| As an auxiliary driving facility, automobile anti-collision radar could give the alarm of potential danger and enhance the driving safety. So it has been widely concerned. In order to improve the performance of automobile anti-collision radar, this thesis gives a comprehensive survey and does some research on key technology of signal processing of millimeter automobile anti-collision radar system.Firstly this thesis reviews some problems of millimeter automobile anti-collision radar system, such as the choice of the work principle of millimeter radar, the principle of distance and speed measurement of FMCW millimeter radar, the requirement of system performance and the choice of main parameters of automobile anti-collision radar, and the basic structure of millimeter automobile anti-collision radar.Then, this thesis gives a detailed discuss on the two main problems for radar signal processing: frequency spectrum analysis and signal denoising. FFT is chosen as the main method of frequency spectrum analysis. It analyzes the inherent antinomy between the computation amount and the accuracy of FFT, and proposes FFT-Chrip Z transform method to improve FFT. Both in theory and experimental results show the proposed method is superior to traditional FFT.The denoising of intermediate frequency signal is the other main content in this thesis. In this content two main denoising methods based on wavelet transform:denoising based on threshold value and denoising based on modulus maximum. It does some research on denoising based on threshold value, analyzes the drawback of soft threshold and hard threshold denoising methods, proposes two improved methods, the experimental results show the proposed methods greatly improve the effect of denoising. It also does some research on denoising based on modulus maximum. In order to reduce the amount of computation of AP algorithm, an improved method is proposed. The proposed method has good effect of denoising, and it reduces the amount of computation. |
PDF Full Text Request