A novel TDR signal processing technique for measuring apparent dielectric spectrum

Abstract

Conventional time-domain reflectometry (TDR) signal interpretation generated single-valued apparent dielectric constant, while sophisticated, inconvenient measurement system and analysis is needed for measuring spectral complex dielectric permittivity (CDP). Niching between these approaches, a novel phase velocity analysis (PVA) method is developed to efficiently measure apparent dielectric spectrum (ADS) directly from TDR signals in a simple, quick, model-free, and inversion-free manner. The proposed PVA method extracts the two reflections from the top and end of the sensing probe by proper window selection and calculates their phase shift at each frequency, from which the phase velocity and corresponding apparent dielectric constant can be determined. Numerical and experimental results demonstrated that PVA is capable of measuring ADS in a frequency band, typically from 100 MHz-1 GHz. The effect of signal truncation was identified as the main cause of poor results outside the effective frequency band. Factors that affect the frequency band of effective ADS were numerically investigated. For highly dispersive materials, the end reflection pulse does not fully develop before the arrival of subsequent multiple reflections, resulting in severe truncation error and decreasing the upper frequency of effective ADS. Methods to estimate and extend the frequency band of effective ADS were further proposed. The simple procedures behind PVA, and its computationally efficient nature, make this method suitable for field monitoring. The resultant ADS is a potential improvement to current applications utilizing a travel time approach.