GPR Data Processing Using the Component-Separation Methods PCA and ICA

Abstract

This paper illustrates clutter reduction in stepped- frequency ground penetrating radar (SFGPR) data for anti- personal landmines detection. For the purpose of clutter reduc- tion, two subspace projection techniques have been studied and applied to experimental data, namely the principle component analysis (PCA) and the independent component analysis (ICA). Their output SNR have also been compared. These two algorithms have been applied and compared for experimental data set with non-metallic AP landmines. The experimental data were collected by using an SFGPR operating on the frequency range from 1 GHz to 20 GHz.