Signal Processing Technology Research With Detection Of Airport Runway Based On GPR

With the rapid development of aviation traffic transportation, as the main infrastructureof the aircraft movements including takeoff and landing, detection reliability of carryingcapacity and working life of the airport runway is directly related to the security operations ofairport. Currently, as a high-performance tool for non-destructive testing, ground-penetratingradar not only has the ability of strong anti-interference, but also could obtain intuitive andaccurate measurement results, and it has been widely applied in the field of the airport runwayquality engineering. However, aiming at the detection of the thin layer structure underneaththe airport runway, especially for detection to escapes spatial, cavity, etc., conventionalground-penetrating radar has been unable to meet a demand for the accurate detection andidentification of underground thin layer structure due to the limitations of the radar workingbandwidth.Therefore, this paper will use the super-resolution spectral estimation technology ofmodern signal processing theory to investigate the problem of time delay estimation for thinlayer structure underneath the airport runway. Firstly, On the basis of a constructedapproximation to model of the actual airport runway, GprMax simulation software forground-penetrating radar is utilized to do forward modeling analysis for the stratified mediamodel of airport runway by finite difference time domain (FDTD) method. Secondly,combined with deconvolution technique, this paper has deeply researched spatial smoothingprocess (SSP), modified spatial smoothing process (MSSP) and improved spatial smoothingprocess (ISS) three kinds of spatial smoothing technique for GPR super-resolution methods oftime delay estimation in order to gain time delay estimation value of the undergroundoverlapped echo, before two kinds of super-resolution spectral estimation methods, multiplesignal classification (MSUIC) and estimation of signal parameters via rotational invariancetechniques (ESPRIT), have been introduced in detail. Simultaneously, the performance ofthree kinds of spatial smoothing technique for GPR time delay estimation would be measuredand evaluated by success probability and root-mean-square error of the algorithm. At last, thispaper had made full use of Matlab simulation software to operate a lot of Monte Carlosimulation statistical experiments to analyze and compare the performance of three kinds ofspatial smoothing algorithms. The results of this paper would not only provide a favorable theoretical guidance andfoundation for the future ground-penetrating radar detection of airport runway, but also have acertain reference value for the application of ground penetrating radar in highway quality test.