CN110609264A - A target echo Doppler frequency estimation method for pulsed lidar - Google Patents

Abstract

The invention discloses a target echo Doppler frequency estimation method for a pulse laser radar, and relates to the technical field of laser radar signal processing. Aiming at the linear frequency modulation echo signal, when the signal-to-noise ratio is high, the Doppler frequency shift can be roughly estimated only according to the signal bandwidth and the frequency of the transmitted signal through Fourier transform, and the technical scheme of the invention can directly and accurately estimate the central frequency of the signal so as to obtain the Doppler frequency shift of the echo; when the signal-to-noise ratio is low, the signal frequency can not be observed by the Fourier transform method, and the Doppler frequency of the echo signal can still be estimated in a small error range by the technical scheme of the invention.

Description

A target echo Doppler frequency estimation method for pulsed lidar

technical field

The invention relates to the technical field of laser radar signal processing, in particular to a method for estimating the Doppler frequency offset of an echo signal of a linear frequency modulated pulse laser radar, which is especially suitable for Doppler frequency offset estimation of a single pulse period target echo in a low signal-to-noise ratio environment. Estimated frequency offset.

Background technique

Lidar is a rapidly developing and widely used measurement method. It has the advantages of short operation time, high measurement accuracy, little weather influence, safe operation, and is not restricted by terrain or region, and can measure ground and non-ground layers at the same time. Due to the constant speed of light, time is related to the distance from the emitter, so spatial information is detected along the electron beam trajectory. The radial velocity of the target is detected by receiving the signal frequency offset from the target. It has extremely high angular resolution, distance resolution and strong anti-interference ability. Therefore, with the continuous development and popularization of technology, lidar will be more widely developed in the military and civilian fields.

The laser radar first transmits laser pulses through the laser transmitter, and then the reflected signal is received by the telescope, and the reflected signal is photoelectrically converted by the photodetector. After sampling, it becomes a digital signal for signal processing to obtain target information.

Due to the complex working environment of the radar, the echo will be mixed with more background noise, resulting in a low signal-to-noise ratio of the lidar echo signal. Aiming at the characteristics of lidar signals, the current common signal processing method is incoherent accumulation. The incoherent accumulation method is used to accumulate the signal in the frequency domain to average the signal spectrum and improve the signal spectrum identification. However, this method only affects the signal-to-noise ratio. The improvement is limited. For the echo signal with low signal-to-noise ratio, the signal spectrum line cannot be effectively identified; if the transmitted signal has full coherence, the coherent accumulation can be carried out directly. However, the lidar is limited by the laser transmitter, and it is easy to introduce phase random mutation, it is difficult to achieve pulse coherence. In addition, due to the characteristics of the chirp signal itself, the spectrum of the echo signal will occupy a certain bandwidth, and it is difficult to directly observe the center frequency of the spectrum under the influence of noise, so the Doppler information of the target cannot be obtained.

Therefore, how to effectively detect the weak echo signal of lidar in real time under the condition of low signal-to-noise ratio and effectively estimate the echo Doppler frequency offset is a key issue in the application of lidar.

SUMMARY OF THE INVENTION

In view of the above problems, the present invention proposes a frequency estimation method with windowed convolution in the frequency domain, which realizes the Doppler frequency offset estimation of the echo signal of the chirp laser radar, and only needs one pulse period to effectively estimate Doppler information of targets under low signal-to-noise ratio conditions.

Assuming that there is a moving target, when the pulse of the reflected signal arrives at the receiver, the signal frequency is shifted due to the existence of the Doppler effect; the shift of the received signal frequency reflects the speed information of the target; The echo data is converted from the time domain to the frequency domain by Fourier transform. The transmitted signal is transformed from the time domain to the frequency domain by Fourier transform, and the window is added in the frequency domain with the bandwidth as the scale to intercept the effective range of the spectrum. The windowed spectrum is convolved with the spectrum of the echo signal in the frequency domain. Peak search is performed on the result after convolution, and the position corresponding to the maximum value is the Doppler frequency of the echo signal. The transmitted signal and the echo signal are both chirp signals, and their frequency spectrum can be approximated as two rectangular windows in the frequency domain, so the convolution of the rectangular window in the frequency domain can be performed at the Doppler frequency. The Doppler frequency of the echo signal can be obtained by performing a peak search on the convolution result.

