Quest: Privacy-Preserving Monitoring of Network Data: A System for Organizational Response to Pandemics
Quest: Privacy-Preserving Monitoring of Network Data: A System for Organizational Response to Pandemics
复制标题
探索：网络数据的隐私保护监测：组织应对流行病的系统
DOI：
10.1109/tsc.2022.3166802
复制
发表时间：
2022-05
影响因子：
8.1
通讯作者：
Shantanu Sharma;S. Mehrotra;Nisha Panwar;N. Venkatasubramanian;Peeyush Gupta;Shanshan Han;Guoxi Wang-Guoxi
中科院分区：
文献类型：
--
作者：
Shantanu Sharma;S. Mehrotra;Nisha Panwar;N. Venkatasubramanian;Peeyush Gupta;Shanshan Han;Guoxi Wang-Guoxi
今天，大多数现代组织都支持网络基础设施，以在其场所提供无处不在的网络覆盖。这种由部署在建筑物不同位置的一组接入点组成的网络基础设施可用于支持使用其移动设备连接到基础设施的个人的粗略定位。本文描述了一个名为Quest的系统，该系统支持基于网络数据的各种应用（例如，识别热点区域，查找可能暴露于COVID-19等条件的人员，区域/楼层/建筑物的占用数），以使组织能够维护其工作场所/场所的安全。Quest在充分保护个人隐私的同时构建了上述功能。Quest结合了计算和信息理论上的安全协议，防止对手获得个人位置历史（基于WiFi数据）的知识。我们描述了Quest中提出的安全/隐私技术的体系结构、设计选择和实现。我们还验证了Quest的实用性，并通过我们组织中超过5000万行的大型数据集的实际校园规模部署来彻底评估它。
Most modern organizations today support network infrastructure to provide ubiquitous network coverage at their premises. Such a network infrastructure consisting of a set of access points deployed at different locations in buildings can be used to support coarse-level localization of individuals, who connect to the infrastructure using their mobile devices. This paper describes a system, entitled Quest that supports a variety of applications (e.g., identifying hotspot regions, finding people who are potentially exposed to a condition such as COVID-19, occupancy count of a region/floor/building) based on network data to empower organizations to maintain safety at their workplace/premises. Quest builds the above functionalities while fully protecting the privacy of individuals. Quest incorporates computationally- and information-theoretically-secure protocols that prevent adversaries from gaining knowledge of an individual's location history (based on WiFi data). We describe the architecture, design choices, and implementation of the proposed security/privacy techniques in Quest. We, also, validate the practicality of Quest and evaluate it thoroughly via an actual campus-scale deployment at our organization over a very large dataset of over 50M rows.