Time dewarping to reduce nonlinear sampling effects of GPR

The performance of ground penetrating radar (GPR) for subsurface target depth measurement can be directly related to the ability to accurately measure the signal time-of-flight. For impulse GPR systems that utilize the successive sampling technique to acquire radar data, the presence of deterministic timing jitter in the sampling electronics will warp traces along the time axis in a nonlinear manner. Not only does this warping negatively affect subsurface target depth measurements, but it can also reduce the performance of more sophisticated signal processing methods, such as inversion, that rely on accurate wavelet shape. This paper presents an algorithm that can be used to reverse the effects of nonlinear sampling by dewarping the time axis of the GPR data. The algorithm has been tested with a GPR system that exhibits clear symptoms of nonlinear sampling characteristics. The results show that the nonlinear timing effects in GPR data due to deterministic timing jitter can be significantly minimized using this method.