Staff View: May the privacy be with us: Correlated differential privacy in location data for ITS

May the privacy be with us: Correlated differential privacy in location data for ITS

With the development of Intelligent Transportation Systems (ITS), a vast amount of location data is being generated from various IoT devices equipped with location positioning sensors. Preserving the privacy of location data release is a critical concern, as the publication of aggregated data often...

Full description

Saved in:

Bibliographic Details
Main Authors:	Chong, Kah Meng, Malip, Amizah
Format:	Article
Published:	Elsevier 2024
Subjects:	QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering
Online Access:	http://eprints.um.edu.my/44740/ https://doi.org/10.1016/j.comnet.2024.110214
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my.um.eprints.44740
record_format	eprints
spelling	my.um.eprints.447402024-07-11T01:22:45Z http://eprints.um.edu.my/44740/ May the privacy be with us: Correlated differential privacy in location data for ITS Chong, Kah Meng Malip, Amizah QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering With the development of Intelligent Transportation Systems (ITS), a vast amount of location data is being generated from various IoT devices equipped with location positioning sensors. Preserving the privacy of location data release is a critical concern, as the publication of aggregated data often reveals private information about the users. Differential Privacy (DP) has recently emerged as a robust framework to guarantee privacy in this context. However, conventional DP mechanisms commonly make no assumption about the distribution of the input data, which could lead to unexpected privacy leakage if the data are correlated. In this paper, we investigate the complex simultaneous impact of user correlation, spatial–temporal correlation and prior knowledge of an adversary on the privacy leakage of a DP mechanism, which has not been addressed in prior work. We derive several closed-form expressions that demonstrate and quantify the privacy leakage under correlated location data, followed by the design of efficient algorithms to compute such privacy leakage. Then, we propose a Δ-CDP (Correlated Differential Privacy) to provide a formal privacy guarantee against the additional privacy leakage incurred by these factors. Extensive comparisons, theoretical analysis, and experimental simulations are presented to validate the correctness and efficiency of the proposed work. © 2024 Elsevier B.V. Elsevier 2024-03 Article PeerReviewed Chong, Kah Meng and Malip, Amizah (2024) May the privacy be with us: Correlated differential privacy in location data for ITS. Computer Networks, 241. ISSN 1389-1286, DOI https://doi.org/10.1016/j.comnet.2024.110214 <https://doi.org/10.1016/j.comnet.2024.110214>. https://doi.org/10.1016/j.comnet.2024.110214 10.1016/j.comnet.2024.110214
institution	Universiti Malaya
building	UM Library
collection	Institutional Repository
continent	Asia
country	Malaysia
content_provider	Universiti Malaya
content_source	UM Research Repository
url_provider	http://eprints.um.edu.my/
topic	QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering
spellingShingle	QA75 Electronic computers. Computer science TK Electrical engineering. Electronics Nuclear engineering Chong, Kah Meng Malip, Amizah May the privacy be with us: Correlated differential privacy in location data for ITS
description	With the development of Intelligent Transportation Systems (ITS), a vast amount of location data is being generated from various IoT devices equipped with location positioning sensors. Preserving the privacy of location data release is a critical concern, as the publication of aggregated data often reveals private information about the users. Differential Privacy (DP) has recently emerged as a robust framework to guarantee privacy in this context. However, conventional DP mechanisms commonly make no assumption about the distribution of the input data, which could lead to unexpected privacy leakage if the data are correlated. In this paper, we investigate the complex simultaneous impact of user correlation, spatial–temporal correlation and prior knowledge of an adversary on the privacy leakage of a DP mechanism, which has not been addressed in prior work. We derive several closed-form expressions that demonstrate and quantify the privacy leakage under correlated location data, followed by the design of efficient algorithms to compute such privacy leakage. Then, we propose a Δ-CDP (Correlated Differential Privacy) to provide a formal privacy guarantee against the additional privacy leakage incurred by these factors. Extensive comparisons, theoretical analysis, and experimental simulations are presented to validate the correctness and efficiency of the proposed work. © 2024 Elsevier B.V.
format	Article
author	Chong, Kah Meng Malip, Amizah
author_facet	Chong, Kah Meng Malip, Amizah
author_sort	Chong, Kah Meng
title	May the privacy be with us: Correlated differential privacy in location data for ITS
title_short	May the privacy be with us: Correlated differential privacy in location data for ITS
title_full	May the privacy be with us: Correlated differential privacy in location data for ITS
title_fullStr	May the privacy be with us: Correlated differential privacy in location data for ITS
title_full_unstemmed	May the privacy be with us: Correlated differential privacy in location data for ITS
title_sort	may the privacy be with us: correlated differential privacy in location data for its
publisher	Elsevier
publishDate	2024
url	http://eprints.um.edu.my/44740/ https://doi.org/10.1016/j.comnet.2024.110214
_version_	1805881164289802240
score	13.211869

May the privacy be with us: Correlated differential privacy in location data for ITS

Similar Items