Target detection method based on fmcw-oam radar target imaging

Abstract

The application provides a target detection method based on FMCW-OAM radar target imaging. Since the vortex electromagnetic wave carries the orbital angular momentum OAM, more target information can be provided in radar imaging and target detection. The vortex phase is added in the traditional frequency-modulated continuous wave FMCW radar, the azimuth angle imaging performance of the FMCW radar is improved, the measurement accuracy is not affected by the actual azimuth angle of the target, the distance and the speed of the target, and the azimuth angle detection range is widened. Compared with the pulse OAM radar, the FMCW-OAM radar has the advantages of small short-distance detection blind area, high measurement accuracy, low power consumption and low sampling frequency.

Description

Technical Field

[0001] This invention relates to radar target detection technology, and particularly to target detection technology based on FMCW-OAM radar target imaging. Background Technology

[0002] Due to the introduction of orbital angular momentum (OAM), vortex electromagnetic EM waves propagate in a spiral manner, making their phase distribution spiral along the azimuth direction. This characteristic allows vortex electromagnetic waves to transmit more information than traditional plane waves. Therefore, vortex electromagnetic waves can bring back more information from the target within the beam radiation range without consuming a long coherence time or occupying a large spectral bandwidth. In recent years, vortex electromagnetic waves have received widespread attention and research in fields such as communication and radar imaging. In 2013, Li Liling et al. introduced OAM into diffraction tomography and realized super-resolution imaging of objects. [5] L.Li and F.Li, “Beating the rayleigh limit: Orbital-angular-momentum based super-resolution diffraction tomography,” Phys. Rev. E, vol. 88, p. 033205, Sep 2013. Liu Kang et al. established echo signal models for MIMO and MISO OAM radar imaging systems and successfully realized two-dimensional range-azimuth imaging. K. Liu, Y. Cheng, Z. Yang, H. Wang, Y. Qin, and X. Li, “Orbital-angular momentum-based electromagnetic vortex imaging,” IEEE AWP Letters, vol. 14, pp. 711–714, 2015. For OAM beams with different modes and different main lobes in different directions, the use of concentric circular arrays to calibrate the radiation position of the main lobes was investigated. K. Liu, Y. Cheng, X. Li, Y. Qin, H. Wang, and Y. Jiang, “Generation of orbital angular momentum beams for electromagnetic vortex imaging,” IEEE AWP Letters, vol. 15, pp. 1873–1876, 2016. These studies expanded and improved the application of vortex electromagnetic waves in the field of staring imaging. Vortex electromagnetic waves are also used in SAR imaging to improve the azimuth resolution of SAR. J.Wang, K.Liu, Y.Cheng, and H.Wang, “Vortex SAR imaging method based on oam beams design,” IEEE Sensors Journal, vol.19, no.24, pp.11 873–11 879, 2019.Advanced techniques were employed to achieve eddy current electromagnetic wave imaging using compressed sensing. Compared to previous range-Doppler (RD) algorithms, this method effectively addresses the azimuth profile problem under high sidelobe levels. (Y. Zeng, Y. Wang, C. Zhou, J. Cui, J. Yi, and J. Zhang, “Super-resolution electromagnetic vortex sar imaging based on compressed sensing,” in 2020 IEEE / CIC International Conference on Communications in China (ICCC), 2020, pp. 629–633.)

[0003] Currently, most radar imaging systems that use vortex electromagnetic waves are pulse radars, also known as OAM radars. OAM radars transmit high-energy pulse signals, requiring the transmitter to have high power. Furthermore, pulse radars need to directly process the echo signals, placing high demands on the sampling frequency of the receiver. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a target detection method that uses continuous frequency modulation waves to realize vortex phase and establishes an FMCW-OAM radar system to improve the imaging performance of existing radars.

