Implementation of Extended Differential Privacy for Electric Vehicles Using the Novel Filtering Approach

As electric vehicles (EVs) become more integrated into intelligent transportation systems, vast amounts of personal and operational data, such as location, driving patterns, and energy consumption, are continuously collected. Ensuring privacy for this sensitive data is critical to prevent tracking, profiling, and unauthorised access. This paper presents the implementation of event-wise differential privacy (DP) to safeguard individual data points in EV ecosystems, focusing on protecting event-level information like GPS updates and charging events. By utilizing the Laplace mechanism, noise is added to each event to guarantee privacy without compromising overall data utility. Additionally, we introduce a Kalman filter to mitigate the impact of noise, improving the accuracy of post-processed data while preserving privacy. The proposed framework demonstrates how event-wise DP can protect user information while still enabling accurate vehicle operations and analytics. Our approach highlights the balance between privacy and functionality, offering a scalable solution to enhance data protection in future smart mobility infrastructures. This research lays the groundwork for further advancements in privacy-preserving technologies within the EV sector, contributing to safer and more secure data-driven systems. To further alleviate the error in the utility of dataset by mitigating the noisy data generated through event-wise differential privacy, we integrate a Kalman filter into our framework. The Kalman filter is a unique and efficient tool for truncating the noise by correct prediction of the mechanism over time. In this research, it helps mitigate the impact of Laplace noise introduced for privacy preservation, ensuring smooth and comparatively accurate data while maintaining user privacy. In the last section of this paper the comparison of results would be provided before and after the implementation of Event-Wise Differential Privacy and the results obtained by the Kalman Filter for improving the utility keeping a trade-off between error and the usefulness of the dataset.