A fmcw radar ranging method and system

Abstract

This invention discloses an FMCW radar ranging method and system, comprising: performing a Fourier transform on a discrete beat signal, and using the frequency at the peak of the Fourier transform spectrum as a coarsely estimated frequency f. FFT ; in f FFT Within the first preset range on both sides, the Fourier transform spectrum is refined by CZT to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine estimation frequency f. CZT ; in f CZT Within the second preset range on both sides, coherent CZT refinement is performed on the CZT spectrum to obtain the coherent CZT spectrum; the frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT Based on f CCZT Calculate the target distance. This invention is based on the characteristic that the phase and frequency of the beat signal of a linear frequency modulated continuous wave radar do not change independently. A phase-coherent signal is multiplied on the CZT spectrum of the beat signal. This signal is represented by frequency information, thus obtaining a noise-insensitive coherent CZT spectrum. At the same time, it can make full use of phase information and has no phase ambiguity problem, which greatly improves the ranging accuracy.

Description

An FMCW radar ranging method and system Technical Field

[0001] This invention belongs to the field of radar signal processing technology, and more specifically, relates to an FMCW radar ranging method and system. Background Technology

[0002] Frequency Modulation Continuous Wave (FMCW) radar has advantages such as long detection range, high range resolution, and susceptibility to external weather conditions such as rain, fog, and clouds. Therefore, it has wide applications in military, traffic control, vehicle collision avoidance, and liquid level measurement. It plays a significant role, especially in non-contact measurement applications such as liquid level measurement, deformation of precision parts, and respiratory and heart rate detection. These applications have high requirements for measurement accuracy. Therefore, it is of great significance to study a high-precision FMCW radar ranging method.

[0003] Existing FMCW radar ranging methods mainly include: amplitude ratio method, neighborhood spectrum refinement method, and phase method. The amplitude ratio method uses the peak value and two spectral lines on either side of the peak to interpolate and correct the peak position, thereby estimating the frequency of the beat signal for ranging. This method is computationally simple, but due to the limited number of reference points, it often has large errors when correcting the peak position, resulting in low ranging accuracy. Furthermore, the ranging accuracy is easily affected by the target distance. The neighborhood spectrum refinement method mainly interpolates the FFT or CZT spectrum to reduce the spectral spacing, thus refining the spectrum and more accurately obtaining the peak position. Compared to the amplitude ratio method, the ranging accuracy is improved. However, it utilizes the frequency information of the beat signal, which is not sensitive enough to changes in target distance; therefore, the measurement accuracy remains low. The phase method measures distance by estimating the phase information of the beat signal. Compared to frequency information, phase information is more sensitive to changes in distance. Therefore, the phase method has higher theoretical measurement accuracy than the neighborhood spectrum refinement method. However, the phase estimation result of the beat signal may not be the true phase, but may be the result of taking the true phase modulo 2π, which has a phase ambiguity problem. Therefore, the ranging result may have errors and the accuracy is still low. Summary of the Invention

[0004] In view of the above-mentioned defects or improvement needs of the prior art, the present invention provides an FMCW radar ranging method and system to solve the technical problem of low ranging accuracy of FMCW radar in the prior art.

[0005] To achieve the above objectives, in a first aspect, the present invention provides an FMCW radar ranging method, comprising the following steps:

[0006] S1. Sample the beat signal after mixing the target-reflected echo signal received by the FMCW radar with its transmitted signal to obtain the discrete beat signal;

[0007] S2. Perform a Fourier transform on the discrete beat signal to obtain its Fourier transform spectrum; use the frequency at the peak of the Fourier transform spectrum as the coarsely estimated frequency f. FFT ;

[0008] S3, in the rough estimation of frequency f FFT Within the first preset range on both sides, the Fourier transform spectrum is refined by CZT to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine estimation frequency f. CZT ;

[0009] S4. Estimate the frequency f in detail. CZT Within the second preset range on both sides, the CZT spectrum is further refined using CZT, then multiplied by the phase expression in frequency terms. The real part of the result is taken to obtain the coherent CZT spectrum. The frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT ;

[0010] S5. Based on the final frequency estimation result f CCZT Calculate the distance between the FMCW radar and the target.