[0005] The technical solution adopted by this invention to solve the above-mentioned technical problems is to provide a target detection method based on FMCW-OAM radar target imaging, comprising the following steps:

[0006] 1) Signal transmission steps:

[0007] The FMCW-OAM radar imaging model uses a uniform circular array UCA with N array elements to generate FMCW-OAM signals. An XYZ coordinate system is established with the array as the origin. The nth element transmits a signal s in the l-th vortex electromagnetic wave mode within a frequency modulation period at the t-th time unit. n (t;l) is:

[0008]

[0009] f c It is the carrier frequency, and K represents the frequency modulation rate of the LFM signal. The phase offset is set to obtain the vortex phase, and is the azimuth angle of each array element. n = 0, ..., N-1, l = 0, ..., L-1, where L is the total number of modes of the FMCW-OAM signal;

[0010] The total transmitted signal of a uniform circular array UCA is

[0011] The transmitter simultaneously transmits L different modes of FMCW-OAM signals;

[0012] 2) Signal reception dechirping steps:

[0013] The FMCW-OAM radar imaging model uses a single-antenna receiver to receive the echo signal from the target and combine it with... The mixed signal s′(t;l) is obtained by mixing; then the sampling frequency f is used to obtain the mixed signal s′(t;l); s Sampling s′(t;l) yields the following signal F(ω;l): Among them, J l For l-order Bessel functions, intermediate quantity a is the radius of the uniform circular array UCA, R T , and θ T These represent the target's distance, azimuth, and elevation angles relative to the radar center, respectively. This represents the frequency shift caused by the linear Doppler frequency, an intermediate quantity. X T Y T The distribution represents the target's X and Y coordinates in the coordinate system, and v is the target's velocity. The angle between the target's velocity and the positive X-axis direction is represented by λ; λ is the carrier frequency f. c The corresponding wavelength, sample number variable p = 0, 1...P-1, where P is the number of samples within the frequency modulation cycle;

[0014] 3) Data storage steps:

[0015] The receiver of the FMCW-OAM radar imaging model samples P times in each scan cycle. A complete detection cycle consists of M scan cycles. One detection yields L*P*M received data samples, which are stored in a three-dimensional data matrix. The sampled data in one scan cycle is stored along the row direction of the matrix; data sampled from different modal signals is stored in different columns of the matrix; and data sampled from different frequency modulation cycles is stored in different pages.

[0016] 4) Target distance measurement steps:

[0017] Calculate the frequency domain data F(ω;l) by performing an FFT along the rows of the three-dimensional data matrix (i.e., p-dimensional).

[0018]

[0019] ω is the frequency domain variable after p-dimensional FFT transformation within the period, calculated by... Obtain the frequency domain value ω T ;

[0020] Get the distance to the target Where B is the bandwidth of the baseband signal and c is the speed of light;

[0021] 5) Target azimuth measurement steps:

[0022] The target's azimuth information (PSF) is obtained by performing an FFT along the l-dimensional column of the three-dimensional data matrix on the data sample. Among them, l m The OAM mode range of the transmitted signal is [-l m ,l m ],

[0023] 6) Target velocity measurement steps:

[0024] The frequency domain data is obtained by performing an FFT on the three-dimensional data matrix of the data samples obtained from M scan cycles in the periodic dimension m.

[0025]

[0026] Ω represents the frequency domain variables after FFT transformation in m-dimensional periodicity, where m = 0, ..., M-1. These variables are calculated... Obtain the frequency domain value Ω T ;

[0027] Obtain the actual speed of the target

[0028] The beneficial effects of this invention are:

[0029] 1) Improve the performance of pulse OAM radar:

[0030] Currently, most radar imaging systems using vortex electromagnetic waves are pulse radars. Compared to traditional pulse OAM radars, FMCW radar systems transmit periodic continuous waves instead of high-energy pulse signals, which significantly reduces the average power of the transmitter. Furthermore, FMCW radar obtains target information by analyzing the spectrum of the beat frequency signal obtained by mixing the transmitted and echo signals, unlike pulse radar which directly processes the echo signal. Therefore, it significantly reduces the sampling frequency requirements of the receiver. This invention innovatively uses continuous frequency modulated waves to achieve vortex phase, establishing an FMCW-OAM radar system that gives OAM radar the advantages of continuous wave radar, such as small short-range detection blind zone, high measurement accuracy, low power consumption, and low sampling frequency.

