AAU Student Projects - visit Aalborg University's student projects portal
An executive master's programme thesis from Aalborg University
Book cover


Automated Satellite Communication Testbed for Complex Interference Signal Separation Using Real-Time AI-Driven Data: Satellite communication interference mitigation testbed using AI driven data

Translated title

Automated Satellite Communication Testbed for Complex Interference Signal Separation Using Real-Time AI-Driven Data

Term

4. semester

Publication year

2025

Submitted on

Pages

83

Abstract

This master’s thesis develops a flexible, reprogrammable testbed for emulating a satellite communication system, including a ground station, satellite transponder, and interference source. The platform that is presented enables automated measurement campaigns to collect high-quality data for AI-based interference mitigation. The thesis also includes system validation across QPSK, BPSK, and 8PSK modulation schemes, which has an error vector magnitude of 0.22% ±0.005% and modulation error rate up to 51.87%, which indicates and confirms an excellent signal integrity in both low and highly increased interference settings. Furthermore, a U-Net implementation separates the interference from real-time raw (no preprocessing) mixed IQ signals. The U-Net is trained on extensive, large real-world data sets of over 200,000 data samples per gain and frequency settings. The U-Net achieved near reference level denoised signal, reflecting correlation values over 98% for QPSK and BPSK 96% and up to 98.45% for 8PSK, even at the highest interference levels. The results demonstrate robust interference suppression, high reconstruction fidelity, and strong generalization capability. The validated automated testbed and U-Net architecture successfully fulfill the thesis objective of enabling real-time AI-driven signal separation of complex interference in satellite communication links.