The National Science Centre (NCN) has announced the results of the sixth and final edition of the Miniatura 8 competition. A record number of 16 grants went to researchers from WUST, including Dr. Mateusz Mądry from the Department of ICT and Telecommunications. They received a total of over PLN 650,000 for their projects.

The Miniatura competition supports scientific activities leading to the preparation of basic assumptions for research projects that will be submitted in NSC competitions or other national and international competitions. The total budget for this year's edition of the competition amounted to PLN 20 million.

The money can be spent on preliminary and pilot studies, literature studies, as well as research internships and trips. In the last edition of the competition, a total of PLN 13.7 million was awarded to 339 people. This group included sixteen researchers from WUST. 

Dr. Mateusz Mądry (Faculty of Information and Communication Technology)

„Research on simultaneous measurement of temperature and relative humidity using a distributed fibre sensor based on the Rayleigh scattering phenomenon and machine learning" – awarded funding of PLN 14,785.

Distributed fibre sensors based on the Rayleigh scattering phenomenon, using an optical frequency domain reflectometer (OFDR), enable relatively fast measurement along the entire length of the fibre, and with high spatial resolution. They are traditionally used for temperature and strain measurements, but can also be used to measure relative humidity.

Recently, there has been an increase in research on the use of machine learning (ML) models in fibre optic sensors, e.g. in Bragg gratings or loop interferometers. However, there are no reports in the literature on the use of ML models for simultaneous measurement of temperature and relative humidity using a sensor based on the Rayleigh scattering phenomenon.

Therefore, the aim of the project is to assess whether and to what extent ML models will allow simultaneous and precise measurement of these environmental parameters using a sensor based on Rayleigh scattering and OFDR. ML models can reduce data processing time, improve prediction accuracy and minimize the number of measurement points.

The project is expected to investigate the spectral responses of different optical fibres under variable temperature and relative humidity conditions. These data will be used to train selected ML models, and then the analysis will cover their impact on the root mean square error during temperature and relative humidity measurement.