Robust and hyperparameter-free RFI mitigation via ℓ1,2-PLAD and group SPICE

Radio-frequency interference (RFI) poses significant threats to proper ultra-wideband (UWB) radar operations in many practical applications. This paper proposes a novel method for RFI mitigation based on the ℓ
1,2
-penalized least absolute deviation (ℓ
1,2
-PLAD) approach and the group SParse Iterative Covariance-based Estimation (group SPICE) algorithm. The proposed method simultaneously exploits the sparse properties of both RFI spectrum and UWB radar echoes. Unlike the existing sparsity-based RFI suppression methods, such as the robust principal component analysis (RPCA) algorithm, our proposed method is robust and hyperparameter-free, and therefore easy to use in practical applications. Additionally, no other prior knowledge of RFI sources is required in our technique. Results from measured data collected by the U.S. Army Research Laboratory (ARL) UWB synthetic aperture radar (SAR) system are used to demonstrate the robustness and effectiveness of our method.