毫米波主动探测系统的信号处理算法及实现

发布时间：2018-04-13 18:29

本文选题：毫米波 + LFMCW　；参考：《南京理工大学》2017年硕士论文

【摘要】：毫米波雷达在近程测距方面有着广泛的应用,高效的测距算法和稳定的硬件电路是实现高精度测距的重要保障。本文以3mm主动探测系统为研究背景,围绕主动雷达测距的高精度算法和系统硬件电路设计进行了深入研究。按照实际应用需求,构建了一种对信号进行处理的系统,主要工作如下:1、介绍了 LFMCW雷达主动探测的基本原理和实现方案,给出了系统的结构模型,建立了差频信号模型并进行了相关参数分析。2、利用比值法测距算法简单、精度高、易实现的特点,首先分析了 Rife算法和幅相联合内插算法(CAPI)的性能,提出了 Rife和幅相联合内插结合算法(R_CAPI);其次剖析了比值法其他两种常用算法Jacobsen和Quinn算法的性能,提出了基于J-Quinn的频率估计算法。两种改进算法均提升了差频信号频率估计性能。3、针对FFT变换存在栅栏效应、频谱泄露和分辨率不高的缺陷,以及目前一些算法在多频率估计方面的不足,本文提出将MVDR模型的现代谱估计理论引入到LFMCW雷达测距中。在传统MVDR谱估计方法基础上进行了改进,提出自适应的最速下降频率估计算法(AFEA-SD),此方法中所使用的依托迭代理论的谱峰搜索区别于通常的搜索方法,结果表明,此算法可以有效克服FFT变换的固有缺陷,并具有分辨率高、估计精度高、计算量小和可用于多频率估计的优势。4、结合项目需要,以FPGA的XC3S50为核心器件,完成了调制信号产生、差频信号采样、差频信号处理三个部分的硬件电路设计与实现,重点包括电源产生、时钟控制和仿真JTAG等辅助模块电路设计,以及数模转换、电压比较、差频信号采样等测距模块电路设计。同时,给出了整体的系统原理图设计和PCB实物图。5、对系统的软、硬件进行联调和户外模拟实验,从而验证了系统的可行性,检验了各算法在雷达测距中的性能和硬件电路设计的可靠性。
[Abstract]:Millimeter-wave radar is widely used in short-range ranging. Efficient ranging algorithm and stable hardware circuit are the important guarantee to achieve high-precision ranging.Based on the 3mm active detection system, the high precision algorithm of active radar ranging and the hardware circuit design of the system are studied in this paper.According to the practical application requirements, a signal processing system is constructed. The main work is as follows: 1. The basic principle and implementation scheme of LFMCW radar active detection are introduced, and the structure model of the system is given.The differential frequency signal model is established and the correlation parameters are analyzed. The ratio method is used to measure the distance, which is simple, accurate and easy to realize. Firstly, the performance of the Rife algorithm and the amplitude-phase interpolation algorithm are analyzed.In this paper, the Rife and amplitude-phase combined interpolation algorithms are proposed, and the performance of the other two common algorithms, Jacobsen and Quinn, are analyzed, and the frequency estimation algorithm based on J-Quinn is proposed.The two improved algorithms both improve the performance of frequency estimation of differential frequency signals. 3. Aiming at the defects of FFT transform, such as fence effect, spectrum leakage and low resolution, and the shortcomings of some current algorithms in multi-frequency estimation, the two improved algorithms improve the performance of frequency estimation of differential frequency signals.In this paper, the modern spectrum estimation theory of MVDR model is introduced into LFMCW radar ranging.Based on the improvement of the traditional MVDR spectral estimation method, an adaptive algorithm for estimating the steepest descent frequency is proposed. The spectral peak search based on the iterative theory is different from the usual search method in this method. The results show that,This algorithm can effectively overcome the inherent defects of FFT transform, and has the advantages of high resolution, high estimation accuracy, small computational complexity and the advantage of multi-frequency estimation. Combined with the project needs, the modulation signal is generated using FPGA's XC3S50 as the core device.The hardware circuit design and implementation of the three parts of differential frequency signal sampling and differential frequency signal processing include power generation, clock control and simulation JTAG, digital to analog conversion, voltage comparison, etc.Circuit design of ranging module such as differential frequency signal sampling.At the same time, the design of the whole system schematic diagram and the PCB physical diagram .5 are given. The software and hardware of the system are combined with the outdoor simulation experiment, which verifies the feasibility of the system.The performance of each algorithm in radar ranging and the reliability of hardware circuit design are tested.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN957.51

【参考文献】