Advanced methods for sub surface object detection with emphasis on data acquisition and interpretation

The detection of objects below the surface is the most important requirement for various civil and military applications (including the detection of land mines). Ground penetrating radar (GPR), which detects electrical discontinuities in the shallow subsurface, is normally used for space-time recording, the radargram. However, conventional radargrams are generated by moving the antenna directly above the Earth's surface and are quite rigid due to direct contact with the surface and the inability to view the object from different sides. This project presents basic research to develop a small radar system for object detection using advanced radar concepts and implementing advanced imaging concepts using multiple views of the examined area. This research project will be focused on: developing concepts of radars, broadband antenna design, low-noise signal generation, dynamical background suppression, bistatic radar synchronization, developing advanced imaging concepts and extract features using convolutional neural networks (CNNs). Ground-penetrating radar (GPR) consists of a transmitter, antennas, and a receiver. All three components require careful design to maximize the output power of the transmitter and adapt the transmitted signal to the radiating element, the antenna. In this research project, we will develop a pulse-based radar and a stepped frequency continuous wave (SFCW) radar using on-board digital processing with a field programmable gate array (FPGA) enabling us oboard processing and further noise reduction. The developed radars will be analyzed deeply to reduce phase noise and improve radar characteristics. The novelty in radar design will be on board dynamical background removal, which replaces automatic gain control and enables much higher output power of the transmitted signal without saturating the nearby receiver and removing background targets. We will develop a bistatic radar for use on two unmanned aerial vehicles (UAVs) used for advanced imaging. The bistatic radars will be synchronized using global navigation satellite systems (GNSS) and internal clocks for spatial and time synchronization. The time and spatial synchronization is needed to ensure advanced imaging techniques, where correct position of UAVs will be determined by controlling and managing UAV flight path. The flight control is needed to enable Synthetic Aperture Radar acquisition and tomography. To use microwave tomography, multiple views of the examined subsurface will be provided by multiple UAV, carrying transmitters and receive sensors to perform measurements of the scattered signal. The Artificial Intelligence will be used to detect objects within the polarimetric data. We will use the developed GPR for several scenarios: subsurface object detection, polarimetric dielectric constant estimation using CNN approach which will enable soil moisture detection and micro Doppler signatures detection for human detection. The objectives of the proposal are: ·To develop a small size GPR that is convenient for attaching to small vehicles or UAV. ·To analyze and model radar system and reduce phase noise of the transmitter and receiver. ·To propose on board dynamical background removal. ·To propose synchronization techniques for bistatic radar. ·To design air coupled polarimetric antennas suitable for UAV. ·To propose optimal flight configuration of polarimetric bistatic radars. ·To develop advanced SAR and microwave tomography imaging techniques using the multi-monostatic mode (single drone) or multi-bistatic mode (moving receive sensor). ·To use Artificial Intelligence to detect and classify buried objects. ·To extract features from GPR data for object detection, through the wall imaging for human activity detection using bistatic radar and soil moisture estimation using GPR data.