Use this URL to cite or link to this record in EThOS:
Title: Signal processing strategies for ground-penetrating radar
Author: Jiang, Wei
ISNI:       0000 0004 2706 7467
Awarding Body: University of Bath
Current Institution: University of Bath
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Access from Institution:
Interpretation of ground penetrating radar (GPR) signals can be a key point in the overall operability of a GPR system. In stepped-frequency and Frequency-Modulated Continuous-Wave (FMCW)GPR systems in particular, the target or object of interest is often located by analysis of Fast Fourier Transform (FFT) derived data. Increasing the GPR system bandwidth can improve resolution, but at the cost of reduced penetrating depth. The challenge is to develop high-resolution signal processing strategies for GPR.A number of Fourier based methods are investigated. However, the main response over a target's position can make it difficult to recognise closely spaced targets. The Least-Suare method is found to be the best autoregression-based estimator. However the method requires high Signal-to-Noise ratio to achieve high- resolution. Furthermore a number of subspace-based methods are investigated. Although the MUItiple Signal Classification (MUSIC) method can theoretically offer infinite resolution, they must be seeded with the number of targets actually present. A superimposed MUSIC technique is proposed to suppress false targets. A novel windowed MUSIC (W-MUSIC) algorithm is developed, and it offers high resolution while still able to minimise spurious responses. Since the performance of any FMCW GPR is critically linked to the linearity of the sweep frequency, the non-linearity in the target range estimation is studied. A Novel Short-Time MUSIC method is proposed and higher time and frequency resolution is achieved than the conventional Short-Time Fourier Transform method. In addition a modified Adaptive Sampling method is proposed to solve the non-linear problem by utilising a reference channel in a GPR system.
Supervisor: Shepherd, Peter Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: IMCW ; time-series ; Spectral Analysis ; GPR ; Signal Processing