Non-uniform sampling rotation MIMO super-resolution imaging method, system and product

Abstract

The invention provides a non-uniform sampling rotation MIMO super-resolution imaging method, system and product, and the method comprises the steps: synthesizing a virtual aperture based on collected radar data, and building a coordinate system based on the virtual aperture; calculating position information of the radar antennas in the coordinate system, and screening all the radar antennas based on a preset step length to obtain effective virtual antennas; acquiring target distance information based on the acquired radar echo signal; constructing a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antenna, and generating a local super-resolution image based on the rotating super-resolution dictionary; and performing coordinate transformation and data splicing on each local super-resolution image to generate a global super-resolution image. Compared with a traditional angle estimation algorithm, the method provided by the invention can greatly improve the angle resolution capability, enables the main lobe of the radar to face the target at any time through the rotatable characteristic of the platform, enables the monitoring area of the radar to be enlarged, improves the target detection capability, and meanwhile, the radar has the advantage of flexible deployment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar, and in particular to a non-uniform sampling rotating MIMO super-resolution imaging method, system and product. BACKGROUND

[0002] With the comprehensive arrival of the intelligent era, high-resolution imaging technology has become a key support for realizing accurate perception and decision-making in the fields of industrial detection, security monitoring, intelligent transportation, etc. In the field of intelligent transportation, accurate imaging helps to realize real-time monitoring and trajectory prediction of vehicles and pedestrians, and improves road safety and traffic efficiency; in the industrial detection scene, high-resolution imaging can find tiny defects and ensure product quality and production safety. With the complication and diversification of application scenarios, the requirements for imaging resolution and accuracy are increasingly stringent, and traditional imaging technology has been difficult to meet the needs.

[0003] In the field of imaging detection, the choice of sensor is crucial. Although visual sensors can intuitively obtain target images, they are greatly affected by environmental factors such as light and weather, and the imaging quality decreases significantly in harsh conditions such as darkness and heavy fog; although ultrasonic sensors can achieve non-contact measurement, they have low resolution and insufficient detection capability for small targets. In comparison, radar sensors, with their advantages of being unaffected by light, smoke and other environmental disturbances, can work around the clock and protect privacy, and have become an important means of imaging detection. Especially millimeter wave radar, with its high frequency band and short wavelength, has higher resolution and accuracy in target detection and imaging, and is widely used in military, transportation, security and other fields. However, existing millimeter wave radars still face many challenges in angle imaging. On the one hand, traditional beamforming angle solving algorithms rely on fixed antenna arrays, which are limited by static installation mode and small-scale arrays, making it difficult to obtain rich target spatial information, resulting in limited angular resolution and failing to meet the requirements of high-precision imaging for detailed resolution. On the other hand, in complex electromagnetic environments or scenarios with obvious multipath effects, the processing capability of traditional algorithms for target echo signals is insufficient, and is easily affected by interference, resulting in target missed detection, imaging blur and other problems, making it difficult to achieve stable and reliable high-resolution imaging. SUMMARY

[0004] The present application aims to provide a non-uniform sampling rotating MIMO super-resolution imaging method, system and product, which can at least solve the problem of low angular resolution of radar in related technologies, resulting in poor positioning accuracy.

[0005] To solve the above technical problems, the first aspect of the embodiments of the present application provides a non-uniform sampling rotating MIMO super-resolution imaging method, comprising:

[0006] Synthesizing a virtual aperture based on the collected radar data, and establishing a coordinate system based on the virtual aperture;

[0007] The position information of the radar antennas in the coordinate system is calculated, and all the radar antennas are screened based on a preset step length to obtain effective virtual antennas;

[0008] Target distance information is obtained based on the collected radar echo signals;

[0009] A rotation super-resolution dictionary is constructed based on the target distance information and the position information of the effective virtual antennas, and a local super-resolution image is generated based on the rotation super-resolution dictionary; wherein the rotation super-resolution dictionary contains target angle information;

[0010] The local super-resolution images are subjected to coordinate conversion and data splicing to generate a global super-resolution image.