[0011] More preferably, step S3 above includes: coarsely estimating the frequency f FFT ±B on both sides CZT Within a range of / 2, the Fourier transform spectrum is refined by CZT to obtain the CZT spectrum;

[0012] Among them, B CZT =f s / N;f s The sampling rate for sampling the beat signal in step S1 is N, where N is the number of corresponding sampling points.

[0013] More preferably, step S4 above includes: finely estimating the frequency f CZT ±B on both sides CCZT Within the range of / 2, the CZT spectrum is further refined using CZT;

[0014] Among them, B CCZT =Mf s / NK1; The value of M is determined based on the confidence principle and the Cramer-Rao bound of frequency estimation; f s N is the sampling rate for sampling the beat signal in step S1, and N is the number of corresponding sampling points; K1 is the number of refinement points for CZT refinement in step S3.

[0015] More preferably, f CCZT Falling into [f CZT -βσ,f CZT The confidence level within the range of +βσ is [value].

[0016] in, Indicates rounding up; β is a preset constant; σ represents the Cramer-Rao bound of the frequency estimation, and its expression is:

[0017]

[0018] SNR is the signal-to-noise ratio of the echo signal.

[0019] More preferably, the phase expression expressed in terms of frequency is as follows:

[0020]

[0021] Where f0 is the start frequency of the transmitted signal; B is the modulation bandwidth of the transmitted signal; f s N is the sampling rate for sampling the beat signal in step S1, and N is the number of corresponding sampling points. f1′ is the starting refinement frequency for further CZT refinement of the CZT spectrum in step S4; k = 0, 1, ..., K2; K2 is the number of refinement points for further CZT refinement of the CZT spectrum in step S4.

[0022] More preferably, in step S2 above, the discrete beat signal is subjected to FFT transformation to obtain the Fourier transform spectrum of the discrete beat signal.

[0023] More preferably, the distance between the FMCW radar and the target is:

[0024]

[0025] Where c is the speed of light; T is the frequency sweep repetition period of the transmitted signal; and B is the modulation bandwidth of the transmitted signal.

[0026] In a second aspect, the present invention provides an FMCW radar ranging system, comprising: a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform an FMCW radar ranging method provided in the first aspect of the present invention.

[0027] Thirdly, the present invention provides an FMCW radar ranging system, comprising:

[0028] The sampling module is used to sample the beat signal after the target reflected echo signal received by the FMCW radar is mixed with its transmitted signal to obtain the discrete beat signal.

[0029] The frequency coarse estimation module performs a Fourier transform on the discrete beat signal to obtain its Fourier transform spectrum; the frequency at the peak of the Fourier transform spectrum is used as the coarse estimated frequency f. FFT ;

[0030] The first frequency fine estimation module is used to perform coarse frequency estimation f. FFT Within the first preset range on both sides, the Fourier transform spectrum is refined by CZT to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine estimation frequency f. CZT ;

[0031] The second frequency fine estimation module is used to finely estimate the frequency f. CZT Within the second preset range on both sides, the CZT spectrum is further refined using CZT, then multiplied by the phase expression in frequency terms. The real part of the result is taken to obtain the coherent CZT spectrum. The frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT ;

[0032] The distance calculation module is used to calculate the distance based on the final frequency estimation result f. CCZT Calculate the distance between the FMCW radar and the target.

[0033] In summary, the above-described technical solutions conceived in this invention can achieve the following beneficial effects:

[0034] 1. This invention provides an FMCW radar ranging method. Based on the characteristic that the phase and frequency of the beat signal of a linear frequency modulated continuous wave radar do not change independently, a phase-coherent signal is multiplied by the CZT spectrum of the beat signal. This signal is represented by frequency information. The real part of the result of the multiplication is taken to obtain the coherent CZT spectrum. The coherent CZT spectrum is not sensitive to noise and can make full use of the phase information of the beat signal. Furthermore, this invention is based on the frequency estimation result of the beat signal rather than the phase to obtain the target distance. Since the mapping relationship between frequency and distance is one-to-one, there is no ambiguity problem, resulting in higher frequency estimation accuracy and greatly improving the accuracy of FMCW radar ranging.

