Pre-processing of ground penetrating impulse radar data for improved detection capability

Ground penetrating radar systems have successfully been used to detect buried objects. But, the detection of shallowly buried non metallic objects poses some problems. The ground surface will give rise to a strong reflection, often much stronger than the target reflex. There also often exist cross-coupling between transmit and receive antennas. For shallowly buried objects will these disturbing signals exist in the same time segment as the target signal. This paper presents a method to reduce the effect of such disturbances. Due to the highly transient character of the impulse radars signals must the method use sub-sample resolution time alignment. The method is evaluated on real impulse radar data and detection performances are compared to standard methods such as spatial averaging