[0031] 2) Improve the azimuth imaging performance of traditional FMCW radar:

[0032] Traditional FMCW radar systems typically employ multi-channel receivers to measure target angles, using methods such as phase comparison and the MUSIC algorithm. However, the angular resolution of the phase comparison method decreases with increasing arrival signal angle, and it also has a maximum observation angle limitation. Furthermore, ambiguity arises when processing multiple targets with the same range and velocity. The MUSIC algorithm involves matrix operations, thus requiring significant computation for peak search. This paper combines FMCW technology with OAM beamforming, resulting in FMCW-OAM signals offering more degrees of freedom compared to traditional FMCW signals. Based on the Fourier transform pair between the mode and azimuth of vortex electromagnetic waves, a Fast Fourier Transform (FFT) can be performed in the mode domain to extract the target azimuth. This method is unaffected by the correlation between the actual target azimuth, range, and velocity, and broadens the azimuth detection range. Moreover, compared to other algorithms, FFT-based computation reduces the overall computational load. Attached Figure Description

[0033] Figure 1 OAM-FMCW radar imaging model

[0034] Figure 2 Schematic diagram of data storage mode

[0035] Figure 3 Here are schematic diagrams of range-azimuth 2D combined imaging; (a) range-azimuth 2D combined imaging, (b) range-velocity 2D combined imaging, (c) magnified result of range-azimuth 2D combined imaging, and (d) magnified result of range-velocity 2D combined imaging.

[0036] Figure 4 This is a multi-target imaging result;

[0037] Figure 5 The diagram illustrates the comparison of effects; (a) comparison of azimuth measurement errors, and (b) comparison of measurement results for targets in different azimuths.

[0038] Figure 6 This is a schematic diagram showing how the resolution varies with the number of antenna elements; (a) the element spacing is fixed, and (b) the total aperture is fixed. Detailed Implementation

[0039] The signal processing procedure, echo signal model, FMCW-OAM system model, and derivation of the resolution of measurement parameters for target detection using vortex electromagnetic waves in FMCW radar are as follows:

[0040] STEP 1: Transmit signal

[0041] FMCW-OAM radar imaging model as follows Figure 1As shown, a uniform circular array (UCA) can be used to generate FMCW-OAM signals. The radius of the uniform circular array (UCA) is 'a'. The radar remains stationary, and the target moves within the detection area. A Cartesian coordinate system O-XYZ, R is established. T , and θ T Let X represent the target's distance, azimuth, and elevation angles relative to the radar center, respectively. H is the known altitude of the radar. T ,Y T () represents the coordinates of the target's position. The target's velocity is v. This represents the angle between the target's velocity and the positive X-axis.

[0042] The feed current to each element in the array is a linear sawtooth frequency-modulated signal with a specific phase offset applied. During the frequency modulation period T, the signal transmitted by the nth element is represented as...

[0043]

[0044] Where t is the time variable within one frequency modulation period, l is the mode number, and f c It is the carrier frequency, and K represents the frequency modulation rate of the LFM signal. This is a phase offset set to obtain the eddy current phase. The azimuth angle of each array element N is the total number of array elements. Therefore, the total transmitted signal of N array elements is:

[0045]

[0046] STEP2: Receive signal

[0047] The signal at the target point can be written as

[0048]

[0049] Among them, J l For l-order Bessel functions, intermediate quantity a is the radius of the uniform circular array UCA, t′ n τ and τ represent the time delays for the target to reach the nth antenna element and the center of the circular array, respectively.

