Research On Intermediate Frequency Signal Processing Algorithm Of Vehicle Millimeter Wave Radar

In recent years,with the development of China's economy,people's demand for automobiles has gradually increased,and collision accidents caused by a driver's lack of concentration have also increased.Advanced driver assistance systems(ADAS)and autonomous driving technologies are increasingly being supported by the government,valued by the scientific research institutions and favoured by the companies.As an important sensor in the vehicle,researching on intermediate frequency signal processing algorithms is of great significance.In this thesis,two waveforms of frequency modulated continuous wave(FMCW)and frequency-modulated interrupted continuous wave(FMICW)are studied.It is proposed to develop a radar intermediate frequency signal processing algorithm with high detection accuracy,good real-time performance and strong system robustness.It mainly studies several aspects of signal denoising,ranging,velocity measurement and angle estimation algorithms:1)Various types of noise,such as mechanical noise and thermal noise,are mixed in the actual radar echo signal.In order to eliminate the noise in the radar echo signal to improve the signal-to-noise ratio,a signal denoising method based on Empirical Mode Decomposition(EMD)and wavelet threshold is proposed in this paper.Firstly,the original signal is decomposed into several Intrinsic Mode Function(IMF)components by EMD.Then according to the distribution characteristics of the signal and noise,the instantaneous half period criterion is used to filter out the IMF sets dominated by the useful signal and noise.The noise-dominated IMFs are processed by wavelet threshold denoising,and the signal-dominated IMFs are not processed.Finally,reconstructing each order IMF components can get the denoised signal.2)Study the application of FMICW radar in vehicle.Firstly,the disadvantage of FMCW radar on direct leakage of transmitted energy to overload the receiver is pointed out.Then,the problem of spurious spectral peaks in-vehicle application using FMICW radar with single switching frequency is analyzed.In this paper,we use a stagger algorithm based on multiple switching frequencies to remove the spurious peaks from frequency domain.Finally,the selection requirements of selecting multiple switching frequencies are given,and the influence of the algorithm on the signal-to-noise ratio of echo signals is analyzed.3)Research angle estimate algorithms.The angle estimation principles of the conventional beamforming(CBF)method,the minimum variance distortionless response algorithm(MVDR),the multiple signal classification(MUSIC)algorithm and estimation of signal parameters via rotational invariance techniques(ESPRIT).Then,we do the simulation experiment to analyse the angular resolution and the applicability of each algorithm in the target angle estimation of vehicle radar.4)We use the phased array system toolbox in Matlab to build the transmitter,radiator,receiver amplifier and other radar modules to establish complex traffic scenes to obtain simulated echo signals of multiple targets.And we select the AWR1642 evaluation board of the vehicle millimeter wave radar launched by Texas Instruments(TI),then build an experimental platform that simulates real traffic conditions to capture raw radar echo data.The radar echo signal data obtained by software simulation and AWR1642 evaluation board is used to verify the intermediate frequency signal processing algorithm.It is shown that the proposed signal denoising method can obtain the maximum signal-to-noise ratio,the maximum signal-to-noise ratio gain and the minimum root-mean-square error,and have a better filtering effect.It is also shown that the stagger algorithm in FMICW radar is close to the equivalent FMCW radar detection performance and is good in multiple targets scenario.