Using ground penetrating radar and time–frequency analysis to characterize construction materials

For decades, applications of nondestructive evaluation-civil engineering (NDE-CE) focus on object identifications (such as steel bars, tendon ducts and backwall reflections) in infrastructures. Because of the advantage of efficient visualization of internal structure, utilization of these methods can probably be extended to material characterization (MC) of aging and adversely exposed infrastructures. However, two factors yield a big gap between NDE and MC. First, for the ease of visualization, the primary focus of NDE-signal processing is object identification, which usually alters the originality of the signal. Second, there is lack of relationship and inverse models bridging the NDE-derived and conventional material properties compared to other disciplines of science, such as geophysics. These disadvantages make laboratory and field-scale NDE-MC still a far-reaching holy grail and is possibly the greatest hurdle to be regularly adopted in CE structures. This paper attempts to address this gap from object identification to MC using ground penetrating radar (GPR) as one of the most frequently used NDE-CE methods, and signal processing with joint time–frequency domain (JTFA) analysis. Three examples of material property characterization regarding the individual effects of steel bar corrosion in concrete, hydration and moisture content distribution of construction materials are given.