The Library
Differential privacy-enabled location data sharing solutions for vehicle ecosystems
Tools
Atmaca, Ugur Ilker (2022) Differential privacy-enabled location data sharing solutions for vehicle ecosystems. PhD thesis, University of Warwick.
PDF
WRAP_Theses_Atmaca_2022.pdf - Submitted Version Embargoed item. Restricted access to Repository staff only until 26 September 2025. Contact author directly, specifying your specific needs. - Requires a PDF viewer. Download (11Mb) |
Official URL: http://webcat.warwick.ac.uk/record=b3970608~S15
Abstract
Connected autonomous vehicles (CAVs) are moving from futuristic realms to becoming increasingly encountered in modern society, with the promise of improved safety, efficiency and sustainability. The functions of CAVs rely on data from multiple sources, including vehicular onboard sensors, other vehicles and the infrastructure. However, the public, industry and government are increasingly concerned that data can be revealing, particularly spatio-temporal data, which is sensitive since it can reveal private information about the user, such as habits, health conditions, etc. There are many ways to protect privacy, but differential privacy remains the only mechanism providing mathematical privacy guarantees that enables the quantification of privacy loss. However, when applying such techniques, the affordance of privacy comes at the cost
of utility. This thesis explores the cost of privacy in real-time location data sharing for CAV functions with respect to efficacy and provides novel techniques to minimise this cost. It is composed of three essential components.
The first study focuses on real-time frequency estimation using central and local differentially private data sharing. The output data of this scheme is for a vehicle route planning function. The second study addresses the efficacy challenge raised in the first study regarding the practical application of local differential privacy for a location-based querying function. Both of these works can be considered to offer privacy-preserving data aggregation through a central server. In contrast, the third study involves the development of a federated mechanism to enable collaboration without the need for sharing raw data with a central server. Extensive experimentation provides results that demonstrate the proposed schemes achieve high efficacy for their respective functions and have the potential to guide research for a broader practical deployment of
privacy-enhancing technologies.
Item Type: | Thesis (PhD) | ||||
---|---|---|---|---|---|
Subjects: | T Technology > TL Motor vehicles. Aeronautics. Astronautics | ||||
Library of Congress Subject Headings (LCSH): | Automated vehicles, Intelligent transportation systems, Transportation -- Data processing | ||||
Official Date: | December 2022 | ||||
Dates: |
|
||||
Institution: | University of Warwick | ||||
Theses Department: | Warwick Manufacturing Group | ||||
Thesis Type: | PhD | ||||
Publication Status: | Unpublished | ||||
Supervisor(s)/Advisor: | Maple, Carsten ; Dianati, Mehrdad | ||||
Sponsors: | Republic of Turkey | ||||
Format of File: | |||||
Extent: | xvi, 132 pages : illustrations, charts | ||||
Language: | eng |
Request changes or add full text files to a record
Repository staff actions (login required)
View Item |