The technical solution of the present invention is a target echo Doppler frequency estimation method for a pulsed laser radar. In the method, the transmitted signal and the echo signal of the pulsed laser radar are both linear frequency modulation signals, and the method steps include:

Step 1: After the receiver receives the echo signal, it first performs frequency-selective filtering and amplification to filter out noise and clutter outside the signal frequency band; then performs frequency mixing and band-pass filtering and amplification to convert the radio frequency signal into an intermediate frequency signal;

Step 2: A/D conversion is performed on the intermediate frequency signal of step 1, and then digital down-conversion is performed after being converted into a digital signal, and two baseband signals of I and Q are obtained after quadrature mixing and low-pass filtering;

Step 3: Synthesize the two-way baseband signals of Step 2 into a complex signal, and select a pulse repetition period with a length of f _s *PRI according to the sampling rate f _s of the receiver and the pulse repetition period PRI of the transmitted signal and the PRI synchronization signal. The complex signal is used as an echo signal, and then Fourier transform is performed to convert it from the time domain to the frequency domain;

Step 4: Preprocess the original baseband transmit signal, perform Fourier transform on the baseband transmit signal with a pulse width of τ and convert it from the time domain to the frequency domain to obtain a spectrum signal, whose spectrum range is [-f _s /2, f _s /2], which is much larger than the bandwidth Bw of the signal, take the center frequency as the center and the bandwidth Bw as the scale to add a window in the frequency domain to intercept the signal within the bandwidth of the spectrum signal, and the window length is the bandwidth Bw;

Step 5: Convolve the spectrum within the bandwidth of the transmitted signal intercepted in step 4 and the echo signal spectrum obtained in step 1 in the frequency domain. According to the convolution process, the frequency axis changes from -f _s /2 to f _s /2 to After -f _s /2～f _s /2+Bw, convert the coordinate axis of the convolution result;

Step 6: Search for the maximum value on the frequency axis of the convolution result after the coordinate axis conversion. The coordinate of the frequency axis corresponding to the maximum value is the Doppler frequency shift of the echo signal plus the half bandwidth Bw/2, minus The corresponding frequency coordinate after Bw/2 is the Doppler frequency offset of the target echo.

Further, the signal after the intermediate frequency signal is A/D converted in the step 2 is:

x(t)=s(t)+n(t)

in: where f ₀ is the carrier frequency of the signal, f _d is the Doppler frequency shift, and n(t) is Gaussian white noise;

The original baseband transmit signal in step 4 is:

where μ is the slope of the frequency change.

The technical scheme of the present invention is first proposed in the aspect of estimating the center frequency of the chirp signal. For the chirp echo signal, when the signal-to-noise ratio is high, the Doppler can only be roughly estimated based on the signal bandwidth and the transmitted signal frequency through Fourier transform. frequency shift, and the technical solution of the present invention can directly and accurately estimate the signal center frequency, and then obtain the Doppler frequency shift of the echo; when the signal-to-noise ratio is low, the Fourier transform method and the signal frequency cannot be observed at all, and the present invention The technical solution can still estimate the Doppler frequency of the echo signal within a small error range.

Description of drawings

Fig. 1 is the algorithm flow chart of the present invention;

Fig. 2 is the spectrogram of the transmitted signal;

Fig. 3 is the echo signal spectrum diagram when the signal-to-noise ratio is 0dB;

Figure 4 is the spectrum of the transmitted signal after windowing in the frequency domain

Figure 5 is the frequency estimation result at 0dB;

Figure 6 is the echo signal spectrum diagram at -5dB;

Figure 7 shows the frequency estimation results at -5dB;

Figure 8 is the echo signal spectrum diagram at -10dB;

Figure 9 shows the frequency estimation results at -10dB.

Detailed ways

In Figure 1, the expression of the transmitted signal is as follows

In the formula μ=3*10 ¹³ Hz/s is the slope of frequency change, and the bandwidth is 60MHz. Its spectrogram is shown in Figure 2.

Suppose the expression of the intermediate frequency after A/D sampling of the received signal is as follows

x(t)=s(t)+n(t) (2)

in the formula is a useful signal, where f ₀ =100MHz is the signal carrier frequency, f _d =200MHz is the Doppler frequency shift, n(t) is Gaussian white noise, and the spectrum after adding Gaussian white noise with a signal-to-noise ratio of 0dB is shown in Figure 3 shown.

The spectrum of the transmitted signal is windowed in the frequency domain with the bandwidth as the scale, and the signal bandwidth range in Figure 2 is intercepted, and the result is shown in Figure 4.

Convolve the windowed spectrum with the spectrum of the echo signal in the frequency domain, and the result is shown in Figure 5. The frequency corresponding to the peak point is the Doppler frequency of the echo signal, 200MHz.

The present invention will be further described below with reference to specific embodiments.