[0035] 2. The FMCW radar ranging method provided by this invention performs CZT refinement and coherent CZT refinement sequentially after obtaining the Fourier spectrum. Compared to performing only a single CZT refinement, the two refinement processes further reduce the number of refinement points, thereby improving the calculation speed. Furthermore, derivation reveals that the coherent CZT spectrum is equivalent to multiplying the CZT amplitude spectrum by a cosine term. Therefore, the coherent CZT spectrum will have multiple peaks within the main lobe of the CZT amplitude spectrum. Without the first CZT refinement, due to the discretization of the cosine function, it is difficult to locate the largest peak position with a limited number of refinement points. If the number of refinement points is inappropriate, the largest peak may not even appear. This invention performs a CZT refinement before the coherent CZT refinement, enabling more accurate location of the interval containing the main lobe peak position of the coherent CZT spectrum. Attached Figure Description

[0036] Figure 1 is a flowchart of the FMCW radar ranging method provided in Embodiment 1 of the present invention;

[0037] Figure 2 is a schematic diagram of the FFT amplitude spectrum, CZT amplitude spectrum and coherent CZT amplitude spectrum provided in Embodiment 1 of the present invention;

[0038] Figure 3 is a schematic diagram of the CZT spectrum obtained by the existing improved CZT algorithm in a scenario without Gaussian noise and the coherent CZT spectrum obtained in this invention, provided in Embodiment 1 of this invention; wherein, (a) is the CZT spectrum obtained by the existing improved CZT algorithm in a scenario without Gaussian noise; Figure (b) is a schematic diagram of the coherent CZT spectrum obtained in this invention in a scenario without Gaussian noise.

[0039] Figure 4 is a schematic diagram of the CZT spectrum obtained by the existing improved CZT algorithm in a scenario with Gaussian noise and the coherent CZT spectrum obtained in the present invention, provided in Embodiment 1 of the present invention; wherein, (a) is the CZT spectrum obtained by the existing improved CZT algorithm in a scenario with Gaussian noise; (b) is a schematic diagram of the coherent CZT spectrum obtained in the present invention in a scenario with Gaussian noise.

[0040] Figure 5 is a schematic diagram showing the distance measurement results obtained by calculating the target distance in 200 experiments using an existing improved CZT method, a PDC algorithm utilizing FMCW radar phase information, and the FMCW radar ranging method provided in this invention.

[0041] Figure 6 is a schematic diagram showing the distance offset obtained by calculating the target distance in 200 experiments using an existing improved CZT method, a PDC algorithm that utilizes FMCW radar phase information, and the FMCW radar ranging method provided in this invention.

[0042] Figure 7 is a schematic diagram showing the root mean square error of distance measurement obtained by calculating the target distance in 200 experiments using an existing improved CZT method, a PDC algorithm utilizing FMCW radar phase information, and the FMCW radar ranging method provided in this invention. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0044] Example 1

[0045] This embodiment provides an FMCW radar ranging method for measuring the distance between an FMCW radar and a target. The FMCW radar ranging method provided in this embodiment is applicable to the field of high-precision ranging with linear frequency modulated continuous wave radar, and the ranging accuracy can reach the micrometer level.

[0046] In this embodiment, the initial target is located on an ultra-high precision micro-motion platform 0.73m away from the radar panel. It moves 200 times in 50μm steps. Each movement is measured using the FMCW radar ranging method provided in this embodiment. During the first 100 measurements, the target moves away from the radar panel, and during the next 100 measurements, the target moves closer to the radar panel. The radar panel parameters are set as follows:

[0047] Table 1

[0048] Symbol Explanation: Value f0 (GHz), Starting Frequency, 60s (MHz / μs), Slope 33.71T idle (μs) Idle time 7T ramp (μs) Ramp time 114.4N chirp Number of chirps in one frame: 1N sample Number of sampling points within a chirp: 512f s (Hz) Sampling rate 5000T frame (ms) The duration of one frame 100 surface

[0049] After obtaining the experimental data, signal processing was performed using the FMCW radar ranging method provided in this implementation via Matlab, as shown in Figure 1. The specific process is as follows:

[0050] S1. The beat signal obtained by mixing the target-reflected echo signal received by the FMCW radar with its transmitted signal is sampled at N points to obtain a discrete beat signal; the sampling rate is f.s ;

[0051] Specifically, the expression for the discrete beat signal is:

[0052]

[0053] Among them, f r The frequency of the beat signal; The phase of the beat signal; t d denoted as , where is the time delay between the radar echo and the transmitted signal; B is the modulation bandwidth of the transmitted signal; and f0 is the starting frequency of the transmitted signal.

