Over the past few decades, few technological advances have transformed daily life as significantly as mobile phones and their applications. These devices have evolved into essential portable personal computers, enabling communication, information access, and entertainment from virtually any location due to their internet connectivity. Achieving today’s high-speed, low-latency mobile services required decades of research and development. Early mobile networks supported only voice calls and text messages, but current fourth and fifth generations of mobile networks facilitate high-definition video streaming and cloud applications. The proliferation of smartphones and mobile apps allows any company or developer to offer their products to consumers anywhere, further embedding these devices into daily routines.

Improvements in mobile networks have encouraged users to rely more on their smartphones for various tasks, reflecting the integration of mobile phones into daily life. Data collected by mobile network operators now provides a detailed view of user behaviors and movements, making it a crucial resource for research. This data assists operators in enhancing network performance and offers insights into population dynamics and urban environments. This has led to the emergence of networks data science, a field focused on managing large-scale data, ensuring its quality, and extracting meaningful insights. The thesis at hand contributes to this field by developing tools and methodologies for analyzing mobile network data and demonstrating its potential in diverse research domains.

The first part of the thesis contextualizes networks data science, discussing the challenges of managing large datasets and the various fields that have benefited from this research, such as network engineering, population studies, and urban planning. It also emphasizes the importance of established routines for data collection and processing.

The second part presents original contributions, exploring different applications of mobile network data and including: the analysis of how the adoption of new technologies by mobile operators influences traffic patterns; how mobile consumption patterns can be used to derive insights into urban spaces and specific locations within cities; and investigation of the impact of special events on mobile networks and how these events affect user behavior and network performance; a characterization of session-level measurements to create synthetic datasets for research purposes, facilitating the testing and validation of datadriven solutions.

In summary, the extensive data generated by mobile networks offers numerous opportunities for innovation in research and technology. This thesis provides an overview of the current state of mobile network data science and explores potential future research directions.

About André Zanella

André Zanella is a PhD researcher at IMDEA Networks Institute. He received his bachelor’s and master’s degrees in Telecommunications Engineering at the Federal University of Parana, in Brazil, where he previously worked on data modeling of microelectronic RF systems. He’s currently a part of the Networks Data Science group, working on mobile traffic analysis and remote sensing with network metadata, where his interests lay in developing techniques that help solving social sciences problems using information gathered by network operators.

Supervisor de tesis: Dr. Marco Fiore, IMDEA Networks Institute, España

Universidad: Universidad Carlos III de Madrid, España

Programa de doctorado: Ingeniería Telemática

Miembros del tribunal:

• Presidente: Esteban Moro, Full Professor of Physics, Northeastern University, USA
• Secretario: Andra Lutu, Senior Researcher, Telefonica Innovacion Digital, Spain
• Miembros del panel: Iñaki Úcar, Assistant Professor of Statistics, Universidad Carlos III de Madrid, Spain