A multi-mode deformation monitoring method based on millimeter wave radar

Abstract

The present invention discloses a multi-mode deformation monitoring method based on millimeter wave radar, including: selecting a radar imaging mode according to demand; if the real-time requirement for deformation monitoring is higher than a preset condition, selecting a real aperture radar imaging mode: using a millimeter wave radar with a multi-input multi-output system, using its echo signal mixing, windowing, FFT transformation, etc. to obtain a continuous multi-frame range-azimuth angle diagram of the observation area; if the accuracy of deformation monitoring is higher than a preset threshold, selecting a synthetic aperture radar imaging mode: carrying the millimeter wave radar on a ground-based track, using a back-projection algorithm for the echo signal to extract the range-azimuth angle diagram in each time period; finally interfering and unwrapping the range-azimuth angle diagrams of different frames / time periods to obtain the phase difference between two adjacent frames, and then obtaining the deformation amount between different frames / time periods. The present invention can select real aperture imaging synthetic aperture radar imaging according to demand to achieve flexible deformation monitoring.

Description

Technical Field

[0001] The invention belongs to the technical field of radar measurement, and in particular relates to a multi-mode deformation monitoring method based on millimeter wave radar. Background Art

[0002] Angular resolution is the minimum distance at which a radar can distinguish and differentiate between two targets of equal size at the same distance, or the ability to distinguish between two objects by angle alone. In millimeter-wave radar, the level of angular resolution directly affects the radar's perception accuracy of the surrounding environment and the imaging effect. When the angular resolution is low, it directly affects the extraction of target deformation information in the observation area. The angular resolution of millimeter-wave radar is related to the number of antennas. The more antennas there are, the higher the angular resolution. However, too many antennas will increase the cost and volume. In order to solve this problem, researchers use a multi-input multi-output system to improve the angular resolution. Although this system can improve the angular resolution, it increases the difficulty of signal processing.

[0003] At present, common deformation monitoring radars are mainly satellite-borne and ground-based. Satellite-borne deformation monitoring radars can obtain large-scale observations at one time, but it is not easy to change the observation range according to needs, and the re-entry cycle time is long, and real-time deformation monitoring cannot be achieved. Ground-based deformation monitoring radars have solved the problems of fixed satellite observation range and long re-entry cycle to a certain extent, but there are still problems such as large radar and orbit volume, which is inconvenient to carry. In addition, the imaging mode is single and the imaging processing time is long. Summary of the invention

[0004] The present invention provides a multi-mode deformation monitoring method based on millimeter wave radar, which can select real aperture imaging and synthetic aperture radar imaging according to needs to realize flexible deformation monitoring.

[0005] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0006] A multi-mode deformation monitoring method based on millimeter wave radar, comprising:

[0007] Step 1: Select the radar imaging method according to your needs:

[0008] If the real-time requirement for deformation monitoring is higher than the preset condition, the real aperture radar imaging method is selected;

[0009] If the accuracy of deformation monitoring is higher than the preset threshold, the synthetic aperture radar imaging method is selected;

[0010] Step 2: If the real aperture radar imaging mode is selected, then: a millimeter wave radar with a multi-input multi-output system is used, and its echo signal is mixed with the transmission signal to obtain an intermediate frequency signal, and the range-azimuth angle diagram of multiple consecutive frames of the observation area is obtained by processing the intermediate frequency signal;

[0011] Step 3, interfering and unwrapping the range-azimuth diagrams of each two adjacent frames obtained in step 2 to obtain the phase difference between the two adjacent frames, and then calculating the deformation amount of the observation area between the two adjacent frames based on the phase difference; and then extracting the deformation position based on the multivariate signal classification method;

[0012] Step 4: If the synthetic aperture radar imaging method is selected, then: the millimeter wave radar is mounted on a ground-based track, and a range-azimuth angle diagram is extracted by using a back-projection algorithm for the echo signal within a time period;

[0013] Step 5, interfering and unwrapping the range-azimuth diagrams of the two time periods to obtain the phase difference between the two time periods, and then calculating the deformation amount of the observation area between the two time periods based on the phase difference.

