SaTC: CORE: Medium: End-to-End Support for Privacy in the Internet -of-things

在过去的几年里，人们对物联网(IoT)的兴趣大幅增加。我们的家庭、建筑甚至城市都开始配备设备、传感器和执行器，这些设备、传感器和执行器将为用户提供新的功能，帮助他们自动完成日常生活中的任务，并最终提高生活质量。然而，物联网长期成功的主要挑战是隐私。物联网传感器和执行器将是多样化的，并嵌入我们周围的基础设施中，主要是在共享空间中，控制和反馈的种类不清楚；此外，使用它们的设备和应用程序通常是由几乎不了解隐私的开发人员创建的。该项目采用基于第一原则的方法来实现全面的隐私控制，以解决这些问题，开发一个开源的物联网堆栈，其中将包括新的隐私机制、隐私模型和对隐私通知的系统支持，允许其他研究人员和从业者在此基础上进行构建。研究成果将作为学生的教育资源，特别是那些来自代表性不足的群体的学生，并通过研究人员教授的新课程和现有课程作为课程项目，如无处不在的个人智能代理、以用户为中心的传感系统和分布式系统。该提案的总体目标是从端到端的角度支持物联网部署中的隐私，采取跨学科的方法来解决物联网隐私中的基本问题，主要有四个方面：(1)进行一系列访谈和调查，以了解用户对物联网隐私的担忧(例如数据类型、粒度、目的等)；(2)在物联网软件堆栈之上开发一套新颖的隐私机制，以使开发人员和最终用户更容易管理隐私，包括新形式的访问控制，使开发人员更容易以隐私敏感的方式获取个人数据的基于流的函数式编程方法，以及对指定数据访问和使用目的的开发人员支持；(3)设计和评估新型隐私通知，以及检查和提供对特定传感器正在做什么的一些控制的方法；以及(4)设计和评估允许管理员和最终用户管理敏感数据的新型策略，包括数据的邻近性和时间降级。这些建议的想法将在大学正在部署的广泛的物联网生活实验室试验台中实施、部署和评估。这项工作将在网络物理系统、人机交互和可用隐私的交叉领域带来显著的跨学科进展。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Interest in the Internet-of-Things (IoT) has increased tremendously over the past few years. Our homes, buildings, and even cities are starting to be instrumented with devices, sensors, and actuators that will provide new capabilities to users, help automate the mundane tasks in their lives, and ultimately to improve quality of life. However, a primary challenge to the long-term success of IoT is privacy. IoT sensors and actuators will be diverse and embedded all around us in the infrastructure, largely in shared spaces with unclear kinds of controls and feedback; further these devices and apps that use them will often be created by developers with little awareness or knowledge of privacy. This project takes a first-principles based approach towards enabling holistic privacy controls to address these concerns, developing an open source IoT stack which will include new privacy mechanisms, privacy models, and systems support for privacy notifications, allowing other researchers and practitioners to build upon them. The research results will serve as an educational resource for students, particularly those from underrepresented groups, and as course projects through new and existing courses taught by the researchers, such as Ubiquitous Personal Smart Agents, User-Focused Sensing Systems and Distributed Systems. The overarching goal of this proposal is to support privacy in Internet of Things deployments from an end-to-end perspective, taking an interdisciplinary approach to address fundamental issues in IoT privacy along four main thrusts: (1) conducting a series of interviews and surveys to understand user concerns about IoT privacy (e.g. type of data, granularity, purpose, and so on); (2) developing a suite of novel privacy mechanisms on top of an IoT software stack to make it easier for developers and end-users to manage privacy, including new forms of access control, a stream-based functional programming approach that makes it easier for developers to get personal data in a privacy-sensitive manner, and developer support for specifying purposes of data access and usage; (3) designing and evaluating new kinds of privacy notifications, as well as ways to inspect and offer some control over what specific sensors are doing; and (4) designing and evaluating new kinds of policies for letting administrators and end-users manage sensitive data, including proximity and temporal degradation of data. These proposed ideas will be implemented, deployed, and evaluated in an extensive IoT living lab testbed that is being deployed at the university. The work will lead to significant interdisciplinary advances at the intersection of cyber-physical systems, human-computer interaction, and usable privacy.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.