Signal processing techniques for clutter parameters estimation and clutter removal in GPR data for landmine detection

Ground penetrating radar (GPR) has become widely accepted as a major technique for subsurface investigations, mainly in civil engineering. Considerable efforts are put in the development of GPR systems for the detection of shallow buried landmines. However, the GPR performs inadequately due to clutter, which dominates the data and obscures the mine information. The clutter varies with surface roughness and soil conditions and lead to uncertainty in the measurements. It is therefore necessary to overcome these surrounding effects when processing GPR data for detecting small, shallow buried objects. We present improved signal processing techniques which can be used to reduce the clutter through data preprocessing. Several approaches are proposed for the GPR clutter reduction techniques, most of them model the clutter statistically. The proposed clutter reduction technique models the clutter using parametric modeling. The clutter contained in the measurements is treated as an ARMA model. The advantage of such approach lies therein that once the clutter is satisfactorily known, any target will show up as a small anomaly in against the known clutter background. This method suggests that the clutter shows a certain amount of correlation. Experimentally it is shown that the dominant interference in GPR data is correlated clutter, i.e., interference, which has a large correlation coefficient for tags greater than zero. However, the clutter environment cannot be considered completely stationary. An ideal filter would then be an adaptive filter, which estimates the slowly varying local clutter parameters, all the time ignoring the small parameter jumps caused by the buried targets to be detected. Kalman filtering is used for the estimation of the clutter parameters in the presence of random noise, detects jumps that occur at unknown points in time, and provides estimates of the new parameter values, without altering the target return