人工智慧物聯網數據隱私保護與基於本地差分隱私的安全數據蒐集機制
Abstract
This project focuses on developing privacy-preserving mechanisms for Artificial Intelligence of Things (AIoT) systems through Local Differential Privacy (LDP) techniques. The research aims to establish secure data collection protocols that protect individual privacy while maintaining the utility of collected data for AI model training and inference. Key areas of investigation include noise injection algorithms, privacy budget allocation strategies, and utility-privacy trade-off optimization in distributed IoT environments. The proposed framework will enable AIoT systems to collect sensitive data with strong privacy guarantees, supporting applications in smart cities, healthcare monitoring, and industrial automation while complying with privacy regulations.