Research And Realization Of Two-dimensional Constant False Alarm Algorithm

Constant false alarm rate(CFAR)detection has always been an important part of radar signal processing.CFAR is mainly in the case of constant false alarm rate to detect the target.Whether in a uniform Gaussian background environment or in the complex clutter background environment,CFAR detection has a very wide range of applications.Several classic CFAR algorithms are discussed in this thesis,such as the mean level type CFAR,ordered statistic type CFAR and so on.The theoretical basis and the implementation procedures of these algorithms are introduced.Then the characteristics of each CFAR detectors are analyzed in the homogeneous clutter background,the clutter edge heterogeneous and the multitarget situations.Then,the detection results in different clutter backgrounds are given through the computer simulations,the performances of these CFAR algorithms are analyzed with the computer simulations,besides,the environments of several CFAR algorithms are summarized,and their respective advantages and disadvantages are given.Since the clutter and noise are existed on the range and the Doppler dimension,the received echo signal is the two-dimensional range-Doppler-matrix(RDM).Therefore,based on the classical CFAR algorithm,the two-dimensional CFAR processing method is discussed in this thesis,then the basic principle of the two-dimensional mean level type CFAR algorithm and the two-dimensional ordered statistic CFAR algorithm are introduced,then their procedures of realization are analyzed.According to the needs of engineering practice,the two-dimensional OSCA-CFAR and the two-dimensional FOSCA-CFAR are introduced in the thesis,these two methods utilize the advantages of classic CFAR algorithms and reduce the computational complexity.The difference between these two two-dimensional CFAR detection methods is that the two-dimensional FOSCA-CFAR is simplified on the two-dimensional OSCA-CFAR.At the same time,according to the engineering needs,this thesis also introduces an efficient two-dimensional meter reference window,the design of this reference window can avoid too much invalid units of the traditional rectangular reference window and solve the problem of low utilization rate of useful information in the cross-reference window.Then the measured data are processed and analyzed,which demonstrate the utility of the meter-shaped reference window.Finally,the two-dimensional CFAR in engineering is realized in FPGA.Because of the limited internal memory resources of FPGA,DDR2-SDRAM is used as the external memory of FPGA,and the hardware platform structure based on FPGA and DDR2-SDRAM is given.The interface between the FPGA and DDR2-SDRAM is to ensure the correctness of data transmission between FPGA and DDR2-SDRAM.At the same time,the design of the system clock and the design of the data transmission module are completed.The meter reference window is selected as the reference window of two-dimensional CFAR,and the two-dimensional FOSCA-CFAR method is selected as two-dimensional CFAR method.Multiple parallel modules are designed and the data in the reference window are processed in parallel to improve the data processing speed and reduce the computational complexity.The real-time performance and the detection performance of the two-dimensional CFAR processing method are improved.