Narrowband Internet of Things Localization

This thesis focuses on the development and testing of techniques for localising Narrowband Internet of Things (NB-IoT) devices, by leveraging machine and deep learning techniques on empirical measurements

The NB-IoT technology is a leading solution in the context of Low Power Wide Area Networks (LPWANs), and exploits the cellular infrastructure to enable several IoT services, including smart cities, industrial automation, logistics tracking, and wearables.

Many of these use cases require or benefit from location information, making localisation a key dimension for NB-IoT. However, the widely used global navigation satellite system (GNSS) is not suitable, due to a) further power consumption and costs needed to support GNSS chips, and b) challenging locations in which many NB-IoT devices may be deployed (e.g., deep indoor), where GNSS systems cannot operate. Alternatives enabling reliable localisation via direct exploitation of NB-IoT signals are thus needed, as highlighted by current investigations in the research community.


The goal is to analytically and experimentally address the challenges in the localisation of NB-IoT devices, deriving accurate localisation algorithms. The latter will be tested on large datasets of NB-IoT traces collected in the cities of Oslo and Rome.

Learning outcome

Knowledge of IoT technologies and localisation techniques, practical experience with real devices and experimental traces, data-driven and Machine Learning (ML)-based result analysis.


Wireless communications and IoT fundamentals, Machine Learning fundamentals, Matlab and/or Python programming, knowledge of data processing techniques.


Collaboration partners

Sapienza University of Rome

Contact person