Accurate image expansion method for target buried in dielectric medium using multi-static UWB radar

Ultra wideband (UWB) pulse radar has high dielectric permeability and range resolution, and it is thus promising for use in non-invasive imaging applications, such as non-destructive testing of pipes buried in soil or in a concrete wall, or tumor detection in early-stage breast cancer. Various imaging methods are available for near-field UWB radar systems, including synthetic aperture radar (SAR) [1], beam forming based imaging, time reversal focusing, and numerical solution of the domain integral equation with multi-dimensional optimization. However, none of these methods accomplish a good performance balance between the amount of computation required and the desired accuracy or spatial resolution. As a solution to this difficulty, we have already proposed a fast and accurate imaging method for targets buried in a dielectric medium based on the advanced principle of the range points migration (RPM) algorithm, which performs accurate surface extraction for targets in free space by using the group mapping from the range points (a set of antenna locations and observed ranges) to target surface points.However, the RPM method assumes a mono-static configuration in the observation event, and often barely reconstructs the whole of the target boundary, particularly for a dielectric medium with a random surface. This then poses a difficulty in distinguishing the exact shape of an internal target, which could be critical in applications such as discrimination of a deformed pipe in a non-destructive testing application. To enhance the imaging area of an internal target, this paper introduces a novel method using multi-static observation, where the principle of the method in is extended appropriately to a single-transmitting and multiple-receiving model. This method can enhance the instantaneous aperture size, which makes it possible to obtain the reflection echo from a wider part of the target boundary. The results of numerical simulations show that the proposed method accur- tely expands the imaging region of the internal target compared with that obtained using the mono-static based method.