NASA-ISRO SAR (NISAR) Mission is equipped with L- and S-band Synthetic Aperture Radar (SAR) to produce high-quality imagery products to detect changes on earth surface, and to estimate and monitor different geophysical quantities such as soil moisture and water surface extent. Both L-band and S-band are susceptible to Radio Frequency Interference (RFI) in many geographic regions. One of the main sources of RFI are powerful electromagnetic signals from ground-based radar and communication platforms. In addition, there are unregulated electromagnetic emitters that could also operate in NISAR passband. RFI observed using existing L-band acquisitions, e.g., from ALOS PALSAR, appear to be modulated wideband and narrow band signals in the range frequency spectrum. RFI often causes haze-like image artifacts on focused L-band SAR images. [1]. In addition, RFI degrades the quality of the Single Look Complex ( SLC) images and interferometric coherence, and can bias the estimation of physical quantities such as ionospheric delay or soil moisture from SAR images. Hence effective RFI detection and mitigation algorithms are desired to address RFI contamination in NISAR products. One of the RFI mitigation algorithms developed is Slow-Time Eigenvalue Decomposition (ST-EVD) which is a Principal Component (PC) based approach that removes RFI Eigenvalues through projection. An adaptive data-driven thresholding algorithm, Slow-Time Eigenvalue Slope Thresholding (ST-EST) is being developed to detect RFI contamination severity and provide ST-EVD with mitigation thresholds.