[0054] S2. Perform a Fourier transform on the discrete beat signal to obtain its Fourier transform spectrum; use the frequency at the peak of the Fourier transform spectrum as the coarsely estimated frequency f. FFT ;

[0055] In this embodiment, an N-point FFT transform is performed on the discrete beat signal to obtain the Fourier transform spectrum (i.e., the FFT spectrum) of the discrete beat signal:

[0056]

[0057] Where k = 0, 1, ..., N.

[0058] Find the position k of the maximum value of the FFT amplitude spectrum max =Bt d Its corresponding frequency is

[0059] Specifically, the obtained FFT amplitude spectrum is shown as the circled line segment in Figure 2. It can be seen that its spectral line density is sparse, and it can only determine the approximate range of the true frequency position.

[0060] S3, in the rough estimation of frequency f FFT Within the first preset range on both sides, the Fourier spectrum is refined by K1-point CZT to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine-estimated frequency f. CZT ;

[0061] Specifically, the aforementioned first preset range can be selected based on experience. To obtain a more accurate fine-estimated frequency, in one optional implementation, the coarse-estimated frequency f... FFT Both sides ±B CZT Within a range of / 2, the Fourier transform spectrum is refined using K1-point CZT to obtain the CZT spectrum; where B CZT =f s / N. Since the error of the FFT coarse estimation result will not exceed one frequency resolution, the true peak value is within one frequency resolution range. Therefore, preferably, in this embodiment, the error of the FFT coarse estimation result f FFT CZT refinement is performed within the Δf range on both sides, that is, CZT refinement is performed within one frequency resolution, i.e., B CZT =f s / N=Δf.

[0062] In this embodiment, K1 is set to 256, and the frequency resolution of the FFT spectrum is Δf = f s / N, therefore B CZT =Δf, at which point the initial refined frequency point of the CZT spectrum is:

[0063]

[0064] Interpolation frequency points are

[0065]

[0066] Therefore, the CZT spectrum is obtained as follows:

[0067]

[0068] Find the CZT amplitude spectrum |X(z) k The peak position of )| is k′ max =K1(Bt) d -f1 / B CZT The fine frequency estimation results are as follows:

[0069]

[0070] Specifically, the resulting CZT amplitude spectrum is shown by the forked line segment in Figure 2. It is within the interval B determined by the FFT spectrum. CZT Interpolation is performed within the bandwidth to obtain denser spectral lines, further approximating the true frequency position.

[0071] S4. Estimate the frequency f in detail. CZT Within the second preset range on both sides, the CZT spectrum is refined using K2-point CZT, multiplied by the phase expression in frequency terms, and the real part of the result is taken to obtain the coherent CZT spectrum; the frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT ;

[0072] Specifically, the aforementioned second preset range can also be selected based on experience. It should be noted that the frequency resolution after the first CZT refinement is Δf = f. s / NK1, generally speaking, the range of coherent CZT refinement should be within one frequency resolution unit, i.e., B. CCZT =Δf, but due to the multi-peak nature of the coherent CZT spectrum and the susceptibility of CZT spectrum peaks to noise, there may be cases where the maximum peak of the coherent CZT spectrum does not fall within the Δf range. Therefore, it is necessary to appropriately expand the refinement range. If the expansion range is too large, the interval between spectra will increase with the same number of refinement points; if the expansion range is too small, it cannot guarantee that the true peak position will be included. Therefore, in order to obtain the final frequency estimation result more accurately, in one optional implementation, the frequency f is further refined. CZT Both sides ±B CCZT Within a range of / 2, the CZT spectrum is refined using K2-point CZT; where B CCZT =Mf s The value of / NK1;M is determined by the confidence principle and the Cramer-Rao bound of frequency estimation. Here, K2 and K1 can be equal or unequal, it is not required.

[0073] Specifically, f CCZT Falling into [f CZT -βσ,f CZT The confidence level within the range of +βσ is [value]. in, Indicates rounding up; β is a preset constant; σ represents the Cramer-Rao bound (CRB) for frequency estimation, and its expression is:

[0074]

[0075] Wherein, SNR is the signal-to-noise ratio of the echo signal.

[0076] In this embodiment, K2 is set to 512. According to the normal distribution distribution theorem, the peak frequency is within [f CZT -6σ,f CZT The probability within the range of +6σ] is approximately 1. In this case, β is preferably 6, which can achieve a better estimation effect.

