원문정보
초록
영어
When the target of interest is determined, the transmitter and receiver positions of bistatic radar are of great importance at the aspect of radar target classification. The radar cross section (RCS) of a target varies with these positions, and the target classification performance is considerably influenced by RCS. In this study, the target classification performance using the bistatic scattering data of wire targets and scale-model targets is analyzed and compared. Time-frequency analysis and effective compression techniques are used for target feature extraction from the bistatic scattering data of each target, and a multilayered perceptron (MLP) neural network is used as a classifier. The optimum receiver position is found by comparing the calculated classification probabilities while changing the position of the bistatic radar receiver. The classification results using calculated data and measured data show that an optimally positioned bistatic radar yields better classification results, demonstrating the importance of the positions of the transmitter and receiver for bistatic radar.
목차
1. Introduction
2. Proposed Method
2.1. Optimum Bistatic Angle Extraction
2.2. Feature Extraction Using Short-time Fourier Transform
2.3. Feature Extraction Using Continuous Wavelet Transform
3. Simulation and Measurement Results
4. Conclusion
Acknowledgements
References