[0050] STEP 3: Dechirping

[0051] The receiver uses a single-antenna reception scheme, combining the received signal with... The baseband signal obtained by mixing is

[0052]

[0053] Since the reception is a two-way distance, the time delay τ is

[0054]

[0055] Among them, intermediate quantity Substitute (5) into (4) and compensate for constant amplitude (N terms) and phase. The mixed signal can then be represented as

[0056]

[0057] (6) The index contains t 2 If the mixing term is omitted, the resulting signal is still a frequency-modulated (FM) signal. However, due to the short sampling time interval and low target speed, the modulation rate is small, meaning the frequency change after mixing is very small relative to the signal phase. Therefore, the FM term can be ignored, and the mixed signal can be approximated as a single-frequency signal. Simultaneously, it is observed that some terms in the equation need to be divided by the square of the speed of light; compared to other terms, the speed of light is a very small value and can also be ignored. Therefore, the signal can be further rewritten as...

[0058]

[0059] It can be seen that the frequency of the signal to be sampled is significantly lower than that of the echo signal. Taking the sampling frequency f... s The signal after sampling is

[0060]

[0061] in This represents the frequency shift caused by the linear Doppler frequency. s λ and f represent the sampling frequency and center frequency (carrier frequency), respectively. c The corresponding wavelengths are p = 0, 1...P-1, where P is the number of samples within the frequency modulation cycle.

[0062] STEP4: Data Storage

[0063] Assume that a UCCA radar simultaneously transmits L different modes of FMCW-OAM signals, each detection signal consisting of M scan cycles. The receiver samples P times in each cycle to obtain L*P*M data samples. These data are... Figure 2 The data is stored in a three-dimensional matrix in the manner shown, i.e., data sampled within one period is stored along the rows of the matrix; data sampled from different modal signals is stored in different columns of the matrix; and data sampled from different frequency modulation periods is stored in different pages.

[0064] STEP 5: Measure the target distance

[0065] Calculating the FFT along the rows of the three-dimensional data matrix, i.e., in p dimensions, yields...

[0066]

[0067] ω is the frequency domain variable after p-dimensional FFT transformation within the period. Let the frequency domain value ω T satisfy The target's distance information and the resolution in the distance direction are obtained.

[0068]

[0069]

[0070] Where B is the bandwidth of the baseband signal. When the following conditions are met...

[0071]

[0072] The target is considered to be moving at a low speed, and the range shift caused by the Doppler frequency is negligible.

[0073] STEP 6: Measure the target azimuth angle

[0074] As can be seen from (3), mode l and azimuth form a Fourier transform pair. Therefore, within one modulation period, the target azimuth can be obtained by calculating the FFT along the mode dimension. Since the radar altitude is fixed, the target elevation can be calculated by combining the target range. Using the target elevation angle information, the influence of the Bessel function on the signal amplitude can be compensated. Then, the target azimuth information is obtained by performing an FFT along the columns of the matrix, i.e., the l-dimensional dimension.

[0075] The azimuth resolution can be measured using the point spread function (PSF). Assume the transmitted OAM mode range is [-l m ,l m Furthermore, under the condition that the influence of the Bessel function on the echo signal amplitude can be compensated, the PSF can be expressed as...

[0076]

[0077] The azimuth resolution can be obtained as follows:

[0078]

[0079] STEP7: Measure the target speed

[0080] To measure target velocity, the radar needs to transmit multiple periods of signal. Assuming M periods of echo signal have been obtained, performing an FFT on the sampled data for each period yields the spectral peak for each period. Since the target's initial position changes in each period, the peak value in the m-th period can be expressed as...

[0081]

[0082] In the above analysis, because the duration of a single cycle is very short, the target's azimuth angle does not change significantly. When considering more frequency modulation cycles, the azimuth angle change caused by the target's motion should be taken into account, i.e., in the above formula... It will change over time. It is stated that in the subsequent process... Let this be the initial azimuth angle. The function of azimuth angle changing with time can be described by the following formula:

[0083]

[0084] Among them, intermediate quantity Substituting (16) into (15), we get

[0085]

[0086] Next, the FFT is calculated in different periodic dimensions m, and the results are obtained.

