CAREER: Getting the Most out of Secure Multi-Party Computation

职业：充分利用安全多方计算

• 批准号: 1420062
• 负责人: Michael Rosulek
• 金额: $ 35.41万
• 依托单位: Oregon State University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-11-04 至 2018-02-28
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1420062&HistoricalAwards=false
• 关键词: CAREER; Getting; Most; out; Secure

Our society is becoming increasingly reliant on powerful and interconnected computing devices that store much of our personal information. These devices present an ever-growing tension between the desire for our personal information to be private, and the desire to put our personal information to good use for our own convenience. In cryptography, problems that involve requirements of useful computation and privacy are understood through the lens of secure multi-party computation (SMPC). Specifically, SMPC refers to the problem of how mutually distrusting parties can securely and collaboratively perform tasks that involve private information.In this project, the PI will develop new techniques for understanding and overcoming the challenges and limitations posed by SMPC protocols. In particular, the following major themes will be explored: (1) is it possible in the real world to design protocols that remain secure regardless of other concurrently running protocols? (2) what security properties can be embedded into objects like ciphertexts and digital signatures? (3) how can shared data be maintained persistently while remaining private? The PI will develop new materials for undergraduate courses in cryptography and algorithms, based on principles of pedagogically sound visualization. The PI will also create resources for the cryptography research community, in the form of an online database of cryptographic constructions and primitives.

我们的社会正变得越来越依赖于存储我们大部分个人信息的强大且互连的计算设备。这些设备在我们的个人信息隐私的愿望和为了我们自己的方便而充分利用我们的个人信息的愿望之间呈现出不断增长的紧张关系。在密码学中，通过安全多方计算（SMPC）的透镜来理解涉及有用计算和隐私要求的问题。具体来说，SMPC指的是相互不信任的各方如何安全和协作地执行涉及私人信息的任务的问题。在这个项目中，PI将开发新的技术来理解和克服SMPC协议带来的挑战和限制。特别是，将探讨以下主要主题：（1）是否有可能在真实的世界中设计的协议，保持安全，而不管其他并发运行的协议？(2)什么样的安全属性可以嵌入到像密文和数字签名这样的对象中？(3)如何在保持私密性的同时持久地维护共享数据？PI将根据教学声音可视化的原则，为密码学和算法的本科课程开发新材料。PI还将为密码学研究社区创建资源，以密码学结构和原语的在线数据库的形式。