MSc thesis project proposal

[2024] AI4RF: Artificial Intelligence for 6G RF Signal Processing


As the world eagerly anticipates the advent of 6G technology, there is a growing need to address the complexities and challenges in radio frequency (RF) signal processing. This project seeks to harness the power of artificial intelligence to revolutionize the way RF signals are processed, specifically targeting improvements in PAPR compression and digital predistortion.

Project Aims and Objectives

The primary goal of this thesis is to develop AI-based methodologies for enhancing 6G RF signal processing. The project will focus on two main aspects:

  1. PAPR Compression: Investigating AI techniques to reduce the peak-to-average power ratio in RF signals is crucial for efficient power usage and minimizing signal distortion in wireless communications.
  2. Digital Predistortion: Applying AI algorithms for digital predistortion to counteract the effects of nonlinearities in RF transmitters, thereby improving signal integrity and transmission quality.

    Opportunity for Advanced Research

    This Master's thesis not only presents an opportunity to be at the forefront of 6G technology but also opens avenues for further research, potentially leading to a Ph.D. project at ELCA.


    Research and Methodology

    The project will involve:

    • Literature Review: Conducting a comprehensive review of existing AI techniques used in RF signal processing and identifying potential areas of innovation.
    • Algorithm Development: Design and develop AI algorithms tailored for PAPR compression and digital predistortion. This could include neural networks, machine learning models, or other AI methodologies.
    • Simulation and Analysis: Testing the algorithms through simulations, analyzing their effectiveness in improving RF signal processing, and comparing them with traditional methods.

    Expected Outcomes

    • The successful implementation of AI in RF signal processing for 6G is expected to yield:
    • Improved efficiency and reliability in RF signal transmission.
    • Enhanced bandwidth utilization and reduced power consumption.


    Ideal Candidates should Have:

    1. Background in wireless communications or signal processing.
    2. Familiar with neural networks
    3. Proficiency in programming languages (e.g., Python) and deep learning frameworks (e.g., PyTorch).


    dr. Chang Gao

    Electronic Circuits and Architectures Group

    Department of Microelectronics

    Last modified: 2023-11-12