Research On Multi-target Parameter Estimation Method Based On Vehicle Millimeter Wave Radar

Smart cars are emerging technologies under the background of a new round of technological and scientific revolution.They are the key to the transformation and upgrading of the automobile industry and are also recognized as the development direction in the world.As one of the core sensors used in smart cars,vehicle radar plays a vital role in achieving intelligent driving of automobiles.Millimeter wave radar is widely used in automobile anti-collision detection systems due to its low transmission power and high resolution.Strengthening the research on vehicle millimeter wave radar and its related technologies can not only improve road traffic safety,but also promote the further development of intelligent transportation.The Linear Frequency Modulation Continuous Wave system is the most common radar system for existing vehicle millimeter wave radars.The LFMCW radar adopts the "transmit-reflection-receive" mode of operation,and performs spectrum analysis on the beat signal after the mixing and receiving of the received and transmitted signals to estimate the distance and speed of the target to be measured.Due to the working mode of LFMCW,there will be a problem of distance-velocity coupling,which will generate false targets under multi-target conditions;and because the length of the radar echo signal sequence is limited,the traditional fast Fourier algorithm will produce a fence effect.This will reduce the accuracy of the radar's estimation of distance and velocity parameters.This paper combines the national key research and development plan "Electric Vehicle Intelligent Assisted Driving Technology R & D and Industrialization"(project number: 2016YFB0101102)to carry out research work,the following is the main work content of this paper:1.The background and significance of the research of this paper are expounded.The domestic and foreign research status of millimeter wave radar,LFMCW radar multitarget parameter estimation algorithm and short-term signal frequency estimation algorithm are introduced respectively.The structure and working principle of vehicle LFMCW radar are introduced in detail,and different derivations are made.The echo signal model of the radar under the modulated waveform and its principle of ranging speed measurement,and analyze the problems of the speed measurement ranging.2.A multi-target parameter estimation method based on interval sweep frequency band difference and FFT-MUSIC is presented.Firstly,a target distance-velocity decoupling method based on interval sweep frequency band difference is presented.Based on the beat signal model of interval sweep frequency band,the distance-speed decoupling principle of the method is deduced.Then,it is proposed.A high-precision target frequency component estimation method based on FFT-MUSIC is proposed.Based on the traditional fast Fourier algorithm,combined with MUSIC algorithm,the accuracy of radar estimation of distance and velocity parameters is improved.Finally,based on interval sweep Based on the target distance-velocity decoupling method of frequency band difference,a multi-target matching method based on interval sweep period is proposed to eliminate the false target generated by the proposed target distance-speed decoupling method.3.The performance analysis of the multi-target parameter estimation method based on interval sweep frequency band difference and FFT-MUSIC is carried out,including multi-target velocity and distance parameter estimation error analysis,and the multitarget parameter estimation method proposed in this paper is derived.The lower bound expression and the matching complexity of the multi-target matching method proposed in this paper under complex road conditions is analyzed.Finally,the feasibility of the proposed method and its parameter estimation performance are verified by experimental simulation.4.The hardware experimental platform is built to verify the multi-target parameter estimation method proposed in this paper.Firstly,the structure and performance of the hardware experimental platform are described in detail.Secondly,the parameter estimation process of the multi-target parameter method proposed in this paper is illustrated by a set of beat signal sampling data of the interval sweep band.Finally,the motion trajectory of the measurement target is obtained by multiple measurements,which proves the feasibility of the multi-target parameter estimation method proposed in this paper.