Equalization of Frequency Domain Adaptive Filter (FDAF) Using Signal Prediction Aided Reference Spectrum Model (SPRSM)

Filtering jamming signals from intermediate frequency (IF) data is a practical anti-jamming solution for low-cost global navigation satellite system (GNSS) receivers, as it does not change antennas or baseband structures. Frequency domain adaptive filter (FDAF) is a commonly used IF anti-jamming filter, which removes jamming components in the frequency domain. However, this filter also filters out normal satellite signals within the jamming bands, resulting in significant signal distortions. These distortions can negatively impact signal quality and positioning accuracy. To address this issue, this study focuses on improving the performance of FDAF by incorporating a signalprediction-aided reference spectrum model (SPRSM) equalizer. Initially, the signal loss caused by FDAF is analyzed. Then, the SPRSM equalizer is developed, which differs from classical equalizers that rely on an empirical normal spectrum. Instead, the SPRSM equalizer utilizes GNSS predictions to compensate for distortions. The effectiveness of the SPRSM equalizer is then theoretically compared to that of the classical equalizer, specifically in terms of carrier-to-noise-density ratio (C/N0) improvement. Real-data experiments are conducted to further evaluate the superiority of the SPRSM equalizer over FDAF and FDAF with the classical equalizer. Specifically, the experimental results show that the SPRSM equalizer can handle narrowband sweep continuous wave interference (SCWI) intensities up to 10 and 6 dB higher than standalone FDAF and the classical equalizer, respectively, while still maintaining a C/N0 over 27 dB•Hz of tracking threshold. Under bandlimited noise (BN) scenarios, the intensities that SPRSM equalizer can handle are up to 6 and 4 dB higher than that of standalone FDAF and the classical equalizer, respectively. Moreover, when subjected to the same jamming attacks, the SPRSM equalizer demonstrates a 11% 60% improvement in horizontal positioning accuracy compared to the classical equalizer.