocol. All the existing protocols for the problem have a common issue of lacking a guarantee to halt within a finite number of steps. In this paper, we investigate feasibility and infeasibility for the problem where both a uniformly random output and a finite runtime is required. First, we propose a way of reducing the original problem, which is to sample a uniform distribution over an inefficiently large set of the derangements, to another problem of sampling a non-uniform distribution but with a significantly smaller underlying set. This result will be a base of a new approach to the problem. On the other hand, we also give (assuming the abc conjecture), under a certain formal model, an asymptotic lower bound of the number of cards for protocols solving the problem using uniform shuffles only. This result would give a supporting evidence for the necessity of dealing with non-uniform distributions such as in the aforementioned first part of our result.
Many works have been done for secure computation of functions. Most of them assume computation on computers. The protocols are difficult for the people who has no knowledge on cryptography. Therefore, secure computation using cards was considered. Since nothing other than the cards are used, the protocols are easy to understand for the people who has no knowledge on cryptography. This paper presents a card-based protocol that generates a random hidden permutation. This problem is applicable to a case when players exchange gifts. The protocol must not output a permutation with a fixed point, which means that every player does not receive the gift he/she prepared. Ishikawa et. al proposed a protocol with two-color cards using random shuffles and a proof of having no fixed points. However, the success probability of the protocol is not so high. Therefore, more efficient random permutation protocol is required. In this paper, we use cyclic shuffles in addition to the random shuffles to lower the possibility of fixed points. We show the success probability of obtaining a permutation without fixed points by our protocol is better than the one by the existing protocol.
Secure computation is a cryptographic primitive that enables mutually distrustful parties to jointly realize a possibly randomized functionality in a secure manner. Card-based secure computation studies the problem of implementing secure computation protocols visually by using physical cards, to help non-experts understand their mechanisms. Although many card-based secure computation protocols have been proposed, most of them focus on deterministic functionalities and only a few protocols for randomized functionalities are tailored to specific applications. In this work, we initiate the study of card-based protocols for generating uniformly random bits, which is the most fundamental randomized functionality in secure computation. We present four constructions of such protocols differing in the types of shuffles they use. We also show a lower-bound result that the number of shuffles required for protocols using complete shuffles must be at least the number of output bits. Furthermore, we demonstrate that our protocols for generating uniformly random bits can be extended to those for generating possibly non-uniform random numbers, when combined with existing card-based protocols for computing deterministic functions.
Executing a card-based cryptographic protocol is an attractive way to perform secure multiparty computation (MPC) with a deck of physical cards. Crèpeau and Kilian at CRYPTO 1993 proposed card-based AND and XOR protocols that can deal with a logical conjunction and exclusive-or of variables. Their protocols use a familiar shuffling action called a random cut that can be easily implemented by humans, while the numbers of required cards and shuffles are not small enough to be efficient and the protocols are Las Vegas algorithms, i.e., they are not finite-runtime. Several researchers have improved upon the protocols for a quarter of a century. Eventually, there are many AND protocols in the literature, some of which are quite efficient and practical to execute. By contrast, there are only three XOR protocols including the Crèpeau--Kilian protocol mentioned above. In this paper, we design an efficient XOR protocol using only random cuts; the numbers of required cards and shuffles are six and two (which is fixed), respectively. Our proposed XOR protocol is the first construction of a finite-runtime XOR protocol if we restrict ourselves to use only random cuts.
Card-based cryptographic protocols provide secure multi-party computation using a deck of physical cards. The most important primitive of those protocols is the shuffling operation, and most known protocols rely on uniform shuffles (such as the random cut and random bisection cut) in which each possible outcome is equally likely. However, several protocols with non-uniform shuffles have recently been proposed by Koch et al. Compared to other protocols, their protocols require fewer cards to securely produce a hidden AND value, although implementation of the non-uniform shuffle appearing in their protocols remains an open problem. This paper presents a secure implementation of their non-uniform shuffle. To implement the shuffle, we utilize physical cases that can store piles of cards, such as boxes and envelopes. Therefore, humans are able to perform the non-uniform shuffle using these everyday objects.
Card-based cryptography, as first proposed by den Boer [den Boer, 1989], enables secure multiparty computation using only a deck of playing cards. Many protocols as of yet come with an “honest-but-curious” disclaimer. However, modern cryptography aims to provide security also in the presence of active attackers that deviate from the protocol description. In the few places where authors argue for the active security of their protocols, this is done ad-hoc and restricted to the concrete operations needed, often using additional physical tools, such as envelopes or sliding cover boxes. This paper provides the first systematic approach to active security in card-based protocols. The main technical contribution concerns shuffling operations. A shuffle randomly permutes the cards according to a well-defined distribution but hides the chosen permutation from the players. We show how the large and natural class of uniform closed shuffles, which are shuffles that select a permutation uniformly at random from a permutation group, can be implemented using only a linear number of helping cards. This ensures that any protocol in the model of Mizuki and Shizuya [Mizuki and Shizuya, 2014] can be realized in an actively secure fashion, as long as it is secure in this abstract model and restricted to uniform closed shuffles. Uniform closed shuffles are already sufficient for securely computing any circuit [Mizuki and Sone, 2009]. In the process, we develop a more concrete model for card-based cryptographic protocols with two players, which we believe to be of independent interest.
Card-based cryptographic protocols provide secure multi-party computations using a deck of physical cards. The most important primitive of those protocols is the shuffling operation, and most of the existing protocols rely on uniform cyclic shuffles (such as the random cut and random bisection cut) in which each possible outcome is equally likely and all possible outcomes constitute a cyclic subgroup. However, a couple of protocols with non-uniform and/or non-cyclic shuffles were proposed by Koch, Walzer, and Härtel at Asiacrypt 2015. Compared to the previous protocols, their protocols require fewer cards to securely produce a hidden AND value, although to implement of such unconventional shuffles appearing in their protocols remains an open problem. This paper introduces “pile-shifting scramble,” which can be a secure implementation of those shuffles. To implement such unconventional shuffles, we utilize physical cases that can store piles of cards, such as boxes and envelopes. Therefore, humans are able to perform the shuffles using these everyday objects. Furthermore, we show that a certain class of non-uniform and/or non-cyclic shuffles having two possible outcomes can be implemented by the pile-shifting scramble. This also implies that we can improve upon the known COPY protocol using three card cases so that the number of cases required can be reduced to two.