In secure multi-party computation $n$ parties jointly evaluate an $n$-variate function $f$ in the presence of an adversary which can corrupt up till $t$ parties. Almost all the works that have appeared in the literature so far assume the presence of authenticated channels between the parties. This assumption is far from realistic. Two directions of research have been borne from relaxing this (strong) assumption: (a) The adversary is virtually omnipotent and can control all the communication channels in the network, (b) Only a partially connected topology of authenticated channels is guaranteed and adversary controls a subset of the communication channels in the network. This work introduces a new setting for (unconditional) secure multiparty computation problem which is an interesting intermediate model with respect to the above well studied models from the literature (by sharing a salient feature from both the above models). We consider the problem of (unconditional) secure multi-party computation when 'some' of the communication channels connecting the parties can be corrupted passively as well as actively. For this setting, some honest parties may be connected to several other honest parties via corrupted channels and may not be able to authentically communicate with them. Such parties may not be assured the canonical guarantees of correctness or privacy. We present refined definitions of security for this new intermediate model of unconditional multiparty computation. We show how to adapt protocols for (Unconditional) secure multiparty computation to realize the definitions and also argue the tightness of the results achieved by us.
In secure multi-party computation n parties jointly evaluate an n-variate function f in the presence of an adversary which can corrupt up till t parties. Almost all the works that have appeared in the literature so far assume the presence of authenticated channels between the parties. This assumption is far from realistic. Two directions of research have been borne from relaxing this (strong) assumption: (a) The adversary is virtually omnipotent and can control all the communication channels in the network, (b) Only a partially connected topology of authenticated channels is guaranteed and adversary controls a subset of the communication channels in the network. This work introduces a new setting for (unconditional) secure multiparty computation problem which is an interesting intermediate model with respect to the above well studied models from the literature (by sharing a salient feature from both the above models). We consider the problem of (unconditional) secure multi-party computation when 'some' of the communication channels connecting the parties can be corrupted passively as well as actively. For this setting, some honest parties may be connected to several other honest parties via corrupted channels and may not be able to authentically communicate with them. Such parties may not be assured the canonical guarantees of correctness or privacy. We present refined definitions of security for this new intermediate model of unconditional multiparty computation. We show how to adapt protocols for (Unconditional) secure multiparty computation to realize the definitions and also argue the tightness of the results achieved by us.
Secure multiparty computation of a multivariate function is a central problem in cryptography. It is well known that secure multiparty computation can be realized by a set of n parties iff the connectivity of the underlying (authenticated) communication network is more than twice the number of corrupted parties. This impossibility result makes secure multiparty computation far less applicable in practice, as most deployed networks have a degree much lower than O(n) and one would ideally like to be able to tolerate Θ(n) corrupted parties. In this work we consider a recently proposed model for (Unconditional) secure multiparty computation for networks of low degrees for which authenticated channels are available only between very few pairs of parties. Not all honest parties may be expected to achieve traditional security guarantees of multiparty computation for this model because of theoretical limitations posed by lack of infrastructure. Honest parties which are not given canonical guarantees of Correctness and Privacy are called ”sacrificed” and the resulting notion is called almost everywhere secure computation. In this conceptual note, we investigate the previous results about this model by Garay and Ostrovsky'2008. We explain why these results hold only for weak honest-but-curious type passive corruptions. We emphasize why the results for almost everywhere secure computation are theoretically interesting and practically relevant only when active malicious corruptions are allowed. We argue the limitation of hybrid argument for realizing privacy property when Byzantine corruptions are allowed. From this we deduce that simulation based reduction approach is necessary to realize even input indistinguishability type definition of privacy for the stand alone setting. We present a conceptual exposition of the simulator based approach for defining privacy of a.e.s.c. Finally, we present a brief overview of technicalities involved in realizing a.e.s.c. when malicious corruptions are allowed.
This paper improves on the classical results in uncondition- ally secure multi-party computation among a set of n players, by consid- ering a model with three simultaneously occurring types of player cor- ruption: the adversary can actively corrupt (i.e. take full control over) up to ta players and, additionally, can passively corrupt (i.e. read the entire information of) up to tp players and fail-corrupt (i.e. stop the compu- tation of) up to tf other players. The classical results in multi-party computation are for the special cases of only passive (ta = tf =0 ) or only active (tp = tf = 0) corruption. In the passive case, every function can be computed securely if and only if tp <n = 2. In the active case, every function can be computed securely if and only if ta <n =3; when a broadcast channel is available, then this bound is ta <n =2. These bounds are tight. Strictly improving these results, one of our results states that, in addi- tion to tolerating ta <n =3 actively corrupted players, privacy can be guaranteed against every minority, thus tolerating additional tp n=6 passively corrupted players. These protocols require no broadcast and have an exponentially small failure probability. We further show that the bound t<n =2 for passive corruption holds even if the adversary is additionally allowed to make the passively corrupted players fail. Moreover, we characterize completely the achievable thresholds ta, tp and tf for four scenarios. Zero failure probability is achievable if and only if 3ta +2 tp + tf <n ; this holds whether or not a broadcast chan- nel is available. Exponentially small failure probability with a broadcast channel is achievable if and only if 2ta +2 tp +tf <n ; without broadcast, the additional condition 3ta + tf <n is necessary and sucient. In this corrected version, an error pointed out by Damg ard (Dam99) is corrected.