[0011] The second aspect of the embodiments of the present application provides a non-uniform sampling rotation MIMO super-resolution imaging system, comprising:

[0012] An aperture synthesis module is configured to synthesize virtual apertures based on the collected radar data, and to establish a coordinate system based on the virtual apertures;

[0013] An antenna screening module is configured to calculate the position information of the radar antennas in the coordinate system, and to screen all the radar antennas based on a preset step length to obtain effective virtual antennas;

[0014] A target acquisition module is configured to obtain target distance information based on the collected radar echo signals;

[0015] A first generation module is configured to construct a rotation super-resolution dictionary based on the target distance information and the position information of the effective virtual antennas, and to generate a local super-resolution image based on the rotation super-resolution dictionary; wherein the rotation super-resolution dictionary contains target angle information;

[0016] A second generation module is configured to perform coordinate conversion and data splicing on the local super-resolution images to generate a global super-resolution image.

[0017] The third aspect of the present application provides a radar, comprising a memory and a processor, wherein the processor is configured to execute a computer program stored in the memory, and the processor, when executing the computer program, implements each step of the non-uniform sampling rotation MIMO super-resolution imaging method described in the first aspect of the embodiments of the present application.

[0018] The fourth aspect of the present application provides a computer readable storage medium having a computer program stored thereon, and the computer program, when executed by a processor, implements each step of the non-uniform sampling rotation MIMO super-resolution imaging method described in the first aspect of the embodiments of the present application.

[0019] From the above, the embodiment of the application first synthesizes a virtual aperture based on the collected radar data, establishes a coordinate system based on the virtual aperture, then calculates the position information of the radar antenna in the coordinate system, and filters all the radar antennas based on a preset step to obtain effective virtual antennas. Subsequently, target distance information is obtained based on the collected radar echo signals, and a rotating super-resolution dictionary is constructed based on the target distance information and the position information of the effective virtual antennas, and a local super-resolution image is generated based on the rotating super-resolution dictionary. Finally, the global super-resolution image is generated by performing coordinate conversion and data splicing on each local super-resolution image. The non-uniform sampling rotating MIMO super-resolution imaging method proposed in the application can combine a rotating platform to construct a time-space coordinated virtual large aperture through radar rotation, which can greatly improve the angle resolution compared with the traditional angle estimation algorithm. In addition, the rotatable feature of the platform can make the radar main lobe always point to the target, so that the radar monitoring area is larger and the target detection capability is improved, which is conducive to resource optimization allocation, and the radar has the advantage of flexible deployment.

[0020] It should be understood that the content described in this part is not intended to identify key or important features of the application, nor is it used to limit the scope of the application. Other features of the application will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the related art or the embodiments of the application, the following will briefly introduce the drawings needed to be used in the description of the related art or the embodiments of the application. Obviously, the drawings in the following description are only some embodiments of the application, and not all embodiments. Those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0022] Figure 1 The flowchart of the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiment of the application is shown in the figure.

[0023] Figure 2 The two-dimensional turntable radar model diagram in the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiment of the application is shown in the figure.

[0024] Figure 3 The local super-resolution image turntable radar model diagram in the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiment of the application is shown in the figure.

[0025] Figure 4 The detailed flowchart of the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiment of the application is shown in the figure.

[0026] Figure 5A program module schematic diagram of the non-uniform sampling rotating MIMO super-resolution imaging system provided by the embodiment of the present application is provided.

[0027] Figure 6 A module block diagram of the radar provided by the embodiment of the present application is provided.

[0028] Figure 7 A module block diagram of the computer readable storage medium provided by the embodiment of the present application is provided. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the present application more obvious and easy to understand, the present application will be described clearly and completely below in combination with the embodiments of the present application and the drawings, wherein the same or similar reference signs represent the same or similar elements or elements with the same or similar functions throughout. It should be understood that the embodiments of the present application described below are only used to explain the present application and do not limit the present application, that is, all other embodiments obtained by those skilled in the art without creative labor based on the various embodiments of the present application belong to the scope of protection of the present application. In addition, the technical features involved in the various embodiments of the present application described below can be combined with each other as long as there is no conflict.