[0014] Furthermore, the echo signal is mixed with the transmission signal to obtain an intermediate frequency signal, which is expressed as:

[0015]

[0016]

[0017] In the formula, represents the intermediate frequency signal, Indicates the transmission signal. is the carrier frequency of the transmitted signal, To adjust the frequency, Fast time for signal; is the time delay for the transmitted signal to return from the target, is the distance from the radar to the target, The speed of light.

[0018] Furthermore, after the intermediate frequency signal is obtained by mixing, the intermediate frequency signal is first windowed, and then the windowed intermediate frequency signal is subjected to a fast FFT transformation to obtain the range-azimuth angle of multiple consecutive frames in the observation area; the windowing process is expressed as:

[0019]

[0020] In the formula, represents the windowed IF signal, is the length of the window function, Time index within the window function.

[0021] Furthermore, the intermediate frequency signal is subjected to fast Fourier transform to extract the target distance in the observation area, and the azimuth is estimated using the beamforming method.

[0022] Furthermore, the method based on multivariate signal classification extracts the deformation position, that is, extracts the azimuth of each deformation position with higher precision, specifically:

[0023] Perform a fast Fourier transform on the intermediate frequency signal and then calculate the covariance matrix:

[0024]

[0025] In the formula, is the number of snapshots, It is the signal obtained by fast Fourier transform of intermediate frequency signal. is the covariance matrix, represents the transpose operation;

[0026] Covariance matrix Perform eigenvalue decomposition to obtain the signal subspace and the noise subspace :

[0027]

[0028] In the formula, is the matrix composed of the first k eigenvalues ​​with the largest values ​​obtained from eigenvalue decomposition, and the matrix composed of the remaining eigenvalues ​​is recorded as ; The number of eigenvalues ​​k in is determined according to the number of azimuth angles estimated using the beamforming method;

[0029] Constructing spatial spectrum function using noise subspace :

[0030]

[0031] In the formula, is the steering vector, determined by the millimeter-wave radar;

[0032] By traversing all angles The angle with the largest peak value of the spatial spectrum function is found, which is the azimuth of the target with higher precision.

[0033] Furthermore, in step 4, a back-projection algorithm is used on the echo signal to extract the distance and azimuth, specifically:

[0034] Step 4.1, perform windowing processing on the echo signal within a time period:

[0035]

[0036] In the formula, is the echo signal of the transmitted signal of the mth transmitting antenna reflected by the target and received by the nth receiving antenna, is the windowed signal; For time; is the window function;

[0037] Step 4.2, for the windowed signal To perform distance pulse pressure:

[0038]

[0039] In the formula, is a matched filter, To adjust the frequency, is the range matched filter signal;

[0040] Step 4.3: Match the range-direction filtering signal Perform back-projection imaging processing to obtain the position coordinates of each target point in the observation area:

[0041]

[0042]

[0043] In the formula, Target point The signal strength, Target point The transmitted signal of the mth transmitting antenna passes through the target point The propagation time of the signal reflected and received by the nth receiving antenna, is the position coordinate of the mth transmitting antenna, is the position coordinate of the nth receiving antenna, is the location coordinate of the target point; M is the number of transmitting antennas, and N is the number of receiving antennas;

[0044] Step 4.4, extract the distance and azimuth according to the position coordinates of each target point in the observation area to obtain a distance-azimuth diagram.

[0045] Further, in step 4, the CPU is used to complete step 4.1 and step 4.2 and obtain a plurality of continuous distance matched filter signals. The task is divided into multiple small tasks according to the time frame, and CUDA is used to assign the multiple small tasks to multiple GPUs; then each GPU completes its own small task: extract the position coordinates of each target point in the observation area from each divided signal according to step 4.3; then the CPU is used to merge the position coordinates of all target points obtained from all small tasks to obtain the three-dimensional SAR imaging of the observation area; finally, the CPU is used to filter the three-dimensional SAR imaging and extract the range-azimuth diagram.

[0046] Furthermore, step 3 specifically includes:

[0047] Step 3.1: perform complex conjugate multiplication of the range-azimuth diagrams of each two adjacent frames to obtain the interference phase :

[0048]

[0049] In the formula, represents the phase operation, and Represents the range-azimuth diagram of two adjacent frames;

[0050] Step 3.2: Unwrap the interference phase to get the real phase difference ;

[0051] Step 3.3: Calculate the deformation between two adjacent frames based on the actual phase difference :

[0052]

[0053] In the formula, is the wavelength of millimeter radar waves.

