CAREER: Building Reliable Network of Unreliable Things

职业：建立不可靠事物的可靠网络

• 批准号: 2154059
• 负责人: Lu Su
• 金额: $ 50.89万
• 依托单位: Purdue University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-10-01 至 2024-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2154059&HistoricalAwards=false
• 关键词: CAREER; Building; Reliable; Network; Unreliable

Recent years have witnessed the rise of Internet of Things (IoT), a newly emergent networking paradigm that connects humans and the physical world through ubiquitous sensing, computing, and communicating devices. With the ultimate goal of building reliable, robust, and secure IoT systems that are usually composed of multitudes of unreliable wireless devices - sometimes even carried by malicious users - this project develops TRIP, a 3-in-1 integrated framework of TRuth discovery, Incentive, and Privacy preserving mechanisms for IoT systems. This framework consists of 1) a truth discovery mechanism that can distill true information from the deluge of sensory data generated by the ubiquitous IoT devices, 2) a security and privacy mechanism that can not only protect user privacy but also defend against malicious attack, and 3) an incentive mechanism that can select reliable participants in order to maximize the quality of collected information. The successful completion of this project, which for the first time systematically investigates an integrated design of IoT systems, will facilitate not only a whole spectrum of applications that have significant natural and societal impact, but also various curriculum development and outreach activities that can benefit a large group of students, especially the female and minority students.In order to realize the proposed TRIP system, this project addresses a series of important research challenges. First, to infer truth information from the noisy and sparse sensory data collected by human-carried IoT devices, this project delivers a novel truth discovery method that can capture the variety in the reliability of different users as well as the correlations among the objects. Second, this project investigates not only privacy preserving mechanisms that allow the server to conduct truth discovery operations without knowing the genuine values of user observations, but also security solutions that defend against malicious users who try to inject false information for the purpose of sabotage or financial rewards. Third, an effective incentive mechanism is developed to motivate reliable users to contribute to the IoT tasks. Finally, the above mechanisms are seamlessly integrated into networked IoT systems in a distributed and parallel manner, and evaluated on real testbeds in order to validate the effectiveness, efficiency, and the practicality of the proposed research.

近年来，物联网（IoT）的兴起，这是一种新兴的网络范式，通过无处不在的传感，计算和通信设备连接人类和物理世界。最终目标是构建可靠、强大和安全的物联网系统，这些系统通常由大量不可靠的无线设备组成（有时甚至由恶意用户携带），该项目开发了TRIP，这是TRuth发现、激励和隐私保护的3合1集成框架物联网系统的机制。该框架包括：1）一个真相发现机制，可以从无处不在的物联网设备产生的大量传感数据中提取真实信息; 2）一个安全和隐私机制，不仅可以保护用户隐私，还可以抵御恶意攻击; 3）一个激励机制，可以选择可靠的参与者，以最大限度地提高收集信息的质量。该项目首次系统地研究了物联网系统的综合设计，其成功完成不仅将促进具有重大自然和社会影响的各种应用，还将促进各种课程开发和推广活动，使大量学生，特别是女性和少数民族学生受益。为了实现拟议的TRIP系统，该项目涉及一系列重要的研究挑战。首先，为了从人类携带的物联网设备收集的嘈杂和稀疏的传感数据中推断真相信息，该项目提供了一种新颖的真相发现方法，可以捕获不同用户的可靠性以及对象之间的相关性。其次，该项目不仅研究了隐私保护机制，允许服务器在不知道用户观察的真实值的情况下进行真相发现操作，而且还研究了安全解决方案，以防止恶意用户试图注入虚假信息以进行破坏或获得经济奖励。第三，开发有效的激励机制，以激励可靠的用户为物联网任务做出贡献。最后，将上述机制以分布式并行的方式无缝集成到物联网系统中，并在真实的测试平台上进行了测试，验证了所提研究的有效性、高效性和实用性。

项目成果

Joint Charging and Relocation Recommendation for E-Taxi Drivers via Multi-Agent Mean Field Hierarchical Reinforcement Learning

通过多智能体平均场分层强化学习为电动出租车司机提供联合收费和搬迁建议

• DOI: 10.1109/tmc.2020.3022173
• 发表时间: 2022-04
• 期刊: IEEE Transactions on Mobile Computing
• 影响因子: 7.9
• 作者: Enshu Wang;Rong Ding;Zhaoxing Yang;Haiming Jin;Chenglin Miao;Lu Su;Fan Zhang;Chunming Qiao;Xinbing Wang
• 通讯作者: Xinbing Wang

Adversarial Attacks against LiDAR Semantic Segmentation in Autonomous Driving

• DOI: 10.1145/3485730.3485935
• 发表时间: 2021-11
• 期刊: Proceedings of the 19th ACM Conference on Embedded Networked Sensor Systems
• 作者: Yi Zhu;Chenglin Miao;Foad Hajiaghajani;Mengdi Huai;Lu Su;Chunming Qiao

Towards Backdoor Attacks against LiDAR Object Detection in Autonomous Driving

• DOI: 10.1145/3560905.3568539
• 发表时间: 2022-11
• 期刊: Proceedings of the 20th ACM Conference on Embedded Networked Sensor Systems
• 作者: Yan Zhang;Yi Zhu;Zihao Liu;Cheng-yi Miao;Foad Hajiaghajani;Lu Su;Chunming Qiao

Soil Moisture Sensing with UAV-Mounted IR-UWB Radar and Deep Learning

• DOI: 10.1145/3580867
• 发表时间: 2022-03
• 期刊: Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies
• 作者: Rong Ding;Haiming Jin;Dong Xiang;Xiaocheng Wang;Yongkui Zhang;Dingman Shen;Lu Su;Wentian Hao;Ming Tao;Xinbing Wang;Cheng Zhou

Peer-to-Peer Federated Continual Learning for Naturalistic Driving Action Recognition

• DOI: 10.1109/cvprw59228.2023.00553
• 发表时间: 2023-04
• 期刊: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
• 作者: Liangqi Yuan;Yunsheng Ma;Lu Su;Ziran Wang