Studies On Key Techniques Of Post Processing For Ground Penetrating Radar Signals

Ground penetrating radar(GPR) is a nondestructive detection method using electromagnetic radiation to locate shallow geological subsurface features and underground utilities buried in the ground. It has virtues such as convenience, high detection efficiency,and high resolution power. The usage of GPR tends to be more and more broad and diversified in the engineering exploration field. GPR has made remarkable achievements and development in the past few decades, but the results mainly benefit from the gradual improvements of system hardware. The post processing techniques of measured GPR signals occur with quantitative change and without dramatic improvement, which restrain the GPR system performance. The emphasis of scientific research in this field is to find post processing techniques of measured GPR signals and quantitative analysis of the measured targets’ parameters. There are lots of difficulties to design a GPR signals post processing system, not just because of the complexity of environment factors, but because of the variety of targets and their attributes. In order to improve the GPR’s capability of quantitative analysis and semi-quantitative analysis for targets, we study on post processing of key technique in GPR signals in this paper. The research content mainly includes characteristic analysis of GPR echo signals, signal enhancement, target imaging,target reconstruction, waveform inversion, and target recognition. The main works and results are as follows:1. The received signals from GPR contain round trip echoes, clutters and complex noise signals. These jamming signals seriously affect the interpretation precision of shallow subsurface geological information. For better analysis of the characteristics of GPR signals,we analyze the propagation characteristics of the pulse signal transmitted from a GPR based on Maxwell’s equations. On this basis, we introduce a new algorithm, the intrinsic component decomposition(ICD), for efficient and precise analysis of GPR signals. The ICD algorithm generates intrinsic component libraries(ICLs) by changing the phase of the transmitted GPR signal at first. Then, we can get the sparse representation of the received GPR signal using the ICLs as dictionary. Then, we decompose each GPR signal into a plurality of intrinsic component function(ICF) components and a residue by an adaptive threshold. Based on the signal reconstruction principle of the sparse representation theory,each ICF component can be obtained. The correlation between each ICF component and the components in ICLs are depressed in turn. Finally, we perform experiments to examine the performance of the ICD algorithm by employing simulated and real data, which can well demonstrate the capability of the proposed approach. The experimental results suggest that the ICD algorithm has a great ability for clutter suppression and GPR signals enhancement. With ICD algorithm, it becomes easier to analyze the shallow geological subsurface information of GPR signals.2. In order to recover the shape of buried objects in GPR imaging, a self-correlation back-projection(SBP) algorithm is proposed, which is fast imaging and can distinguish the object’s shape. It improves the existing BP algorithms in the following aspects. First, the reflection echo signals of a specific imaging point obtained from its nearest exploration point have high correlation with the one from its multiple nearest neighbors. By setting up a correlation threshold, the valid echo information sequence of the imaging points can be adaptively chosen, which enables the SBP algorithm to have faster calculation speed and better resolution. Then, the imaging result is amended by using a depth energy compensation algorithm. It can improve the imaging resolution of the deep underground objects. The experimental results show that the proposed SBP algorithm is superior to the existing BP algorithms in terms of computing speed and imaging accuracy, which can effectively recover objects with complex shapes.3. GPR system is very effective in detecting and locating the buried underground utilities,spheres, and landmines, circular is their common characteristic in 2D profile. A novel reconstruction algorithm is proposed to get the perfect reconstruction of circular targets,which works well when the permittivity of the background medium is unknown. At first,we derive the mathematical relationship between circular targets’ parameters and the targets echo signals’ curve from basic electromagnetic laws. We introduce the circular targets reconstruction algorithm on this basis and perform experiments with simulation data and real data to examine the efficiency of the proposed algorithm. The experimental results show that the proposed algorithm can be applied on buried circular targets reconstruction from GPR signals. It can reconstruct buried targets’ position, buried depth,size, dielectric properties, and the relative permittivity of the background medium, besides has the characteristics of excellent real-time ability and high reconstructive precision.4. There are some problems on the existing waveform inversion methods of GPR data for stratified media, which include poor convergence of error analysis function, local optimum,and poor stability of inversion results. To solve these problems, we propose a novel waveform inversion method with first reaching time restrictions, and formulate multi-level subdivision search model to solve the optimal solutions of the error analysis function. The proposed waveform inversion method uses the first reaching time of echo signals located in the interface of adjacent medium as a priori information to reduce the computing complexity of the problem-solving methods. By using the multi-level subdivision search method, it can not only improve the efficiency and decrease the complexity, but also guarantee inversion results with good stability. The experimental results show that the proposed waveform inversion method can identify the geoelectric structure thickness and permittivity parameters, which can be applied on the stratified media reconstruction of GPR data. Besides, it has advantages of low time complexity, high precision of inversion results and good stability.5. We derive the mathematical relationship between the target echo signals’ curve and the system function of underground medium. Theoretical derivation indicates that the mathematical relationship is a complex nonlinear implicit expression. To achieve the geoelectric structure parameters recognition, a novel method of extraction technology based on making the best of GPR signal’s dielectric spectrum is proposed to pick up the signal that in different frequency bands. First, according to power spectral distribution of impulse source signal, we select the frequency band of GPR data’s main energy spectrum.Second, we divide the frequency band into different sub-band by band-pass filter group.The features of "slow-decay point" and "fast-decay point" are separately extracted from each sub-band, and encoded to build the feature vector. The experimental results show that the classification results used the proposed feature vector agree well with the parameter values set during experiments. The proposed algorithm can be applied on targets classification of GPR signals. Besides, it has advantages of low time complexity and strong anti-noise ability.