Abstract. Universally composable (UC) multi-party computation has been studied in two settings. When a majority of parties are honest, UC multi-party computation is possible without any assumptions. Without a majority of honest parties, UC multi-party computation is impossible in the plain model, but feasibility results have been obtained in various augmented models. The most popular such model posits a common reference string (CRS) available to parties executing the protocol. In either of the above settings, some assumption regarding the protocol execution is made: i.e., that many parties are honest in the first case, or that a legitimately-chosen string is available in the second. If this assumption is incorrect then all security is lost. A natural question is whether it is possible to design protocols secure if either one of these assumptions holds, i.e., a protocol which is secure if either at most s players are dishonest or if up to t> s players are dishonest but the CRS is chosen in the prescribed manner. We show that such protocols exist if and only if s + t < n. 1
Abstract. Universally composable (UC) multi-party computation has been studied in two settings. When a majority of parties are honest, UC multi-party computation is possible without any assumptions. Without a majority of honest parties, UC multi-party computation is impossible in the plain model, but feasibility results have been obtained in various augmented models. The most popular such model posits a common reference string (CRS) available to parties executing the protocol. In either of the above settings, some assumption regarding the protocol execution is made: i.e., that many parties are honest in the first case, or that a legitimately-chosen string is available in the second. If this assumption is incorrect then all security is lost. A natural question is whether it is possible to design protocols secure if either one of these assumptions holds, i.e., a protocol which is secure if either at most s players are dishonest or if up to t> s players are dishonest but the CRS is chosen in the prescribed manner. We show that such protocols exist if and only if s + t < n. 1
A desirable goal for cryptographic protocols is to guarantee security when the protocol is composed with other protocol instances. Universally composable (UC) protocols provide this guarantee in a strong sense: A protocol remains secure even when composed concurrently with an unbounded number of instances of arbitrary protocols. However, UC protocols for carrying out general tasks are known to exist only if a majority of the participants are honest, or in the common reference string (CRS) model where all parties are assumed to have access to a common string that is drawn from some pre-defined distribution. Furthermore, carrying out many interesting tasks in a UC manner and without honest majority or set-up assumptions is impossible, even if ideally authenticated communication is provided. A natural question is thus whether there exist more relaxed set-up assumptions than the CRS model that still allow for UC protocols. We answer this question in the affirmative: we propose alternative and relaxed set-up assumptions and show that they suffice for reproducing the general feasibility results for UC protocols in the CRS model. These alternative assumptions have the flavor of a "public-key infrastructure": parties have registered public keys, no single registration authority needs to be fully trusted, and no single piece of information has to be globally trusted and available. In addition, unlike known protocols in the CRS model, the proposed protocols guarantee some basic level of security even if the set-up assumption is violated.
Abstract. The recently proposed universally composable (UC) security framework, for analyzing security of cryptographic protocols, provides very strong security guarantees. In particular, a protocol proven secure in this framework is guaranteed to maintain its security even when deployed in arbitrary multi-party, multi-protocol, multi-execution environments. Protocols for securely carrying out essentially any cryptographic task in a universally composable way exist, both in the case of an honest majority (in the plain model, i.e., without set-up assumptions) and in the case of no honest majority (in the common reference string model). However, in the plain model, little was known for the case of no honest majority and, in particular, for the important special case of two-party protocols. We study the feasibility of universally composable two-party function evaluation in the plain model. Our results show that very few functions can be computed in this model so as to provide the UC security guarantees. Specifically, for the case of deterministic functions, we provide a full characterization of the functions computable in this model. (Essentially, these are the functions that depend on at most one of the parties’ inputs, and furthermore are “efficiently invertible ” in a sense defined within.) For the case of probabilistic functions, we show that the only functions computable in this model are those where one of the parties can essentially uniquely determine the joint output. 1
A desirable goal for cryptographic protocols is to guarantee security when the protocol is composed with other protocol instances. Universally Composable (UC) security provides this guarantee in a strong sense: A UC-secure protocol maintains its security properties even when composed concurrently with an unbounded number of instances of arbitrary protocols. However, many interesting cryptographic tasks are provably impossible to realize with UC security in the standard, “plain” model of computation. Impossibility holds even if ideally authenticated communication channels are provided. In contrast, it has been demonstrated that general secure computation can be obtained in a number of idealized models. Each one of these models represents a form of trust that is put in some of the system’s components. This survey examines and compares some of these trust models, both from the point of view of their sufficiency for building UC secure protocols, and from the point of view of their practical realizability. We start with the common reference string (CRS) model, and then describe several relaxations and alternatives including the Defective CRS model, the key registration models, the hardware token model, the global and augmented CRS models, and a timing assumption. Finally, we briefly touch upon trust models for obtaining authenticated communication.
Universal composability and concurrent general composition consider a setting where secure protocols are run concurrently with each other and with arbitrary other possibly insecure protocols. Protocols that meet the definition of universal composability are guaranteed to remain secure even when run in this strongly adversarial setting. In the case of an honest majority, or where there is a trusted setup phase of some kind (like a common reference string or the keyregistration public-key infrastructure of Barak et al. in FOCS 2004), it has been shown that any functionality can be securely computed in a universally composable way. On the negative side, it has also been shown that in the plain model where there is no trusted setup at all, there are large classes of functionalities which cannot be securely computed in a universally composable way without an honest majority. In this paper we extend these impossibility results for universal composability. We study a number of public-key models and show for which models the impossibility results of universal composability hold and for which they do not. We also consider a setting where the inputs to the protocols running in the network are fixed before any execution begins. The majority of our results are negative and we show that the known impossibility results for universal composability in the case of no honest majority extend to many other settings.
Universally composable (UC) protocols retain their security properties even when run concurrently alongside arbitrary other protocols. Unfortunately, it is known that UC multiparty computation (for general functionalities, and without assuming honest majority) is impossible without some form of setup. To circumvent this impossibility, various complete setup assumptions have been proposed. With only a few exceptions, past work has viewed these setup assumptions as being implemented by some ideal, incorruptible entity. Any such entity is thus a single point of failure, and security fails catastrophically in case the setup entity is subverted by an adversary. We propose here a clean, general, and generic approach for distributing trust among m arbitrary setups, by modeling potential corruption of setups within the UC framework, where such corruption might be fail-stop, passive, or arbitrary and is in addition to possible corruption of the parties themselves. We show several feasibility and impossibility results in this model, for different specifications of the corruptible sets. For example, we show that given m complete setups, up to t of which might be actively corrupted in an adaptive manner, general multiparty computation with no honest majority is possible if and only if t < m/2.
In this work, we study the intrinsic complexity of black-box Universally Composable (UC) secure computation based on general assumptions. We present a thorough study in various corruption mod-elings while focusing on achieving security in the common reference string (CRS) model. Our results involve the following: • Static UC secure computation. Designing the first static UC secure oblivious transfer protocol based on public-key encryption and stand-alone semi-honest oblivious transfer. As a corollary we obtain the first black-box constructions of UC secure computation assuming only two-round semi-honest oblivious transfer. • One-sided UC secure computation. Designing adaptive UC secure two-party computation with single corruptions assuming public-key encryption with oblivious ciphertext generation. • Adaptive UC secure computation. Designing adaptively secure UC commitment scheme assum-ing only public-key encryption with oblivious ciphertext generation. As a corollary we obtain the first black-box constructions of adaptive UC secure computation assuming only (trapdoor) simu-latable public-key encryption (as well as a variety of concrete assumptions). We remark that such a result was not known even under non-black-box constructions.