[0087]

[0088] Ω represents the frequency domain variable after FFT transformation in m dimensions with different periods. Let the frequency domain value Ω T satisfy

[0089] Similarly, the target's actual velocity and velocity resolution will be solved as follows:

[0090]

[0091]

[0092] To obtain higher velocity resolution, we can select a suitable OAM mode signal for processing according to (20). After determining the mode, we can select the corresponding two-dimensional matrix from the three-dimensional matrix. Finally, we perform a two-dimensional FFT operation on the two-dimensional matrix to obtain the target velocity.

[0093] Simulation Experiment

[0094] A. Imaging Results

[0095] This section presents imaging results to verify the effectiveness of the proposed FMCW-OAM system. The main simulation parameters are shown in Table 1. A target is placed in the imaging scene at a distance R. T Azimuth With velocity v = (80m, 30°, 10m / s), and velocity direction... It is 30°.

[0096] Table 1 Simulation Parameters

[0097] parameter Value <![CDATA[Carrier frequency f c > 3GHz bandwidth 300MHz Frequency modulation rate 750GHz / s OAM mode [-20,20] Radar Altitude 50m Target scattering coefficient 1

[0098] Table 2 Target Parameters

[0099]

[0100] Figure 3 (a) and Figure 3 (b) shows the 2-D combined imaging results of the distance to target 2 and the target's azimuth and velocity measurements, respectively. Figure 3 (c) and Figure 3 (d) shows the corresponding magnified result. It can be seen that the system has high range and azimuth resolution and can accurately measure the target's velocity.

[0101] Figure 4 Imaging results for multiple targets are shown in Table 2, with target parameter settings as shown. Although the influence of the Bessel function term on amplitude is difficult to compensate for directly under multi-target conditions, the system can still exhibit good resolution when UCCA generates OAM at the transmitter.

[0102] B. Comparison and analysis of degree measurement performance

[0103] Traditional FMCW radar systems use the phase comparison method to measure target azimuth. Since there is a phase difference between the echoes received from different channels, and this phase difference is related to the target's azimuth angle, the phase comparison method can obtain the target's azimuth angle information by calculating the phase difference. If a uniform linear array is used to receive the signal, the phase difference between different array elements varies with the target azimuth angle, as shown below:

[0104]

[0105] Where ΔΦ represents the phase difference between channels, and d represents the element spacing on the uniform linear array. Due to the nonlinear relationship between the phase difference and the azimuth angle, the angle measurement accuracy of this method is affected by the target azimuth angle. The modes and azimuth angle of the OAM vortex wave form a Fourier transform pair, and the target azimuth angle can be obtained by calculating the FFT over the mode range. Figure 5 (a) shows the relationship between the measurement error and the actual azimuth angle for the two measurement methods described above. The basic simulation parameters of the OAM system are the same as those in Table 1. When simulating the phase comparison method, a uniform linear array is used, comprising a total of 41 elements, with a spacing of d = λ / 2 between each element. The carrier frequency is the same as that used in the OAM system. It is easy to see that the measurement error of the phase comparison method increases with the increase of the target azimuth angle. However, FMCW-OAM is not affected by the target azimuth angle.

[0106] Another drawback of the phase comparison method is its limited angle measurement range. Figure 5 (b) shows the results of target azimuth measurements using two methods. Figure 5 In (b), the horizontal axis represents the target's true azimuth angle, while the vertical axis represents the measurement result obtained using the corresponding method. As can be seen from the figure, if the target azimuth angle exceeds 60°, the measurement result obtained by the phase comparison method appears to be incorrect because the distance between the antennas has been modified. As the distance between antennas increases, the azimuth range measured by the phase comparison method decreases. In contrast, the proposed FMCW-OAM system can obtain accurate measurement results.