[0054] Furthermore, step 5 specifically includes:

[0055] Step 5.1: Perform complex conjugate multiplication of the range-azimuth diagrams in two different time periods to obtain the interference phase :

[0056]

[0057] In the formula, represents the phase operation, and A range-azimuth diagram showing two different time periods;

[0058] Step 5.2: Unwrap the interference phase to get the real phase difference ;

[0059] Step 5.3: Calculate the deformation between two different time periods based on the true phase difference :

[0060]

[0061] In the formula, is the wavelength of millimeter radar waves.

[0062] The present invention takes a relatively small millimeter-wave radar as the object and invents a multi-mode deformation monitoring method, which can not only realize real aperture imaging monitoring, but also realize synthetic aperture radar imaging monitoring. Real aperture imaging monitoring mainly utilizes multiple-input multiple-output (MIMO) to realize high angular resolution imaging to monitor the target area. This method can quickly and efficiently obtain deformation information of the target area. Synthetic aperture radar imaging monitoring is to mount the millimeter-wave radar on a stable moving platform and synthesize a large aperture through motion to improve the imaging resolution. This method can obtain high-resolution radar images of the target area, but it will reduce the temporal resolution. The present invention supports the above two deformation monitoring modes, and the monitoring mode can be selected according to needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 It is a flow chart of the multi-mode deformation monitoring method based on millimeter-wave radar described in an embodiment of the present application. DETAILED DESCRIPTION

[0064] The following is a detailed description of an embodiment of the present invention. This embodiment is based on the technical solution of the present invention, and provides a detailed implementation method and a specific operation process to further explain the technical solution of the present invention.

[0065] This embodiment provides a multi-mode deformation monitoring method based on millimeter wave radar. Figure 1 As shown, the following steps are included:

[0066] Step 1: Select the radar imaging method according to the needs: if the real-time requirement for deformation monitoring is higher than the preset condition, select the real aperture radar imaging method of the multi-input multi-output system; if the accuracy of deformation monitoring is higher than the preset threshold, select the synthetic aperture radar imaging method.

[0067] Step 2: If the real aperture radar imaging method is selected, then: a millimeter wave radar with a multi-input multi-output system is used, and its echo signal is mixed with the transmission signal to obtain an intermediate frequency signal, and the intermediate frequency signal is processed to obtain a continuous range-azimuth angle diagram of multiple frames in the observation area.

[0068] At present, millimeter-wave radars with multiple-input multiple-output (MIMO) systems are mainly used for intelligent perception in the fields of robots and autonomous driving. Research is conducted on how to generate high-quality, high-density, and high-precision point cloud data and back-end applications, but research is rarely conducted on front-end signal processing. The present invention mainly generates high-resolution radar images by processing the front-end echo signals, and then interferes with the radar images to extract deformation information of the target area.

[0069] The range resolution of the radar is determined by the signal bandwidth. The larger the signal bandwidth, the higher the range resolution, which can reach the millimeter level at most. The azimuth resolution is determined by the number of antennas. The more antennas there are, the higher the angular resolution. Therefore, the present invention uses the MIMO system to virtualize multiple antennas to form a large-size aperture to improve the azimuth angular resolution. First, the fast Fourier transform (FFT) is used to complete the range imaging, and then the fast Fourier transform is used to complete the azimuth estimation. However, the resolution of the azimuth estimation based on FFT is low, so the beamforming (Beamforming, BF) method can be used to complete the azimuth estimation. After the original echo signal is imaged in the range and estimated in the azimuth, the imaging result of the target area can be obtained, and then the imaging result is processed by differential interferometry technology to obtain high-precision deformation information of the target area.

[0070] Assuming that the electromagnetic wave emitted by the millimeter wave radar is a frequency modulated continuous wave, the signal expression is:

[0071]

[0072] In the formula, Indicates the transmission signal. is the carrier frequency of the transmitted signal, To adjust the frequency, Fast time for signal;

[0073] In order to further improve the angular resolution of the radar, the present invention mixes the echo signal with the transmission signal to obtain an intermediate frequency signal, which is expressed as:

[0074]

[0075] In the formula, represents the intermediate frequency signal, is the time delay for the transmitted signal to return from the target, is the distance from the radar to the target, The speed of light.

