DIFFERENTIAL PRIVACY MECHANISMS FOR SMART CITY IOT DATA STREAMS: UTILITY–PRIVACY TRADE-OFFS
Subjects/Theme:
Differential Privacy, Smart Cities, IoT Data Streams, Privacy Preservation, Utility-Privacy Trade-off, Edge Computing, Data Analytics, Real-Time SystemsDescription
Security and Privacy in AI Systems,
Edited By: Dr. Sunita Chaudhary, Dr. Joydeb Patra
ISBN (978-81-685212-9-2)
The proliferation of Internet of Things (IoT) devices in smart cities generates massive real-time data streams that enable intelligent decision-making, resource optimization, and improved urban services. However, such data often contain sensitive personal and behavioral information, raising critical privacy concerns. Differential Privacy (DP) has emerged as a robust mathematical framework for privacy preservation in data analytics. This paper explores various differential privacy mechanisms applied to smart city IoT data streams and analyzes the inherent trade-offs between data utility and privacy. We evaluate Laplace, Gaussian, and randomized response mechanisms under streaming constraints and assess their performance using real-world IoT datasets. Experimental results demonstrate that while stronger privacy guarantees (lower ε) reduce information leakage, they significantly impact data utility and model accuracy. We further propose an adaptive privacy budget allocation strategy to optimize the utility–privacy balance in dynamic smart city environments.
