Zanella, André Felipe https://orcid.org/0000-0001-8828-6629
Rubrichi, Stefania
Smoreda, Zbigniew
Fiore, Marco
Funding for this research was provided by:
Horizon 2020 (860239)
Comunidad de Madrid (2019-T1/TIC-16037, 2023-5A/TIC-28944)
Agence Nationale de la Recherche (ANR-22-CE25-0016)
Article History
Received: 6 February 2024
Accepted: 29 October 2024
First Online: 7 November 2024
Declarations
:
: Our work builds on mobile network traffic generated by users of a nationwide cellular infrastructure. The traffic measurements used to derive the data set were collected by the operator for network management and research purposes, and temporarily stored within a secure platform at their own premises. The aggregation at the level of the radio access antennas was also carried out in the same platform by personnel of the network operator, in full compliance with Article 89 of the General Data Protection Regulation (GDPR) of the European Commission. The data collection and processing was approved by the Data Protection Officer (DPO) of the operator, and authorized by the French National Commission on Informatics and Liberty (CNIL), within the context of a collaborative research project.We remark that the original network measurements contained personal identifiers (<i>e.g.</i>, the International Mobile Subscriber Identifier, or IMSI) and sensitive data (<i>e.g.</i>, locations of visited antennas, or mobile services consumed) about individual users, and were deleted upon aggregation. Instead, the aggregated data consist of time series of total traffic at the antenna level with a temporal granularity of one hour, and do not contain personal identifiers or sensitive information, such as the device type, preference in terms of application consumption, or trajectories. In addition, the level of spatiotemporal aggregation ensures that no data subject can be re-identified, and that the statistics do not configure as personal data in the GDPR acceptation.The researchers involved in the work presented in this paper only had access to such aggregated and privacy-preserving statistics for the purpose of carrying out the study. Ultimately, our dataset and research do not involve risks for the mobile subscribers.
: The authors declared that there is no conflict of interest.