Optimization of ground-penetrating radar data to predict layer thicknesses in flexible pavements

Ground-penetrating radar has been used successfully to predict layer thicknesses in flexible pavements. Currently, a time-domain method is used to measure the two-way travel time between reflections from which the layer thickness is predicted. In this method, the dielectric constants are assumed to be real and are calculated from simplified equations. This paper presents a new method to measure layer thicknesses in flexible pavements. The method is based on predicting the reflected frequency spectrum through a multiple reflection model. The dielectric constants of the materials are assumed to be constant over the entire frequency range of the radar system, although dielectric and conduction losses are considered in the model. An optimization procedure is used by first calculating the reflected frequency spectrum using several values for the thicknesses and the complex dielectric constants, then choosing the solution with the minimum root-mean-square error between the predicted and measured frequency spectrums. The method was used in three pavement sections with known layer thicknesses. A percentage error of less than 3.6% was found between the measured and predicted values.