University of Twente Student Theses
Identification of wireless devices in the 2.4GHz band using machine learning
Vink, L.W.A. (2024) Identification of wireless devices in the 2.4GHz band using machine learning.
|
PDF
7MB |
| Abstract: | With the increasing popularity of wireless devices, it has become increasingly important to monitor the electromagnetic spectrum and enforce rules on the use of it. The radio communications agency of the Netherlands, Rijksinspectie Digitale Infrastructuur, is responsible for monitoring the frequency spectrum and ensuring its proper use. Processing the data measured by the measurement setups spread throughout the Netherlands is labour-intensive which is why there are machine-learning models in place for processing data to detect outlandish signals, reducing the need for human intervention. Further analysis of the 2.4GHz unlicensed band is requested, with focus on detecting specific devices transmitting at this frequency. This can later be used for statistical purposes or to enforce regulations. The analysis will be done using supervised machine learning due to its proven effectiveness and faster data processing capabilities compared to traditional signal processing methods. This paper explores the use of the Random Forest model and the Gradient Boosted Trees model for detecting specific devices in the 2.4 GHz unlicensed band. First, a dataset will be made, consisting of spectrograms of WiFi, Bluetooth, and Fixed carrier (Babyphone) signals. The dataset is later expanded upon with several combinations of the three signals. Next, a Random Forest and Gradient Boosted Trees model is trained, compared, and shown to be effective. After the expansion of the dataset, the models are retrained and shown to be working. Finally, a real-world data sample is input to both models where it is shown that they correctly identify the specific signal. The best-performing algorithm is the Gradient Boosted Trees model, achieving an accuracy of 97.3%. |
| Item Type: | Essay (Bachelor) |
| Faculty: | EEMCS: Electrical Engineering, Mathematics and Computer Science |
| Subject: | 53 electrotechnology, 54 computer science |
| Programme: | Electrical Engineering BSc (56953) |
| Link to this item: | https://purl.utwente.nl/essays/101673 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page