[0076] After mixing to obtain the intermediate frequency signal, the intermediate frequency signal is subjected to windowing processing:

[0077]

[0078] In the formula, represents the windowed IF signal, is the length of the window function, Time index within the window function.

[0079] Then the windowed IF signal A fast FFT transform is performed to extract the target distance in the observation area, and the beamforming method is used to estimate the azimuth. The distances and azimuths of all target points in the observation area constitute a range-azimuth diagram of the observation area. One frame of echo signal corresponds to one frame of range-azimuth diagram.

[0080] Step 3, interfering and unwrapping the range-azimuth diagram of each two adjacent frames obtained in step 2 to obtain the phase difference between the two adjacent frames, and then calculating the deformation amount of the observation area between the two adjacent frames based on the phase difference; and then extracting the deformation position based on the multivariate signal classification method.

[0081] Step 3.1: perform complex conjugate multiplication of the range-azimuth diagrams of each two adjacent frames to obtain the interference phase :

[0082]

[0083] In the formula, represents the phase operation, and Represents the range-azimuth diagram of two adjacent frames;

[0084] Step 3.2: Unwrap the interference phase to get the real phase difference ;

[0085] Step 3.3: Calculate the deformation between two adjacent frames based on the actual phase difference :

[0086]

[0087] In the formula, is the wavelength of millimeter radar waves.

[0088] However, the resolution of the azimuth obtained by the beamforming method is low. On the basis of obtaining the deformation amount at each position, the multivariate signal classification method is used to accurately estimate the azimuth, that is, the azimuth of each deformation position with higher accuracy. The specific process includes the following

[0089] Perform a fast Fourier transform on the windowed intermediate frequency signal obtained in step 2, and then calculate the covariance matrix:

[0090]

[0091] In the formula, is the number of snapshots, It is the signal obtained by fast Fourier transform of the windowed intermediate frequency signal. is the covariance matrix, represents the transpose operation;

[0092] Covariance matrix Perform eigenvalue decomposition to obtain the signal subspace and the noise subspace :

[0093]

[0094] In the formula, is the matrix composed of the first k eigenvalues ​​with the largest values ​​obtained from eigenvalue decomposition, and the matrix composed of the remaining eigenvalues ​​is recorded as ; The number of eigenvalues ​​k in is determined by the number of azimuths estimated by the BF method;

[0095] Constructing spatial spectrum function using noise subspace :

[0096]

[0097] In the formula, is the steering vector, determined by the millimeter-wave radar;

[0098] By traversing all angles The angle with the largest peak value of the spatial spectrum function is found, which is the azimuth of each deformation position of the target in the observation area. The resolution of the azimuth obtained in step 2 is higher.

[0099] Step 4: If the synthetic aperture radar imaging method is selected, then: the millimeter wave radar is mounted on a ground-based track, and the corresponding range-azimuth angle diagram is extracted by using a back-projection algorithm for the echo signal within a time period.

[0100] Among them, the distance and azimuth are extracted by using the back-projection algorithm for the echo signal within a time period, specifically:

[0101] Step 4.1, perform windowing processing on the echo signal within a time period:

[0102]

[0103] In the formula, is the echo signal of the transmitted signal of the mth transmitting antenna reflected by the target and received by the nth receiving antenna, is the windowed signal; For time; is the window function;

[0104] Step 4.2, for the windowed signal To perform distance pulse pressure:

[0105]

[0106] In the formula, is a matched filter, To adjust the frequency, is the range matched filter signal;

[0107] Step 4.3: Match the range-direction filtering signal Perform back-projection imaging processing to obtain the position coordinates of each target point in the observation area:

[0108]

[0109]

[0110] In the formula, Target point The signal strength, Target point The transmitted signal of the mth transmitting antenna passes through the target point The propagation time of the signal reflected and received by the nth receiving antenna, is the position coordinate of the mth transmitting antenna, is the position coordinate of the nth receiving antenna, is the location coordinate of the target point; M is the number of transmitting antennas, and N is the number of receiving antennas;

[0111] Step 4.4, extract the distance and azimuth according to the position coordinates of each target point in the observation area to obtain a distance-azimuth diagram.