We consider a recently proposed model for secure computation appropriate to the setting of low degree networks called almost everywhere secure computation. This model of multiparty computation allows a few honest parties to not achieve the canonical guarantees of Correctness and Privacy. Such honest parties may not be able to communicate reliably or securely with other honest parties in the network due to lack of infrastructure. We explain why a straightforward hybrid argument employed in the previous work can be used to realize privacy only when honest-but-curious type passive corruptions are considered. We further note that the notion of almost everywhere secure computation is theoretically challenging and practically relevant only when malicious corruptions are allowed. We argue and emphasize why simulation based reduction approach taken by the author is the only way to meaningfully realize privacy of almost everywhere secure computation. We present such an approach and show how to realize a.e.s.c. for general Byzantine corruptions, resolving the principle open problem in this line of research. Finally, we note several technical and conceptual improvements to the results given in previous work.
The notion of secure computation is central to cryptography. Introduced in the seminal works of Yao [FOCS'82, FOCS'86] and Goldreich, Micali and Wigderson [STOC'87], secure multiparty computation allows a group of (mutually) distrustful parties to jointly compute any functionality over their individual private inputs in such a manner that the honest parties obtain the correct outputs and no group of malicious parties learn anything beyond their inputs and the prescribed outputs. General feasibility results for secure computation were given by Yao [FOCS'86] and Goldreich et al. [STOC'87] more than two decades ago. Subsequent to these works, designing secure computation protocols that can tolerate more powerful adversaries and satisfy stronger notions of security has been a very active area of research. In this dissertation, we study two such new frontiers in the area of secure computation.
In the first part of this dissertation, we initiate a study of designing leakage-resilient interactive protocols. Specifically, we consider the scenario where an adversary, in addition to corrupting a subset of parties, can leak (potentially via physical attacks) arbitrary information from the secret state of any honest party. This is in contrast to the standard notion of secure computation where it is assumed that the adversary only has black-box'' access to the honest parties. In particular, we formalize a meaningful definition of leakage-resilient zero knowledge proof systems and provide constructions that satisfy our notion. We also discuss various applications of our results.
The second part of this dissertation concerns with the general question of whether it is possible to securely run cryptographic protocols over an insecure network environment such as the Internet. It is well-known that the standard notion of secure computation is only relevant to the stand-alone setting where a single protocol is being executed in isolation; as such it does not guarantee security when multiple protocol sessions may be executed concurrently under the control of an adversary who is present across all sessions. We consider the open problem of constructing secure password-based authenticated key exchange protocols in such a setting in the plain model (i.e., without assuming any trusted infrastructure or random oracles). We give the first construction of such a protocol based on standard cryptographic assumptions. Our results are in fact much more general, and extend to other functionalities w.r.t. a (necessarily) weakened notion of concurrently secure computation.
The results presented in this dissertation stem from two papers which are joint works with Sanjam Garg and Amit Sahai, and with Vipul Goyal and Rafail Ostrovsky, respectively.
Secure multiparty computation (MPC) is perhaps the most popular paradigm in the area of cryptographic protocols. It allows several mutually untrustworthy parties to jointly compute a function of their private inputs, without revealing to each other information about those inputs. In the case of unconditional (information-theoretic) security, protocols are known which tolerate a dishonest minority of players, who may coordinate their attack and deviate arbitrarily from the protocol specification. It is typically assumed in these results that parties are connected pairwise by authenticated, private channels, and that in addition they have access to a “broadcast ” channel. Broadcast allows one party to send a consistent message to all other parties, guaranteeing consistency even if the broadcaster is corrupted. Because broadcast cannot be simulated on the point-to-point network when more than a third of the parties are corrupt, it is impossible to construct general MPC protocols in this setting without using a broadcast channel (or some equivalent addition to the model). A great deal of research has focused on increasing the efficiency of MPC, primarily in terms of round complexity and communication complexity. In this work we propose a refinement of the round complexity which we term broadcast complexity. We view the broadcast channel as an expensive resource
Historically, cryptography was created and used to deliver messages between people in such a way that only the intended receiver would be able to read them. The two persons communicating trusted each other, but not the rest of the world - for instance, not the message bearer. But what happens if we assume instead that even the receiver of a message may not be trusted by the sender, while both still have an interest in communicating? This problem is addressed by secure multi-party computation, that studies communication and shared computation between mutually distrusting parties. In a world where most of the communication happens over the Internet, and where an ever-increasing amount of computation is outsourced, for example to the cloud, multi-party computation is a subject of crucial importance. However, unconditionally secure multi-party protocols have never found practical application: the lack of realistic noisy channel models, that are required to achieve security against computationally unbounded adversaries, prevents implementation over real-world, standard communication protocols.\n\nIn this thesis, we study the properties and characteristics of noisy channels, and we introduce new models based on real-world error sources that have never been studied in this context, such as delays. On the contrary, previous works mostly focused on theoretical noisy channels that have little in common with the reality of network communication. We show that the models we propose are suitable for multi-party computation by building on each of them a protocol for oblivious transfer, a fundamental cryptographic primitive that implies any secure computation. Thanks to the realism of our noisy channels, we are able to provide the first working implementation of unconditionally secure oblivious transfer over both wired and wireless networks and different Internet protocols, which we test thoroughly in an extensive set of experiments. Our constructions are not only more realistic, but also more flexible, as they allow larger ranges of error probabilities. This will eventually open the way for the widespread use of unconditionally secure multi-party computation.