RFI localization in synthetic aperture interferometric radiometers based on sparse Bayesian inference

Abstract

ABSTRACTThe presence of radio frequency interference (RFI) sources emitting in the L-band, which is reserved for passive measurements by International Telecommunications Union (ITU) regulations, has seriously deteriorated the data quality of many brightness temperature (BT) snapshots in the Soil Moisture and Ocean Salinity (SMOS) project. In order to obviate the Gibbs-like contamination on the BT maps, one effective way is to locate the positions of RFI sources and switch them off. This article discusses a new method for RFI localization that is tailored to the scenario of synthetic aperture interferometric radiometry. The novel aspect lies in addressing the problem of RFI localization from a probabilistic viewpoint. By introducing the sparsity of RFI distribution in the spatial domain as a priori knowledge, we have employed the sparse Bayesian inference (SBI) strategy to estimate the locations of RFI sources. In addition, we have also tested the proposed method using numerical simulations and actual SMOS data. The results indicate that the proposed method has advantages in both accuracy and resolution of RFI source localization over the conventional direction-of-arrival (DOA) methods used in the beamforming technique.