Title: NERF2-localzation
DNr: Berzelius-2024-40
Project Type: LiU Berzelius
Principal Investigator: Iliar Rabet <>
Affiliation: Mälardalens universitet
Duration: 2024-01-26 – 2024-06-01
Classification: 20203


With the recent developments in the area of generative AI, it is now possible to synthesize RF signals based on 3D visual models and realistic RF datasets. This project aims at studying the application of generative AI for predicting RF signals. Although significant works have been conducted for some wireless technologies such as cellular and BLE, there is a gap regarding applying Gen-AI methods to model specific radio signals such as IEEE 802.15.4. We will apply some existing methods on our collected data on low-power 802.15.4 devices and introduce improvements to improve the accuracy of the results.