[0077] For the CZT spectrum in ±B CCZT Performing K2-point CZT refinement within the range of / 2, the resulting spectrum is as follows:

[0078]

[0079] Where, f1′=f CZT -BCCZT / 2 is the starting refinement frequency for further CZT refinement of the CZT spectrum in step S4.

[0080] It should be noted that the range resolution of FMCW radar is determined by the sweep bandwidth; the larger the bandwidth, the higher the range resolution. However, in practice, the bandwidth cannot be increased indefinitely, and excessive bandwidth will reduce the maximum measurement range of the radar. Therefore, improving the ranging accuracy of FMCW radar mainly lies in spectrum optimization. The beat signal is obtained by mixing the echo signal and the transmitted signal of FMCW radar. The frequency of the beat signal is related to the target distance and is equivalent to a noisy sine wave signal. Therefore, improving the ranging accuracy of FMCW radar is essentially a problem of frequency estimation of the noisy sine wave signal. Specifically, since the phase and frequency of the beat signal of linear frequency modulated continuous wave radar do not change independently, this invention multiplies the CZT spectrum of the beat signal by a phase-coherent signal. This signal is represented by frequency information. Taking the real part of the result of the multiplication yields the coherent CZT spectrum. Compared with the CZT amplitude spectrum, it is less sensitive to noise, fully utilizes the phase information of the beat signal without phase ambiguity, and has higher frequency estimation accuracy, which can greatly improve the ranging accuracy of millimeter-wave radar.

[0081] Furthermore, after obtaining the Fourier spectrum, this invention performs one CZT refinement and one coherent CZT refinement sequentially. Compared to performing only a single CZT refinement, this two-stage refinement process further reduces the number of refinement points, thereby improving computational speed. Additionally, derivation reveals that the coherent CZT spectrum is equivalent to multiplying the CZT amplitude spectrum by a cosine term. Therefore, the coherent CZT spectrum will contain multiple peaks within the main lobe of the CZT amplitude spectrum. Without the preceding CZT refinement, due to the discretization of the cosine function, it is difficult to locate the largest peak position with a limited number of refinement points. If the number of refinement points is inappropriate, the largest peak may not even appear. This invention performs a CZT refinement before the coherent CZT refinement, enabling more accurate location of the interval containing the main lobe peak position of the coherent CZT spectrum.

[0082] Specifically, the phase expression in terms of frequency is as follows:

[0083]

[0084] in,

[0085] After refining the CZT spectrum using the K2-point CZT method, the expression obtained by multiplying it by the phase expression in terms of frequency is:

[0086]

[0087] When X c (z k At the peak position, only the real part of the information is contained, and the imaginary part is zero. Discarding the imaginary part can also reduce the noise influence caused by the imaginary part. Therefore, the real part of the coherent spectrum is usually used for frequency estimation. Thus, the obtained coherent CZT spectrum is:

[0088]

[0089] Find the coherent CZT spectrum Re[X] c (z k The peak position k′ of )] m ′ ax The final frequency estimation result is as follows:

[0090]

[0091] Specifically, the resulting coherent CZT spectrum is shown by the asterisked line segment in Figure 2, which is within interval B determined by the CZT spectrum. CCZT Interpolation is performed within the range for further refinement. It can be seen that the coherent CZT spectrum has a negative spectrum due to the influence of the cosine term. If only the positive spectrum is considered, it can be seen that the spectrum of coherent CZT is sparser than the amplitude spectrum of CZT with the same number of points. This indicates that the spectrum of coherent CZT is less sensitive to noise, which also shows that the present invention has better performance than the traditional CZT algorithm.

[0092] To more clearly illustrate the advantages of the frequency estimation method based on coherent CZT spectrum in this invention, it is compared with an existing improved CZT algorithm (see S. Scherr, S. Ayhan, B. Fischbach, A. Bhutani, M. Pauli and T. Zwick, “An Efficient Frequency and Phase Estimation Algorithm With CRB Performance for FMCW Radar Applications”, IEEE Trans. Instrum. Meas., vol. 64, no. 7, pp. 1868-1875, 2015). Specifically, this invention magnifies the coherent CZT spectrum near the peak and the CZT spectrum obtained based on the improved CZT algorithm to show more details. As shown in Figure 3, Figure (a) shows the CZT spectrum obtained based on the improved CZT algorithm, and Figure (b) shows the coherent CZT spectrum obtained in this invention. Both the improved CZT algorithm and the coherent CZT spectrum obtained in this invention can find the correct peak in the discrete frequency domain, and the CZT spectrum has a denser concentration of discrete frequency points. However, the above analysis does not consider the influence of noise. In reality, additive noise is unavoidable. When the influence of white Gaussian noise is added to the beat signal, due to the similar height of several spurious peaks, the improved CZT algorithm can no longer accurately determine the position of the correct peak, as shown in Figure 4(a). Conversely, as shown in Figure 4(b), the method based on the coherent CZT spectrum provided by this invention is less affected by Gaussian noise and can still find the correct peak in the discrete frequency domain.

