【作者】 石磊；

【导师】 郑朝霞；

【作者基本信息】 华中科技大学 ， 集成电路工程， 2020， 硕士

【摘要】 随着科技的不断发展,汽车已经成为了人们出行的一种重要手段,然而交通事故频频发生,有关汽车的安全性问题亟待解决,车载毫米波雷达系统可以通过对目标信息进行检测从而做出应对,以此来避免交通事故的发生。由于车载毫米波雷达的相关技术基本被国外公司垄断了,因此开展关于毫米波雷达的研究是十分必要的。本文主要对毫米波雷达的信号处理算法进行研究与电路实现,本文基于毫米波雷达系统中的锯齿波雷达系统,对锯齿波雷达系统的信号处理算法进行了研究。首先,针对锯齿波雷达测距测速过程中的距离速度耦合问题提出了二维FFT算法,将目标的距离信息和速度信息分离开来,然后基于雷达回波信号中叠加的噪声提出了恒虚警检测算法,提高了对目标的检测概率,最后基于目标信息的提取提出了峰值检索算法。之后,基于Matlab平台验证了算法的有效性,根据仿真结果分析确定了算法中的距离维FFT点数为512,速度维FFT点数为128,对目标的检测概率为99.4%,关于目标检测的距离误差率为0.84%,关于目标检测的速度误差率为2.3%,与其他文献对比发现本文的检测概率、距离误差率以及速度误差率比较好。最后,本文对算法进行了硬件加速,基于软件部分设计了硬件电路并且进行了功能仿真,验证了电路的功能正确。随后进行了FPGA性能分析与验证,基于SMIC180nm工艺对本文所设计的电路进行了ASIC实现并进行了性能分析。更多还原

【Abstract】 With the continuous development of science and technology,automobile has become an important means for people to travel.However,traffic accidents happen frequently,and the safety problems of automobile need to be solved urgently.Vehicle mounted millimeter wave radar system can detect the target information to make response,so as to avoid traffic accidents.As the related technologies of vehicle borne MMW radar are basically monopolized by foreign companies,it is necessary to carry out research on MMW radar.In this paper,the signal processing algorithm of millimeter wave radar is studied and implemented.Based on sawtooth radar system,this paper studies the signal processing algorithm of sawtooth radar system.Firstly,a two-dimensional FFT algorithm is proposed to separate the range information and velocity information of the target.Then a CFAR detection algorithm is proposed based on the superimposed noise in the radar echo signal,which improves the detection probability of the target.Finally,a peak retrieval algorithm is proposed based on the extraction of target information.After that,the validity of the algorithm is verified based on MATLAB platform.According to the analysis of the simulation results,it is determined that the number of distance dimension FFT points in the algorithm is 512,the number of speed dimension FFT points is 128,the detection probability of the target is 99.4%,the distance error rate of the target detection is 0.84%,and the speed error rate of the target detection is 2.3%.Compared with other literature,it is found that the detection probability,distance error rate and speed error rate in this paper are better.Finally,the algorithm is accelerated in hardware,and the hardware circuit is designed based on the software part and the function simulation is carried out to verify the correct function of the circuit.Then the performance of FPGA is analyzed and verified.Based on the technology of SMIC 180 nm,the circuit designed in this paper is implemented in ASIC and the performance is analyzed.更多还原