On the Localization of Multisource RFI for Multichannel SAR Based on the Time-Frequency-Domain Least L¹-Norm Method.

Synthetic Aperture Radar (SAR) is easily affected by various radio frequency interference (RFI), which leads to the degradation of image quality. In multi-channel SAR, the position of the RFI source is a prerequisite for most spatial filtering methods to remove interference. For dual-channel SAR systems, the existing RFI source localization methods can only obtain the position of a single jammer, while in a complex electromagnetic environment, there may be multiple RFI sources. To address this problem, a multisource RFI localization method for dual-channel SAR is proposed in this letter. First, the time-frequency analysis is employed to establish the feature mapping of RFI and SAR signals in the time-frequency domain. Then, ridge path detection and recombination algorithms are proposed to separate and extract the received interference signals. Finally, the least L ¹ -norm models are established for each extracted interference signal based on the phase difference of the two channels, and the high precision azimuth position of the RFI sources can be obtained using the interior point method. Simulations indicate that the azimuth localization accuracy of the proposed method can be within 20 m when the Signal-to-Interference Ratio (SIR) is below 0dB. The spaceborne SAR data acquired by Chinese Gaofen-3 (GF-3) also verify the effectiveness of the proposed method.