[0030] With the comprehensive arrival of the intelligent era, high-resolution imaging technology has become a key support for realizing accurate perception and decision-making in the fields of industrial detection, security monitoring, intelligent transportation, etc. In the field of intelligent transportation, accurate imaging helps to realize real-time monitoring and trajectory prediction of vehicles and pedestrians, and improves road safety and traffic efficiency; in the industrial detection scene, high-resolution imaging can find tiny defects and ensure product quality and production safety. With the complication and diversification of application scenarios, the requirements for imaging resolution and accuracy are increasingly stringent, and traditional imaging technology has been difficult to meet the needs.

[0031] In the field of imaging detection, the selection of sensors is crucial. Although visual sensors can intuitively obtain target images, they are greatly affected by environmental factors such as light and weather, and their imaging quality decreases significantly in harsh conditions such as darkness and heavy fog. Ultrasonic sensors can achieve non-contact measurement, but they have low resolution and insufficient detection capability for small targets. In contrast, radar sensors, which are not affected by environmental disturbances such as light and smoke, can work around the clock and provide privacy protection, making them an important means of imaging detection. In particular, millimeter wave radars have higher resolution and accuracy in target detection and imaging due to their high frequency band and short wavelength, and are widely used in military, transportation, security, and other fields. However, existing millimeter wave radars still face many challenges in angle resolution imaging. On the one hand, traditional beamforming algorithms for angle resolution rely on fixed antenna arrays, which are limited by static installation patterns and small-scale arrays, making it difficult to obtain rich target spatial information, resulting in limited angular resolution and failing to meet the requirements of high-precision imaging for detailed resolution. On the other hand, in complex electromagnetic environments or scenarios with significant multipath effects, the processing capability of traditional algorithms for target echo signals is insufficient, and they are easily affected by interference, resulting in target miss detection, imaging blur, and other problems, making it difficult to achieve stable and reliable high-resolution imaging.

[0032] To break through the above technical bottlenecks, the present application proposes a non-uniform sampling rotating MIMO super-resolution imaging method. By rotating the radar, a virtual large aperture with spatio-temporal coordination can be constructed to improve the angular resolution of the target. This method does not infringe on personal privacy and is not sensitive to light, dust, smoke, and temperature, and can work around the clock and in all weather conditions. In addition, this algorithm built on a turntable radar also has the advantage of dynamic installation.

[0033] The present application provides a non-uniform sampling rotating MIMO super-resolution imaging method, specifically, please refer to Figure 1 , Figure 1 The flowchart of the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiments of the present application, the non-uniform sampling rotating MIMO super-resolution imaging method comprises the following steps 101 to 105.

[0034] Step 101, based on the collected radar data, a virtual aperture is synthesized, and a coordinate system is established based on the virtual aperture.

[0035] When the radar starts working, the turntable rotates at a constant speed. After the radar records N frames of data, it starts to generate a local super-resolution image. The two-dimensional super-resolution image generated is based on the coordinate system established at the center time of the synthesized virtual aperture. Specifically, the center position of the synthesized virtual aperture is set as the origin of the coordinate system, the horizontal direction of the radar at that time is set as the positive direction of the x-axis (right is positive), and the vertical direction is set as the positive direction of the y-axis (up is positive). The first radar image is recovered from the radar echo data from the 1st frame to the Nth frame, and the center time of the virtual aperture corresponds to the Nth frame. The first chirp of the frame begins, the second radar image is recovered from data from frame 2 to frame N+1, and so on. The radar begins generating the corresponding local super-resolution image frame by frame upon acquiring the Nth frame of data. After scanning each image, the 360° range of the disk is completed. Then, the data is fused based on the rotation angle of each image relative to the indoor stationary coordinate system, and finally a global super-resolution image is obtained.

