Location and Data Privacy Preservation in Intelligent Systems



Journal Title

Journal ISSN

Volume Title



Due to the ubiquitous mobile devices with embedded sensors and connectivity over the internet, the Internet of things (IoT) has evolved. The IoT brings the explosive growth of devices connected and controlled by the Internet. The enormous collection of connected sensors and devices makes a significant contribution to the volume of data collected, which brings us to the big data era. Intelligent system (IS) becomes an emerging paradigm for integrating big data, analytics, privacy, and artificial intelligence. The IS is any formal or informal system to manage data gathering, to obtain and process the data, to interpret the data, and to provide reasoned judgments to decision makers as a basis for action. In order to keep up with the continuous influx of data, machine learning is one of the best solutions for big data analysis, which is fast evolving during the last decade. With the development of machine learning technologies, it plays a critical role in IS. The IS, which integrates computations, communications and decision making, interacts with humans through many new modalities. However, privacy is an essential concern in IS since a large volume of users’ daily and sensitive data is used in constituting systems, and users become increasingly concerned about the compromise of their personal information. Therefore, it is necessary to develop innovative privacy preserving approaches to prevent users' confidential information from illegal revealing while efficiently utilizing massive data generated from users. In fact, there are trade-offs between the effectiveness of privacy protection and the convenience of data collection, communications, and energy consumption, which need proper considerations in system designs. The objectives of this dissertation are to develop efficient and reliable data analysis methods in various IS applications and protect the data privacy against malicious attacks through a combination of theoretical, simulation, and experimental studies. Given the challenge of privacy preservation and reliable data analysis, this work endeavors to develop a series of privacy preserving data analytic and processing methodologies through machine learning, optimization and differential privacy; and focuses on effectively integrating the data analysis and data privacy preservation techniques to provide the most desirable solutions for the state-of-the-art IS with various application-specific requirements.



Differential Privacy, Intelligent Systems


Portions of this document appear in: Wu, Maoqiang, Xinyue Zhang, Jiahao Ding, Hien Nguyen, Rong Yu, Miao Pan, and Stephen T. Wong. "Evaluation of inference attack models for deep learning on medical data." arXiv preprint arXiv:2011.00177 (2020); and in: Zhang, Xinyue, Jiahao Ding, Maoqiang Wu, Stephen TC Wong, Hien Van Nguyen, and Miao Pan. "Adaptive privacy preserving deep learning algorithms for medical data." In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pp. 1169-1178. 2021; and in: Zhang, Xinyue, Jiahao Ding, Sai Mounika Errapotu, Xiaoxia Huang, Pan Li, and Miao Pan. "Differentially private functional mechanism for generative adversarial networks." In 2019 IEEE Global Communications Conference (GLOBECOM), pp. 1-6. IEEE, 2019; and in: Zhang, Xinyue, Jiahao Ding, Xuanheng Li, Tingting Yang, Jie Wang, and Miao Pan. "Mobile crowdsensing task allocation optimization with differentially private location privacy." In ICC 2020-2020 IEEE International Conference on Communications (ICC), pp. 1-6. IEEE, 2020.