ITR Collaborative Research: Biometrics - Performance, Security, and Social Impact

ITR 协作研究：生物识别 - 性能、安全性和社会影响

• 批准号: 0325640
• 负责人: Arun Ross
• 金额: --
• 依托单位: West Virginia University Research Corporation
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-02-15 至 2011-01-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0325640&HistoricalAwards=false
• 关键词: ITR; Collaborative; Research; Biometrics; Performance

Biometric systems automatically measure a physiological or behavioral "signature" of an individual, from which a decision can be rendered to either authenticate or determine the identity of that individual. In this way, biometric systems provide the means to bind the physical presence of an individual user with her cyber action. Despite the increasing use of these systems, neither an analytic framework for modeling and predicting their performance, nor a set of standard multibiometric data is available to researchers for testing purposes. This research addresses these deficiencies by determining an analytic framework in which the performance of biometric systems can be modeled and predicted, removing key barriers to biometric system performance through research in multibiometrics and effective vulnerability countermeasures, and understanding the relationships among biometric applications, privacy, security, and user acceptance that are essential for both informed public policy and system design. Addressing this interrelated set of key technical barriers and societal issues will advance the understanding of biometric system acceptance, performance, and design, and will enable trustworthy authentication and identification functions essential to pervasive computing and supporting homeland security needs. The data sets generated by the coordinated tasks of this research are also being made available as a resource to the biometrics research, developer, and user communities.

生物识别系统自动测量个体的生理或行为“签名”，由此可以做出决定以认证或确定该个体的身份。 通过这种方式，生物识别系统提供了将个人用户的物理存在与她的网络行为绑定在一起的手段。尽管这些系统的使用越来越多，但研究人员既没有用于建模和预测其性能的分析框架，也没有一组标准的多生物特征数据用于测试目的。本研究通过确定一个分析框架，在该框架中，可以对生物识别系统的性能进行建模和预测，消除生物识别系统性能的关键障碍，通过研究多生物特征和有效的漏洞对策，并了解生物识别应用程序，隐私，安全和用户接受度之间的关系，这对知情的公共政策和系统设计都是必不可少的。解决这一系列相互关联的关键技术障碍和社会问题将促进对生物识别系统的接受、性能和设计的理解，并将实现对普适计算和支持国土安全需求至关重要的可信认证和识别功能。这项研究的协调任务所产生的数据集也可作为生物识别研究，开发人员和用户社区的资源。