抄録
A GPR (Ground Penetration Radar) is a narrow bandwidth, high radar range device, where improvement of resolution is greatly desired to resolve the closely buried targets and clear imaging of targets. However, it is not easy to get high resolution, as the GPR signals are very weak and enveloped inside the noise. The MUSIC (Multiple Signal Classification) algorithm, which has its fame for its super resolution capacity, has been implemented. In addition, conventional spectral estimation technique, FFT (Fast Fourier Transform), has also been implemented for high precision receiving signal level. As well as this, CPM (Combined Processing Method) was performed for the signal processing and image reconstruction of closely located GPR targets. In this paper we proposed the method to improve the SNR (Signal to Noise ratio) of MUSIC and CPM. In order to support the proposal, a detailed simulation has been preformed analyzing SNR. Moreover, a laboratory experiment has also been performed for the thorough investigation and maximum resolution capacity of CPM. The experiment result demonstrates the CPM has the vertical resolution of 2 cm at 800 MHz frequency bandwidth in a water medium. All the simulation and experimental results are presented.