[0093] S5. Based on the final frequency estimation result f CCZT Calculate the distance between the FMCW radar and the target.

[0094] Specifically, the distance between the FMCW radar and the target is:

[0095]

[0096] Where c is the speed of light; T is the frequency sweep repetition period of the transmitted signal; and B is the bandwidth of the transmitted signal.

[0097] Furthermore, to verify the superior performance of the FMCW radar ranging method proposed in this invention, the improved CZT method, a PDC algorithm utilizing FMCW radar phase information (see Qi Guoqing and Jiaxinle, “Improvement of Measurement Accuracy of FMCW Level Radar”, in 2001 CIE International Conference on Radar Proceedings, 2001), and the FMCW radar ranging method provided in this invention were used to calculate the target distance in 200 experiments. In the first 100 experiments, the target gradually moved away from the radar, and the measured distance gradually increased. In the last 100 experiments, the target gradually moved closer to the radar, with each movement step being 50 μm.

[0098] The ranging results of the three methods are shown in Figure 5. It can be seen that the PDC algorithm cannot accurately distinguish each displacement. The improved CZT algorithm performs better than PDC, but is far inferior to the FMCW radar ranging method provided in this invention. The FMCW radar ranging method provided in this invention can clearly distinguish each displacement, and the calculated trajectory is consistent with the actual motion trajectory.

[0099] Figure 6 shows the distance offset results between two adjacent measurements for the three methods. The average distance offset for measurements 1-100 is 50 μm; the average distance offset for measurements 101-200 is -50 μm, which is consistent with the actual distance traveled. Furthermore, most adjacent distance offsets of the FMCW radar ranging method provided by this invention are close to 50 μm. However, the distance offsets of the other two algorithms exceed 100 μm, with the maximum distance offset approaching 300 μm.

[0100] The root mean square error of the distance measurement for the three methods is shown in Figure 7. It can be clearly seen from the figure that the root mean square error of the coherent CZT method is lower than that of the improved CZT method and the PDC method.

[0101] Example 2

[0102] An FMCW radar ranging system includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform an FMCW radar ranging method provided in Embodiment 1 of the present invention.

[0103] The relevant technical solutions are the same as in Embodiment 1, and will not be repeated here.

[0104] Example 3

[0105] An FMCW radar ranging system includes:

[0106] The sampling module is used to perform N-point sampling on the beat signal obtained by mixing the target reflected echo signal received by the FMCW radar with its transmitted signal, to obtain a discrete beat signal; the sampling rate is f. s ;

[0107] The frequency coarse estimation module performs a Fourier transform on the discrete beat signal to obtain its Fourier transform spectrum; the frequency at the peak of the Fourier transform spectrum is used as the coarse estimated frequency f. FFT ;

[0108] The first frequency fine estimation module is used to perform coarse frequency estimation f. FFT The Fourier spectrum is refined using CZT at K1 points within the first preset range on both sides to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine-estimated frequency f. CZT ;

[0109] The second frequency fine estimation module is used to finely estimate the frequency f. CZT Within the second preset range on both sides, the CZT spectrum is refined using K2-point CZT, multiplied by the phase expression in frequency terms, and the real part of the result is taken to obtain the coherent CZT spectrum; the frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT ;

[0110] The distance calculation module is used to calculate the distance based on the final frequency estimation result f. CCZT Calculate the distance between the FMCW radar and the target.

[0111] The relevant technical solutions are the same as in Embodiment 1, and will not be repeated here.

