Research on Multimodal RFI Cancellation in QTT-PAF Beamforming Based on the PCA Algorithm

Abstract This study tackles the multimodal radio frequency interference (RFI) challenges encountered by the Qitai Radio Telescope by developing a comprehensive numerical model. The model integrates four critical interference types: coupling effects, fixed-position interference, moving-position interference, and white noise. These interferences exhibit dynamic variations in both temporal and spatial domains. Innovatively, this work introduces the principal component analysis (PCA) algorithm to beamforming for the first time, conducting a systematic performance comparison with the conventional subspace projection (SP) algorithm. The interference cancellation capability of PCA is rigorously evaluated under two scenarios: (i) complete spatial separation between signal of interest (SOI) and RFI, and (ii) partial or full spatial overlap between SOI and RFI distributions. Results demonstrate that PCA achieves a 30 dB S/N improvement over SP while maintaining the same robust coupling effects as SP. Crucially, PCA better handles overlapping SOI-RFI scenarios, in which traditional methods exhibit significant performance degradation. This research provides a foundational framework for future beamformer design, offering significant potential to advance RFI mitigation technologies and enhance the observational robustness of radio telescopes in complex electromagnetic environments.