Access Control for Information in Pervasive Computing Environments

美国国家科学基金会分布式系统研究CISE/ cnsabstract提案编号：0411116首席研究员：Peter steenkisteu机构：卡内基梅隆大学提案标题：普适计算环境中的信息访问控制普适计算环境收集和使用各种个人信息，如位置和活动信息。然而，由于信息的独特性，现有的访问控制机制并不适用：它通常是按需从传感器获得的，是动态的，具有不同的精度级别，并且本质上是以分布式方式收集的。这个项目正在开发访问控制机制，使用户可以控制谁可以访问有关他们的信息。用户在信息级别指定策略，而不必关心如何派生该信息。访问控制机制支持在普适计算环境中至关重要的几个特性，例如指定和利用信息关系的能力、上下文敏感约束的使用以及对指定服务信任的支持。该系统将在Aura全面计算系统的背景下实施和评估。虽然该项目的重点是普适计算环境，但其结果也应适用于处理动态、高度分布式信息（如传感器网络）的其他环境。今天，隐私问题使普及计算技术的部署变得困难，如果不是不可能的话——建议的研究将使普及计算应用程序的部署变得可行。项目成果将通过出版物、我们的原型系统的发布以及教育活动（如本科和研究生课程项目）来传播。Brett D. fleisch CISE/ cns2004年6月21日

National Science FoundationDistributed Systems Research CISE/CNSABSTRACTPROPOSAL NUMBER: 0411116 PRINCIPAL INVESTIGATOR: Peter SteenkisteINSTITUTION: Carnegie-Mellon UniversityPROPOSAL TITLE: Access Control for Information in Pervasive Computing EnvironmentsPervasive computing environments collect and use a variety of personal information, such as location and activity information. However, existing access control mechanisms do not apply because of the unique nature of the information: it is often obtained on-demand from sensors, is dynamic, comes in different levels of precision, and is inherently collected in a distributed fashion. This project is developing access control mechanisms that give users control over who can access information about them. Users specify poli-cies at the level of information without having to be concerned about how that informa-tion is derived. The access control mechanisms support several features that are critical in pervasive computing environments, such as the ability to specify and exploit informa-tion relationships, the use of context-sensitive constraints, and support for specifying ser-vice trust. The system will be implemented and evaluated in the context of the Aura per-vasive computing system. While the focus of the project is on pervasive computing envi-ronments, the results should also apply to other environments that handle dynamic, highly distributed information such as sensor networks. Today, privacy concerns make deploy-ment of pervasive computing technology difficult if not impossible - the proposed re-search will make the deployment of pervasive computing applications feasible. Project results will be disseminated through publications, the release of our prototype system, and educational activities such as undergraduate and graduate course projects.Dr. Brett D. FleischProgram Director, CISE/CNSJune 21, 2004..