Signal processing of ground penetrating radar based on KL transform in wavelet domain

The target reflected waves of the ground penetrating radar(GPR) are usually disturbed by the clutter(or the direct wave) and the stochastic noise and causes the target signal to distinguish with difficulty.In this paper,we study a method based on KL transform in wavelet domain and apply it to reduce the clutter of the GPR.It uses the time-frequency localization characteristic of the wavelet analysis and their correlations on the direct wave,target reflected signals and stochastic noise in multichannel GPR data,we obtain the multi-frequency bands wavelet sections via the wavelet decompose of the GPR trace and suppress the clutter by using KL transform and realize the clutter reduction.The validity of this method has been proved by the experimental data and real data.