Example 1:

In step 1, the center frequency of the transmitted signal is 100MHz, the pulse width is 2us, the bandwidth is 60MHz, the pulse repetition period is 10us, and the laser wavelength is 1500nm.

Step 2: Set the sampling rate to 1GHz, the analog signal-to-noise ratio to -5dB, and the Doppler frequency shift of the analog target to be 100MHz, that is, the speed of the analog target is 75m/s, and the spectrum of the echo signal is shown in Figure 6.

Step 3, adding a window with a bandwidth of 60 MHz to the spectrum of the transmitted signal in the frequency domain, and intercepting the effective spectrum.

Step 4: Convolve the windowed spectrum and the echo signal spectrum in the frequency domain, and the result is shown in FIG. 7 . The frequency measurement is 100MHz, and the frequency estimation error is 0Hz.

Step 5, do 500 times of Monte Carlo simulation, the average Doppler frequency detection result is 99999200Hz, and the average measurement error is 800Hz.

Example 2:

In step 1, the center frequency of the transmitted signal is 100MHz, the pulse width is 2us, the bandwidth is 60MHz, the pulse repetition period is 10us, and the laser wavelength is 1500nm.

Step 2: Set the sampling rate to 1GHz, the analog signal-to-noise ratio to -10dB, and the analog target Doppler frequency shift to 55MHz, that is, the analog target speed is 41.25m/s. The echo signal spectrum is shown in Figure 8. At this time, the spectrum cannot be observed. valid range.

Step 3, adding a window with a bandwidth of 60 MHz to the spectrum of the transmitted signal in the frequency domain, and intercepting the effective spectrum.

Step 4: Convolve the windowed spectrum and the echo signal spectrum in the frequency domain, and the result is shown in FIG. 9 . The frequency measurement is 100MHz, and the frequency estimation error is 0Hz.

Step 5, do 500 times of Monte Carlo simulation, the average Doppler frequency detection result is 55005600Hz, and the average measurement error is 5600Hz.

Claims (2)

1. A target echo Doppler frequency estimation method for pulsed laser radar, in which the transmit signal and the echo signal of the pulsed laser radar are both linear frequency modulation signals, and the method steps include: Step 1: After the receiver receives the echo signal, it first performs frequency-selective filtering and amplification to filter out noise and clutter outside the signal frequency band; then performs frequency mixing and band-pass filtering and amplification to convert the radio frequency signal into an intermediate frequency signal; Step 2: A/D conversion is performed on the intermediate frequency signal of step 1, and then digital down-conversion is performed after being converted into a digital signal, and two baseband signals of I and Q are obtained after quadrature mixing and low-pass filtering; Step 3: Synthesize the two-way baseband signals of Step 2 into a complex signal, and select a pulse repetition period with a length of f _s *PRI according to the sampling rate f _s of the receiver and the pulse repetition period PRI of the transmitted signal and the PRI synchronization signal. The complex signal is used as an echo signal, and then Fourier transform is performed to convert it from the time domain to the frequency domain; Step 4: Preprocess the original baseband transmit signal, perform Fourier transform on the baseband transmit signal with a pulse width of τ and convert it from the time domain to the frequency domain to obtain a spectrum signal, whose spectrum range is [-f _s /2, f _s /2], which is much larger than the bandwidth Bw of the signal, take the center frequency as the center and the bandwidth Bw as the scale to add a window in the frequency domain to intercept the signal within the bandwidth of the spectrum signal, and the window length is the bandwidth Bw; Step 5: Convolve the spectrum within the bandwidth of the transmitted signal intercepted in step 4 and the echo signal spectrum obtained in step 1 in the frequency domain. According to the convolution process, the frequency axis changes from -f _s /2 to f _s /2 to After -f _s /2～f _s /2+Bw, convert the coordinate axis of the convolution result; Step 6: Search for the maximum value on the frequency axis of the convolution result after the coordinate axis conversion. The coordinate of the frequency axis corresponding to the maximum value is the Doppler frequency shift of the echo signal plus the half bandwidth Bw/2, minus The corresponding frequency coordinate after Bw/2 is the Doppler frequency offset of the target echo. 2. a kind of target echo Doppler frequency estimation method for pulse laser radar as claimed in claim 1, it is characterized in that the signal after intermediate frequency signal carries out A/D conversion in described step 2 is: x(t)=s(t)+n(t) in: where f ₀ is the carrier frequency of the signal, f _d is the Doppler frequency shift, and n(t) is Gaussian white noise; The original baseband transmit signal in step 4 is: where μ is the slope of the frequency change.