碩士 / 中原大學 / 電子工程研究所 / 105 / This thesis aims to realize the autoregressive high-resolution spectrum estimator based on the Burg method and constant false alarm rate (CFAR) detection for analyzing the Doppler spectrum from the radar echo over strong sea clutters. We adopt the software/hardware co-simulation approach to design the high-resolution spectrum estimator by using the MATLAB software and System Generator tool. The better estimation performance can be achieved by using the Burg-algorithm-based high-resolution spectrum estimation instead of using the traditional moving target identification (MTI) method in the radar signal processing. As for the hardware implementation, the design steps are described as follows. First, we design the finite-length-response-Lattice (FIR-Lattice) filter framework to realize the AR-Burg algorithm architecture with eight-order. In doing so, the reflection factors and filter coefficients of the AR-Burg algorithm can be obtained. Second, the hardware of the infinite-length response (IIR) with eight-order and fast Fourier transform (FFT) is designed to obtain the frequency contents of the radar echo over the strong sea clutters. Third, the CFAR detector is designed to determine whether the target in the sea exists. Finally, we verify the correctness of the co-simulation approach by comparing the simulation results of the MATLAB and those from the System Generator outputs. Simulation results show that the software/hardware results are very consistent with each other. Therefore, the proposed algorithm is proved to have high-accuracy estimation accuracy and to improve the detection performance.
| Identifer | oai:union.ndltd.org:TW/105CYCU5428023 |
| Date | January 2017 |
| Creators | Jen-Chieh Hsu, 徐仁捷 |
| Contributors | Rih-Lung Chung, 鍾日龍 |
| Source Sets | National Digital Library of Theses and Dissertations in Taiwan |
| Language | zh-TW |
| Detected Language | English |
| Type | 學位論文 ; thesis |
| Format | 76 |
Page generated in 0.0159 seconds