SaTC: CORE: Medium: Collaborative: Privacy-Aware Trustworthy Control as a Service for the Internet of Things (IoT)

SaTC：核心：媒介：协作：物联网 (IoT) 的隐私意识可信控制即服务

• 批准号: 1705135
• 负责人: Mani Srivastava
• 金额: $ 85万
• 依托单位: University of California-Los Angeles
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-09-01 至 2023-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1705135&HistoricalAwards=false
• 关键词: SaTC; CORE; Medium; Collaborative; Privacy

The Internet of Things (IoT) includes a variety of devices such as smart appliances, cars, and other physical systems that are deeply embedded in our everyday lives and that are at risk from new kinds of threats to security and privacy from hackers or state actors. In IoT systems, sensors are used to probe the physical state of the system (e.g., temperature in a building or rotational speed of a wheel of a car) and then software control systems use algorithms to determine appropriate adjustments to the system (e.g., run the air conditioning for 5 minutes or apply the brakes). The project is focused on protecting those control systems and algorithms to ensure security and privacy for users.The research addresses trustworthy and privacy-aware control architectures for IoT through mechanisms drawn from control, cryptography, software, and hardware. These include: (i) A framework for formally reasoning about safety and privacy properties of control software in conjunction with dynamical models of the physical world and associated sensing and actuation channels; (ii) Lightweight domain-specific mechanisms, for policing flow of information through software applications, while leveraging the semantics of machine learning and control algorithms, physics of the system, and numerical properties; (iii) Enforcing desired safety and information leakage properties via a combination of principled sensor data perturbation, control algorithms optimized for efficient computation over encrypted data, and a hardware-supported trusted computing base tailored to protecting sensed data and control algorithm parameters; (iv) A resilient control and timing infrastructure that protects against attacks on timing information through a hybrid use of edge and cloud resources and physical models. The success of the mechanisms is being assessed on experimental testbeds for smart home, industrial automation and smart vehicles, but have broader applicability to many other IoT applications. The project team is also creating a new graduate class on IoT security and developing educational material on IoT security for high-schoolers through the Los Angeles Computing Circle initiative at the University of California - Los Angeles.

物联网（物联网）包括各种设备，例如智能电器，汽车和其他物理系统，这些设备被深深嵌入我们的日常生活中，并且从黑客或州演员的新型威胁和隐私面临着风险。在物联网系统中，传感器用于探测系统的物理状态（例如，在汽车轮的建筑物或旋转速度中的温度或旋转速度），然后软件控制系统使用算法​​来确定对系统的适当调整（例如，运行空调5分钟或施加制动器）。该项目的重点是保护这些控制系统和算法，以确保用户的安全性和隐私。该研究通过从控制，密码学，软件和硬件中汲取的机制来解决物联网的可信赖和隐私感知的控制体系结构。其中包括：（i）与物理世界的动态模型以及相关的感应和驱动渠道一起正式推理控制软件安全和隐私属性的框架； （ii）轻巧的域特异性机制，用于通过软件应用程序进行警务流动，同时利用机器学习和控制算法，系统物理和数值属性的语义； （iii）通过合并有原始的传感器数据扰动的组合，优化了用于在加密数据上有效计算的控制算法的组合以及硬件支持的受信任的计算基础，以保护感知的数据和控制算法参数； （iv）一种弹性控制和计时基础架构，可通过混合使用边缘，云资源和物理模型来防止对定时信息的攻击。该机制的成功正在对智能家居，工业自动化和智能车辆的实验测试床上进行评估，但对许多其他物联网应用程序具有更广泛的适用性。项目团队还通过加利福尼亚大学洛杉矶分校的洛杉矶计算圈倡议来创建有关物联网安全的新研究生课程，并为高中生提供有关物联网安全的教育材料。

项目成果

Generating Natural Language Adversarial Examples

• DOI: 10.18653/v1/d18-1316
• 发表时间: 2018-04
• 期刊: ArXiv
• 作者: M. Alzantot;Yash Sharma;Ahmed Elgohary;Bo-Jhang Ho;M. Srivastava;Kai-Wei Chang

TinyNS: Platform-aware Neurosymbolic Auto Tiny Machine Learning

• DOI: 10.1145/3603171
• 发表时间: 2024-05-01
• 期刊: ACM TRANSACTIONS ON EMBEDDED COMPUTING SYSTEMS
• 影响因子: 2
• 作者: Saha，Swapnil Sayan;Sandha，Sandeep Singh;Srivastava，Mani
• 通讯作者: Srivastava，Mani

SecDeep: Secure and Performant On-device Deep Learning Inference Framework for Mobile and IoT Devices

• DOI: 10.1145/3450268.3453524
• 发表时间: 2021-05
• 期刊: Proceedings of the International Conference on Internet-of-Things Design and Implementation
• 作者: Renju Liu;L. Garcia;Zaoxing Liu;Botong Ou;M. Srivastava

Decentralized Resilient State-Tracking

去中心化的弹性状态跟踪

• DOI: 10.1109/cdc45484.2021.9683313
• 发表时间: 2021
• 期刊: 2021 60th IEEE Conference on Decision and Control (CDC
• 作者: Mao, Yanwen;Tabuada, Paulo
• 通讯作者: Tabuada, Paulo

Hard-Label Black-Box Adversarial Attack on Deep Electrocardiogram Classifier

对深度心电图分类器的硬标签黑盒对抗攻击

• DOI: 10.1145/3417312.3431827
• 发表时间: 2020
• 期刊: SecICPS '20: Proceedings of the 1st ACM International Workshop on Security and Safety for Intelligent Cyber-Physical Systems
• 作者: Lam, Jonathan;Quan, Pengrui;Xu, Jiamin;Jeyakumar, Jeya Vikranth;Srivastava, Mani
• 通讯作者: Srivastava, Mani