GEOTEC members presented a couple of papers in the 13th International Conference on Indoor Positioning and Indoor Navigation (IPIN 2023), hosted by the Fraunhofer Institute for Integrated Circuits in Nuremberg, Germany during the last week of September.

Our colleague Miguel Matey Sanz presented a paper entitled “Temporal Stability on Human Activity Recognition based on Wi-Fi CSI“, with the collaboration of Joaquín Torres-Sospedra and Adriano Moreira. This paper was the outcome of Miguel’s international research stay at Universidade do Minho in Guimarães (Portugal), during the past spring.

In short, the work investigates the usage of Wi-Fi Channel State Information data collected from an ESP32-S2 for localized human activity recognition using machine learning. The experiments showed a clear degradation in the classification accuracy of the model over time.

The second paper was co-authored by Tomas Bravenec, Joaquín Torres-Sospedra, Michael Gould and Tomas Fryza. Its title is “UJI Probes: Dataset of Wi-Fi Probe Requests“, and is a direct result of the Tom’s PhD research project. The paper is indeed a data paper, focusing on the creation of a new, publicly available Wi-Fi probe request dataset.