Composable Security in the Tamper-Proof Hardware Model Under Minimal Complexity
Composable Security in the Tamper-Proof Hardware Model Under Minimal Complexity
复制标题
最小复杂度下防篡改硬件模型中的可组合安全性
DOI：
10.1007/978-3-662-53641-4_15
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Muthuramakrishnan Venkitasubramaniam
中科院分区：
文献类型：
--
作者：
Carmit Hazay;Antigoni Polychroniadou;Muthuramakrishnan Venkitasubramaniam
我们提出了一种新的防篡改硬件的全球通用可组合GUC框架中介绍的Canetti等。在TCC 2007年。几乎所有以前的作品都依赖于Katz在Eurocrypt 2007中的公式，并且该公式不能完全捕获并发设置中的令牌。我们通过依赖GUC框架来解决这些缺点，在GUC框架中我们做出了以下贡献：1.我们使用无状态令牌构造了具有最佳轮复杂度和计算假设的安全两方计算2PC协议。更确切地说，我们展示了如何在单向函数OWF的最小假设下，在两轮GUC安全的两方设置中实现任意功能。此外，我们的构造以黑盒的方式依赖于底层函数。作为推论，在最小OWF假设下，我们得到了具有GUC-安全性的多方计算MPC的可行性。作为一个独立的贡献，我们确定了一个问题的索赔在以前的工作由戈亚尔，Ishai，Sahai，Venkatesan和Wadia在TCC 2010中关于无状态令牌的UC安全计算的可行性的讨论，假设抗冲突哈希函数和仅基于单向函数的扩展. 2.然后，我们构造了一个3轮MPC协议来安全地实现GUC的任意功能，安全性从任何半诚实的安全MPC协议开始。对于这种构造，我们需要在STOC 2002中Canetti，Lindell，Ostrovsky和Sahai的原始工作中引入的所谓的一对多提交并证明原语，该原语在底层承诺中是轮效率和黑盒的。使用专门设计的？输入延迟？我们在我们的框架中使用无状态令牌和单向函数来实现这个原语，其中底层的单向函数以黑盒的方式使用。
We put forth a new formulation of tamper-proof hardware in the Global Universal Composable GUC framework introduced by Canetti et al. in TCC 2007. Almost all of the previous works rely on the formulation by Katz in Eurocrypt 2007 and this formulation does not fully capture tokens in a concurrent setting. We address these shortcomings by relying on the GUC framework where we make the following contributions:1.We construct secure Two-Party Computation 2PC protocols for general functionalities with optimal round complexity and computational assumptions using stateless tokens. More precisely, we show how to realize arbitrary functionalities in the two-party setting with GUC security in two rounds under the minimal assumption of One-Way Functions OWFs. Moreover, our construction relies on the underlying function in a black-box way. As a corollary, we obtain feasibility of Multi-Party Computation MPC with GUC-security under the minimal assumption of OWFs. As an independent contribution, we identify an issue with a claim in a previous work by Goyal, Ishai, Sahai, Venkatesan and Wadia in TCC 2010 regarding the feasibility of UC-secure computation with stateless tokens assuming collision-resistant hash-functions and the extension based only on one-way functions.2.We then construct a 3-round MPC protocol to securely realize arbitrary functionalities with GUC-security starting from any semi-honest secure MPC protocol. For this construction, we require the so-called one-many commit-and-prove primitive introduced in the original work of Canetti, Lindell, Ostrovsky and Sahai in STOC 2002 that is round-efficient and black-box in the underlying commitment. Using specially designed ?input-delayed? protocols we realize this primitive with a 3-round protocol in our framework using stateless tokens and one-way functions where the underlying one-way function is used in a black-box way.