Date of Award

May 2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical Engineering

First Advisor

Ryan Adams

Abstract

This dissertation presents a significant advancement in Ground Penetrating Radar (GPR) detection technology through the development of a novel method for the detection and classification of buried conductive objects using custom-designed Archimedean spiral antennas. The research demonstrates that the transient electromagnetic response of an Archimedean spiral antenna is strongly correlated with the Radar Cross Section (RCS) characteristics of buried objects operating within the radar resonance scattering, or Mie, region. By leveraging this relationship, the proposed approach enables not only the detection of buried objects, but also generalized classification based on object geometry and size.A primary contribution of this work is the demonstration that the dominant Mie frequency response of conductive spheres can be accurately determined from the antenna’s transient time-domain response. The research further establishes that the temporal location of the Mie response within the received signal can also be predicted from the transient data, providing complementary frequency and time-domain information that significantly improves object identification confidence. Through modeling, simulation, and experimental validation in both over-air and in-sand bistatic GPR configurations, the research confirms that Mie-region scattering principles—traditionally associated with monostatic far-field radar systems are equally applicable to bistatic GPR.

A second major contribution is the extension of this transient-response based classification methodology to conductive cylindrical objects and inert ordnance projectiles. Unlike spheres, cylindrical objects were shown to produce multiple transient peaks within the Mie scattering region when excited by an Archimedean spiral antenna. These peaks exhibit consistent temporal spacing that correlates directly with the expected RCS behavior of cylindrical geometries, thereby providing a distinctive and reliable signature for cylindrical object classification. This work demonstrates that transient response characteristics generated by the spiral antenna geometry can be used to reliability and unambiguously differentiate buried object shapes.The dissertation also contributes a comparative analysis between Archimedean spiral antennas and conventional bowtie antennas commonly used in GPR systems. Both antenna types were optimized for operation across the 250 MHz to 2 GHz frequency range. The results demonstrate that the polarization and transient response characteristics of the Archimedean spiral antenna provide superior object classification capability relative to the bowtie antenna, highlighting the value of spiral antenna technology for advanced GPR sensing applications.

An additional contribution includes the development of antenna optimization techniques that improve low-frequency performance and radiation patterns while maintaining reduced antenna dimensions. These improvements were achieved through impedance matching, noise management and beam steering methods which were applied to the spiral and bowtie antenna designs. To further enhance operational flexibility, a computer-controlled impedance matching circuit was developed using semiconductor switching technology to enable selectable wideband and impedance-matched narrowband operation, allowing antennas to operate efficiently at frequencies below their conventional operating range.

Collectively, this research establishes a new transient-response-based framework for buried object detection and classification using bistatic GPR technology. The work advances the theoretical understanding of Mie scattering behavior in near-field GPR applications while introducing practical antenna and impedance-matching innovations that improve detection confidence, low-frequency performance, and suitability for autonomous and mobile GPR platforms.

Download

Share

COinS