[0112] SAR imaging algorithms include back projection algorithm (BP), range migration algorithm (RMA) and other algorithms. Among them, the BP algorithm is a time domain imaging algorithm. Compared with the RMA algorithm, it has lower requirements on radar tracks and is suitable for a variety of tracks such as straight lines and circular tracks. It has the characteristics of high accuracy and good robustness. However, the amount of calculation is large and it is difficult to meet the requirements of real-time imaging. Therefore, in a better embodiment, GPU acceleration can be used to achieve three-dimensional BP imaging.

[0113] Specifically, firstly, the CPU is used to complete step 4.1 and step 4.2 and obtain a plurality of continuous distance-directed matched filtering signals. Divide into multiple small tasks, use CUDA to assign the multiple small tasks to multiple GPUs; then each GPU completes its own small task: extract the position coordinates of each target point in the observation area from each divided signal according to step 4.3; then use the CPU to merge the position coordinates of all target points obtained by all small tasks to obtain the three-dimensional SAR imaging of the observation area; finally, use the CPU to filter the three-dimensional SAR imaging and extract the range-azimuth diagram. This embodiment is based on the joint acceleration of BP three-dimensional imaging solution by multiple GPUs, and multiple GPUs are processed in parallel, thereby significantly accelerating BP imaging.

[0114] Step 5, interfering and unwrapping the range-azimuth diagrams of the two time periods to obtain the phase difference between the two time periods, and then calculating the deformation amount of the observation area between the two time periods based on the phase difference.

[0115] Step 5.1: Perform complex conjugate multiplication of the range-azimuth diagrams in two different time periods to obtain the interference phase :

[0116]

[0117] In the formula, represents the phase operation, and A range-azimuth diagram showing two different time periods;

[0118] Step 5.2: Unwrap the interference phase to get the real phase difference ;

[0119] Step 5.3: Calculate the deformation between two different time periods based on the true phase difference :

[0120]

[0121] In the formula, is the wavelength of millimeter radar waves.

[0122] The above embodiments are preferred embodiments of the present application. Ordinary technicians in this field can also make various changes or improvements on this basis. Without departing from the overall concept of the present application, these changes or improvements should fall within the scope of protection required by the present application.

Claims

1. A multi-mode deformation monitoring method based on millimeter wave radar, characterized in that: include: Step 1: Select the radar imaging method according to your needs: If the real-time requirement for deformation monitoring is higher than the preset condition, the real aperture radar imaging method is selected; If the accuracy of deformation monitoring is higher than the preset threshold, the synthetic aperture radar imaging method is selected; Step 2: If the real aperture radar imaging mode is selected, then: a millimeter wave radar with a multi-input multi-output system is used, and its echo signal is mixed with the transmission signal to obtain an intermediate frequency signal, and the range-azimuth angle diagram of multiple consecutive frames of the observation area is obtained by processing the intermediate frequency signal; Step 3, interfering and unwrapping the range-azimuth diagrams of each two adjacent frames obtained in step 2 to obtain the phase difference between the two adjacent frames, and then calculating the deformation amount of the observation area between the two adjacent frames based on the phase difference; Then, the deformation position is extracted based on the multivariate signal classification method; Step 4: If the synthetic aperture radar imaging method is selected, then: the millimeter wave radar is mounted on a ground-based track, and a range-azimuth angle diagram is extracted by using a back-projection algorithm for the echo signal within a time period; In step 4, the distance and azimuth are extracted by applying a back-projection algorithm to the echo signal, specifically: Step 4.1, perform windowing processing on the echo signal within a time period: ； In the formula, is the echo signal of the transmitted signal of the mth transmitting antenna reflected by the target and received by the nth receiving antenna, is the windowed signal; For time; is the window function; Step 4.2, for the windowed signal To perform distance pulse pressure: ； In the formula, is a matched filter, To adjust the frequency, is the range matched filter signal; Step 4.3: Match the range-direction filtering signal Perform back-projection imaging processing to obtain the position coordinates of each target point in the observation area: ； ； In the formula, Target point The signal strength, Target point The transmitted signal of the mth transmitting antenna passes through the target point The propagation time of the signal reflected and received by the nth receiving antenna, is the position coordinate of the mth transmitting antenna, is the position coordinate of the nth receiving antenna, is the location coordinate of the target point; M is the number of transmitting antennas, and N is the number of receiving antennas; c is the speed of light; Step 4.4, extract the distance and azimuth according to the position coordinates of each target point in the observation area to obtain a distance-azimuth diagram; Step 5, interfering and unwrapping the range-azimuth diagrams of the two time periods to obtain the phase difference between the two time periods, and then calculating the deformation amount of the observation area between the two time periods based on the phase difference.

2. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: The echo signal is mixed with the transmission signal to obtain the intermediate frequency signal, which is expressed as: ； ； In the formula, represents the intermediate frequency signal, Indicates the transmission signal. is the carrier frequency of the transmitted signal, To adjust the frequency, To signal fast time; is the time delay for the transmitted signal to return from the target, is the distance from the radar to the target, The speed of light.

3. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: After the intermediate frequency signal is obtained by mixing, the intermediate frequency signal is first windowed, and then the windowed intermediate frequency signal is fast FFT transformed to obtain the range-azimuth angle of multiple consecutive frames in the observation area; the windowing process is expressed as: ； In the formula, represents the windowed IF signal, is the length of the window function, Time index within the window function.

4. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: The intermediate frequency signal is subjected to fast Fourier transform to extract the target distance in the observation area, and the azimuth is estimated using the beamforming method.

5. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: The method based on multivariate signal classification extracts the deformation position, that is, extracts the azimuth angle of each deformation position with higher precision, specifically: Perform a fast Fourier transform on the intermediate frequency signal and then calculate the covariance matrix: ； In the formula, is the number of snapshots, It is the signal obtained by fast Fourier transform of intermediate frequency signal. is the covariance matrix, represents the transpose operation; Covariance matrix Perform eigenvalue decomposition to obtain the signal subspace and the noise subspace : ； In the formula, is the matrix composed of the first k eigenvalues ​​with the largest values ​​obtained from the eigenvalue decomposition, and the matrix composed of the remaining eigenvalues ​​is recorded as ; The number of eigenvalues ​​k in is determined according to the number of azimuth angles estimated using the beamforming method; Constructing spatial spectrum function using noise subspace : ； In the formula, is the steering vector, determined by the millimeter-wave radar; By traversing all angles The angle with the largest peak value of the spatial spectrum function is found, which is the azimuth of the target with higher precision.

6. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: In step 4, the CPU is used to complete steps 4.1 and 4.2 and obtain a plurality of continuous distance-directed matched filter signals. Divide into multiple small tasks, use CUDA to assign multiple small tasks to multiple GPUs; then each GPU completes its own small task: extract the position coordinates of each target point in the observation area from each divided signal according to step 4.3; then use the CPU to merge the position coordinates of all target points obtained from all small tasks to obtain the three-dimensional SAR imaging of the observation area; finally, use the CPU to filter the three-dimensional SAR imaging and extract the range-azimuth diagram.

7. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: Step 3 specifically includes: Step 3.1: perform complex conjugate multiplication of the range-azimuth diagrams of each two adjacent frames to obtain the interference phase : ； In the formula, represents the phase operation, and Represents the range-azimuth diagram of two adjacent frames; Step 3.2: Unwrap the interference phase to get the real phase difference ; Step 3.3: Calculate the deformation between two adjacent frames based on the actual phase difference : ； In the formula, is the wavelength of millimeter radar waves.

8. The multi-mode deformation monitoring method based on millimeter wave radar according to claim 1 is characterized in that: Step 5 specifically includes: Step 5.1: Perform complex conjugate multiplication of the range-azimuth diagrams in two different time periods to obtain the interference phase : ； In the formula, represents the phase operation, and A range-azimuth diagram showing two different time periods; Step 5.2: Unwrap the interference phase to get the real phase difference ; Step 5.3: Calculate the deformation between two different time periods based on the true phase difference : ； In the formula, is the wavelength of millimeter radar waves.

Images