Research On Signal Processing Method For Chirp Sequence Automotive Radar

Automotive radars have been widely used in advanced driver assistant systems(ADAS)and automated driving systems,due to their superior performances at night and in bad weather conditions such as rain and fog,etc.The chirp sequence automotive radar is one type of the most commonly used frequency-modulated continuous wave(FMCW)automotive radars.In the chirp sequence radar,the range-Doppler processing method based on 2D fast Fourier transform(FFT)can be used to extract the range and velocity parameters of multiple targets from the beat signals.Compared with other kinds of FMCW automotive radars,the chirp sequence automotive radar has higher target resolution capability,higher measurement precision,and thus has attracted considerable attention in recent years.In this dissertation,we focus on three critical problems in the signal processing method of the chirp sequence automotive radar,and propose four novel signal processing method.The main contents are illustrated as follows:To solve the problem that the target detection performance of the chirp sequence auto-motive radar degrades in high-target-density situations,a new type of truncated statistic(TS)constant false alarm rate(CFAR)detection method based on singly right truncated Rayleigh distribution model is proposed.The method applies a 2D window to slide on the 2D magnitude spectrum of the beat signal.In the first step,outliers in the 2D sliding window are eliminated.Subsequently,the singly right truncated Rayleigh distribution model is used to model the re-maining background samples,and the look-up table method based on the maximum-likelihood estimation is adopted to estimate the required threshold parameter.Finally,the threshold level is calculated using the threshold parameter,and the target detection is realized.Simulation results show that the TS-CFAR detection method has more superior anti-interference performance than the ordered statistic(OS)CFAR and cell averaging(CA)CFAR methods,which are commonly applied in automotive radars.Moreover,the proposed method has much lower computational complexity compared with the OS-CFAR method since the look-up table approach is applied.The effectiveness of this method is verified by experimental results.To solve the Doppler ambiguity problem of the chirp sequence automotive radar,a novel waveform named interlaced time-varying interval(ITVI)chirp sequence waveform is proposed.In the waveform,the intervals between two adjacent chirps are changing alternately.The pro-posed waveform can be divided into two interlaced classical chirp sequences with equal chirp repetition intervals.In the corresponding signal processing method,the beat signals of the two chirp sequences are processed separately.The phase difference of the two beat signals is used to resolve Doppler ambiguity.The proposed waveform and its corresponding signal processing method are easy to implement.Moreover,the refresh rate of the proposed method is higher compared with the classical Doppler ambiguity resolution method based on Chinese remainder theorem.Simulation and experimental results verify the effectiveness of the proposed method.To solve the Doppler ambiguity problem of the chirp sequence automotive radar,another novel waveform named jittered chirp sequence waveform is proposed based on the compressed sensing(CS)theory.In the proposed waveform,the initial times of the chirps are randomly jit-tered.As long as the waveform parameters are properly designed,sparse recovery method can be applied to achieve unambiguous Doppler processing.The parameter design principle of the new waveform is established in this paper.Subspace pursuit(SP)algorithm is recommended as the suitable sparse recovery method for Doppler processing due to its low complexity and stable recovery performance.Compared with the ITVI chirp sequence waveform,the jittered chirp sequence waveform has the same refresh rate.Although the computational complexity of its signal processing method is higher,the jittered chirp sequence waveform has stronger anti-mutual-interference capability.The effectiveness of the proposed waveform and its corre-sponding signal processing method are verified by simulation and experimental results.To solve the angle-velocity coupling problem in the switched antenna array(SAA)angle measurement method of the chirp sequence automotive radar,a novel SAA method based on multi-cycle(MC)sequential-reverse-fixed(SRF)time division multiplexing(TDM)technique is proposed.In the method,three switching periods are adopted.In the first switching period,the receiving channel is switched to different receiving antennas in sequential order.In the second switching period,the receiving channel is switched to different receiving antennas in reverse order.In the third switching period,the receiving channel is switched to a fixed receiving antenna.The information extracted from the beat signals of the first two switching periods is used to realize angle-velocity decoupling,and the information extracted from the beat signals of the third period is used to eliminate ghost targets.Therefore,the angle,velocity and range parameters of the real targets can be extracted from the three switching period.Compared with the classical MC repeated sequential TDM SAA method and the MC sequential-reverse TDM SAA method,the proposed method can significantly relax the restriction on the target maximum velocity,and thus make the SAA method more feasible for chirp sequence automotive radar.The effectiveness of the proposed method is verified by simulation results.