Detecting Satellites in Radio-Frequency Data via Semi-Supervised Learning

The paper presents a semi-supervised learning workflow for detecting and classifying satellites in radio-frequency (RF) data, addressing challenges posed by sparse and variable datasets. The method combines Non-negative Matrix Factorization with automatic model determination (NMFk) to identify latent patterns in unlabeled RF observations, which are then interpreted by subject-matter experts to classify events. This approach reduces the dependency on large labeled datasets, enabling more efficient and interpretable satellite monitoring in changing RF conditions, which is crucial for space domain awareness.