[0107] To further compare the performance of these two methods, Figure 6 The resolution of these two methods varies with the number of elements in the antenna array. Figure 6 (a) shows that when the element spacing remains constant, both methods can improve resolution as the number of array elements increases. This is because increasing the number of array elements leads to an increase in the total number of emitted OAM modes, thus improving the resolution of the OAM-based method. When the element spacing is fixed, increasing the number of elements will correspondingly increase the aperture of the linear array, thereby improving the resolution of the phase comparison method. Figure 6 (b) reflects that when the antenna aperture remains constant, the OAM-based method can effectively reduce the required antenna size. Since the antenna aperture remains constant, the resolution of the phase comparison method remains unchanged even when the number of array elements is increased. In addition, the resolution of the phase comparison method is related to the target azimuth angle. The larger the target azimuth angle, the weaker its resolution. This is due to the nonlinear relationship in equation (21).

[0108] Simulation results show that the imaging method proposed in this invention can obtain the range, azimuth, and velocity of multiple targets, and can improve the detection performance of traditional FMCW radar in azimuth.

Claims

1. A target detection method based on FMCW-OAM radar target imaging, characterized in that, Including the following steps: 1) Signal transmission steps: The FMCW-OAM radar imaging model uses a uniform circular array (UCA) with N elements to generate FMCW-OAM signals. An XYZ coordinate system is established with the array as the origin. The nth element transmits a signal in the l-th vortex electromagnetic wave mode within a frequency modulation period at the t-th time unit. for: ， ； It is the carrier frequency. Represents the frequency modulation rate of the LFM signal. The phase offset is set to obtain the vortex phase, and is the azimuth angle of each array element. n=0,…,N-1, l=0,…, L–1, where L is the total number of modes of the FMCW-OAM signal; The total transmitted signal of a uniform circular array UCA is ; The transmitter simultaneously transmits L different modes of FMCW-OAM signals; 2) Signal reception dechirping steps: The FMCW-OAM radar imaging model uses a single-antenna receiver to receive the echo signal from the target and combine it with... Mixing yields the mixed signal Represented as ,in, for Bézier function of order, intermediate quantity , Let U be the radius of the uniform circular array UCA. , and These represent the target's distance, azimuth, and elevation angles relative to the radar center, respectively. The time delay for a two-way trip, intermediate quantity , , The distribution represents the target's X and Y coordinates in the coordinate system. For the target speed, This represents the angle between the target's velocity and the positive X-axis. Then by sampling frequency right Sampling is performed, and the sampled signal is obtained. for ,in, This represents the frequency shift caused by the linear Doppler frequency. carrier frequency Corresponding wavelength, sample number variable , It is the number of samples within the frequency modulation cycle; 3) Data storage steps: The receiver of the FMCW-OAM radar imaging model samples P times in each scan cycle. A complete detection cycle consists of M scan cycles. One detection yields L*P*M received data samples, which are stored in a three-dimensional data matrix. The sampled data in one scan cycle is stored along the row direction of the matrix; data sampled from different modal signals is stored in different columns of the matrix; and data sampled from different frequency modulation cycles is stored in different pages. 4) Target distance measurement steps: Calculate the frequency domain data by performing an FFT along the rows of the three-dimensional data matrix (i.e., p-dimensional). : ； To represent the frequency domain variables after a p-dimensional FFT transformation within a period, the calculation... Obtain frequency domain values ; Get the distance to the target ,in, The bandwidth of the baseband signal. The speed of light; 5) Target azimuth measurement steps: For data samples along the columns of the three-dimensional data matrix The target's azimuth information is obtained by performing an FFT on the target. ,in, OAM mode range of the transmitted signal , The azimuth angle is given by the given information, and PSF is the point spread function. 6) Target velocity measurement steps: The frequency domain data is obtained by performing an FFT on the three-dimensional data matrix of the data samples obtained from M scan cycles in the periodic dimension m. , ； in, For the frequency domain variables after FFT transformation in different periodic dimensions m, m=0,…,M-1, calculate… Obtain frequency domain values ; As an intermediate quantity, ; Obtain the actual speed of the target .

2. The method as described in claim 1, characterized in that, Resolution in the distance direction ; Azimuth resolution Speed ​​resolution .

Images