Impossibility Results for Static Input Secure Computation
Impossibility Results for Static Input Secure Computation
复制标题
静态输入安全计算的不可能性结果
DOI：
10.1007/978-3-642-32009-5_25
复制
发表时间：
2012
期刊：
影响因子：
--
通讯作者：
Ivan Visconti
中科院分区：
文献类型：
--
作者：
Sanjam Garg;Abishek Kumarasubramanian;R. Ostrovsky;Ivan Visconti
考虑一个设定两个相互不信任的政党爱丽丝和鲍勃，他们希望在预先指定的输入中安全地评估某些功能。两党安全计算的良好研究概念使他们可以在独立设置中这样做。考虑确定性函数，例如，爱丽丝和鲍勃无法琐碎的评估，只允许BOB接收输出。我们表明，即使预先指定了它们的输入，爱丽丝和鲍勃也无法在并发设置中安全地计算任何此类功能。我们的不可能结果也扩展到了爱丽丝和鲍勃获得相同输出的所有确定性功能。我们的结果也对有限的连续设置也有影响。
Consider a setting of two mutually distrustful parties Alice and Bob who want to securely evaluate some function on pre-specified inputs. The well studied notion of two-party secure computation allows them to do so in the stand-alone setting. Consider a deterministic function e.g., 1-out-of-2 bit OT that Alice and Bob can not evaluate trivially and which allows only Bob to receive the output. We show that Alice and Bob can not securely compute any such function in the concurrent setting even when their inputs are pre-specified. Our impossibility result also extends to all deterministic functions in which both Alice and Bob get the same output. Our results have implications in the bounded-concurrent setting as well.