Artificial Intelligence-Machine Learning based interference rejection for electric-vehicles


Interference is a common problem in wireless communications and is becoming more severe as the wireless spectrum becomes more crowded. Conventional interference rejection methods focus on radiated interferences, which originates from dedicated wireless transmitters. Signals and propagation models of the interference-signals were studied, and interference rejection methods were developed using these models.


Recently, due to the fast development of electric-vehicles, a new type of interference signals, namely the so called; “un-radiated-interference”, was discovered. For example, it is observed that DC/DC-converters in electric-cars introduce significant interference to radio receivers from the low-frequency (around 1 MHz) Amplitude Modulation (AM) signal-band all the way up to the Digital Audio Broadcasting (DAB) band (around 200 MHz). This new type of interference-signals differ from the conventional interference in the following ways:

  • they are not radiated by dedicated wireless transmitters, hence, “un-radiated”,
  • they are very wide-varying in time, frequency and also among, in this case, different car-models,
  • statistical or deterministic modeling of these signals and their propagation is not yet available and to be investigated what is possible.

Artificial Intelligence(AI)-Machine Learning (ML) based methods seem to be very suitable for this type of interference-signals. Therefore, as a first step in this project, your assignments are the following:

  • building a measurement setup, using an available IQ recorder, to record the “un-radiated interferences”,
  • obtaining and collecting a robust database of these interference-signals,
  • labeling (fragments of) the database on the characteristics, i.e., time, frequency, spatial, and statistical characters, of the recorded interference-signals.


This project is in collaboration with NXP semiconductors and the student is-expected-to/might spend time in both the company and the university. This project can also be further extended to a PhD thesis work, where the student can follow up with;

  • analyzing and characterizing the recorded interference-signals in time-, frequency-, spatial-domain, and their statistical behavior,
  • developing and implementing emulation-models to augment the interference-signal database for more test-scenarios,
  • developing and implementing AI-ML interference rejection-algorithms.

This master project is also related to these two PhD positions.