Project description
Device fingerprinting is used to establish the identity of the transmitter. In literature, various software-based fingerprints are established (firmware versions, MAC address, timing information, etc.) to identify the source of the transmission. These software fingerprints can be modified and compromised. Unlike software identity, radiometric fingerprinting is due to imperfections in the transmitter's hardware (PA, modulator, oscillators, antenna). These fingerprints can have features of frequency offset, phase offset, transient envelope, modulation imperfections, or AGC parameters. A large dataset of radio fingerprints is available. The aim of the project is to apply machine learning and AI algorithms to efficiently identify transmitters. The research questions related to this research theme are- How to develop an efficient method to classify transmitters based on its radiometric fingerprint?
- Can we use AI-based algorithms for feature selection to enhance transmitter identification?
- How can adversarial networks be used to attack radiometric fingerprinting?
Note: You need to register interest in projects from different supervisors (not a number of projects with the one supervisor).
You must also contact each supervisor directly to discuss both the project details and your suitability to undertake the project.