GOEASY submission with title “Privacy-Aware Information Base in the Context of Smart Cities” for W3C Smart Cities Workshop is accepted for presentation on 25th June 2021 as a way to bring specific perspective on the topic of standardization of geolocation modelling vis-a-vis privacy. Here is the abstract (courtesy of Dr. Sisay Chala – Fraunhofer FIT):
Developing flexible and extensible mass-market application that employ signals from multi-constellation of satellites in order to provide dependable location-based services (LBS) requires adoption of a component-based architectural style where the system functions are provided by a set of well-defined, self-contained, modules named “components”, which communicate to each other through well-defined interfaces to make sure that they are decoupled from each other.
In this work, we present the implementation of a modular component-based privacy-aware information base, developed under EU-funded GOEASY project, that utilizes Galileo authentication to enhance security of location services. Among the many components is one that deals with ensuring privacy of users that provide location tracking data. Because it is important to guarantee that the user’s Personally Identifying Information (PII) should be protected in order to motivate users to provide data about their mobility behavior. This mobility behavior makes a key input to optimize public transportation services thereby reducing traffic, impact on environment, and cost. Mobility behaviors provide modes of transportation, points of interchanging mode of transportation, duration, and time of arrival/departure to/from a specific location such as public train/bus station or P+R lots.
GOEASY’s privacy-aware information base implementation hides the journey track id from the external services to protect the recovery of the data subject’s identity. We want to send data (that cannot be used to fetch info from the database in the GOEASY platform) to external services such as third-party mobility mode detection systems. Furthermore, to protect breach of privacy of anonymized data through analysis of patterns, e.g., k-anonymity, we implemented data minimization and differential privacy through addition of noises into the tracks. However, where to add the noise, how to decide the volume of noise to add, and whether these parameters should be constant or randomized remains a relevant challenge for standardization.