[0112] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An FMCW radar ranging method, characterized in that, Includes the following steps: S1. Sample the beat signal obtained by mixing the target-reflected echo signal received by the FMCW radar with its transmitted signal to obtain a discrete beat signal; S2. Perform a Fourier transform on the discrete beat signal to obtain its Fourier transform spectrum; use the frequency at the peak of the Fourier transform spectrum as the coarsely estimated frequency f. FFT S3, in the coarsely estimated frequency f FFT Within a first preset range on both sides, the Fourier transform spectrum is refined using CZT to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine-estimated frequency f. CZT S4, in the detailed estimation of frequency f CZT Within the second preset range on both sides, the CZT spectrum is further refined using CZT, multiplied by the phase expression in frequency terms, and the real part of the result is taken to obtain the coherent CZT spectrum; the frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT S5. Based on the final frequency estimation result f CCZT Calculate the distance between the FMCW radar and the target.

2. The FMCW radar ranging method according to claim 1, characterized in that, Step S3 includes: in the coarsely estimated frequency f FFT ±B on both sides CZT Within a range of / 2, the Fourier transform spectrum is refined using CZT to obtain the CZT spectrum; where B CZT =f s / N;f s The sampling rate for sampling the beat signal in step S1 is N, where N is the number of corresponding sampling points.

3. The FMCW radar ranging method according to claim 1, characterized in that, Step S4 includes: in the fine estimation of frequency f CZT ±B on both sides CCZT Within a range of / 2, the CZT spectrum is further refined using CZT; where, B CCZT =Mf s / NK1; The value of M is determined based on the confidence principle and the Cramer-Rao bound of frequency estimation; f s The sampling rate for sampling the beat signal in step S1 is N, where N is the number of corresponding sampling points; K1 is the number of refinement points for CZT refinement in step S3.

4. The FMCW radar ranging method according to claim 3, characterized in that, f CCZT Falling into [f CZT -βσ,f CZT The confidence level within the range of +βσ is [value]. in, Indicates rounding up; β is a preset constant; σ represents the Cramer-Rao bound of the frequency estimation, and its expression is: SNR is the signal-to-noise ratio of the echo signal.

5. The FMCW radar ranging method according to any one of claims 1-4, characterized in that, The phase expression in terms of frequency is: Where f0 is the start frequency of the transmitted signal; B is the modulation bandwidth of the transmitted signal; f s N is the sampling rate for sampling the beat signal in step S1, and N is the number of corresponding sampling points. f1′ is the starting refinement frequency for further CZT refinement of the CZT spectrum in step S4; k = 0, 1, ..., K2; K2 is the number of refinement points for further CZT refinement of the CZT spectrum in step S4.

6. The FMCW radar ranging method according to any one of claims 1-4, characterized in that, In step S2, the discrete beat signal is subjected to FFT transformation to obtain the Fourier transform spectrum of the discrete beat signal.

7. The FMCW radar ranging method according to any one of claims 1-4, characterized in that, The distance between the FMCW radar and the target is: Where c is the speed of light; T is the frequency sweep repetition period of the transmitted signal; and B is the modulation bandwidth of the transmitted signal.

8. An FMCW radar ranging system, characterized in that, include: A memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the FMCW radar ranging method according to any one of claims 1-7.

9. An FMCW radar ranging system, characterized in that, include: The sampling module is used to sample the beat signal after the target reflected echo signal received by the FMCW radar is mixed with its transmitted signal to obtain the discrete beat signal. The frequency coarse estimation module is used to perform a Fourier transform on the discrete beat signal to obtain the Fourier transform spectrum of the discrete beat signal; the frequency at the peak of the Fourier transform spectrum is taken as the coarse estimation frequency f. FFT The first frequency fine estimation module is used to estimate the frequency f in the coarse estimation module. FFT Within a first preset range on both sides, the Fourier transform spectrum is refined using CZT to obtain the CZT spectrum; the frequency at the peak of the CZT spectrum is taken as the fine-estimated frequency f. CZT The second frequency fine estimation module is used to estimate the frequency f in the fine estimation. CZT Within the second preset range on both sides, the CZT spectrum is further refined using CZT, multiplied by the phase expression in frequency terms, and the real part of the result is taken to obtain the coherent CZT spectrum; the frequency at the peak of the coherent CZT spectrum is taken as the final frequency estimation result f. CCZT ; The distance calculation module is used to calculate the distance based on the final frequency estimation result f. CCZT Calculate the distance between the FMCW radar and the target.

Images