[0036] In this embodiment of the application, a virtual aperture is synthesized based on the acquired radar data, and a coordinate system is established based on the virtual aperture. This includes: controlling a radar turntable with a disk radius of r to rotate uniformly at an angular velocity ω, and controlling the radar to rotate at frame intervals T. f Transmit frame data; where each frame contains N c The time interval is T. c The chirp signal; N frames of first echo data are acquired by radar, and a virtual aperture is synthesized based on the N frames of first echo data; the first... The start time of the first chirp signal in the frame data is taken as the virtual aperture center; a coordinate system is established with the virtual aperture center as the origin, the horizontal orientation of the radar at that time as the positive x-axis, and the vertical orientation of the radar at that time as the positive y-axis.

[0037] Specifically, the two-dimensional turntable radar model diagram in this application embodiment is as follows: Figure 2 As shown, taking the radar used as an example, the radar has 4 azimuth antennas and the antenna spacing is λ / 2. Let the radar frame interval be T. f The number of chirps per frame is N c The time interval between two chirps is T. c The turntable disk has a radius of r and a rotational angular velocity of ω. The virtual large aperture synthesis time is the duration of N frames of radar data. When the radar starts working, the turntable rotates at a constant speed. After acquiring N frames of data, the radar begins to generate local super-resolution images. The generated two-dimensional super-resolution images are all based on the coordinate system established at the center time of the synthesized virtual aperture. Specifically, the center position of the synthesized virtual aperture is set as the origin of the coordinate system, with the horizontal direction of the radar at that moment as the positive x-axis (positive to the right) and the vertical direction as the positive y-axis (positive upward). The first radar image is recovered from the radar echo data from the 1st frame to the Nth frame. The virtual aperture center time corresponds to the [missing information - likely a specific time frame or time value]. The first chirp of the frame begins, the second radar image is recovered from data from frame 2 to frame N+1, and so on. The radar begins generating the corresponding local super-resolution image frame by frame upon acquiring the Nth frame of data. After the image is obtained, the scanning of the 360° range of the disc is completed, and then data fusion is performed according to the angle of rotation of each image relative to the indoor stationary coordinate system, so as to finally obtain a global super-resolution image.

[0038] It can be understood that the embodiments of the present application select a 3-transmit 4-receive MIMO radar module for experiment, but are not limited to the number of multi-transmit multi-receive radar radio frequency modules, and the present application does not make specific limitations thereto.

[0039] In step 102, the position information of the radar antenna in the coordinate system is calculated, and all radar antennas are screened based on a preset step length to obtain effective virtual antennas.

[0040] After the new rectangular coordinate system is established with the synthetic virtual aperture center as the origin, the virtual aperture center represents the direction of the radar at all times, and the radar only detects targets in the field of view of the radar in this direction. Therefore, the specific positions of all antennas in the virtual aperture in the newly established coordinate system need to be calculated first. Further, since there are multiple radar antennas in the virtual aperture, the positions of the antennas may be too close at different slow times, resulting in that the signals received by the antennas are very similar, the phase difference is very small or almost 0, and the array cannot extract the direction information of the target through the phase difference. Therefore, the antennas need to be selected to obtain effective virtual antennas.

[0041] In the embodiments of the present application, the position information of the radar antenna in the coordinate system is calculated, including: calculating the angle θ p,l of the radar in the virtual aperture in the pth frame and the lth chirp relative to the initial position of the radar in the slow time dimension p,l ; wherein the angle θ c represents the angle of the radar in the virtual aperture in the pth frame and the lth chirp, θ0 represents the included angle between the initial direction of the radar and the current direction of the radar, the number of calculated angles is N*N p,l , and the expression of the angle θ p,l is:

[0042]

[0043] Based on the angle θ p,l , the angle difference Δθ p,l of each radar position relative to the virtual aperture center is calculated p,l ; wherein the expression of the angle difference Δθ p,l is:

[0044]

[0045] Based on the angle difference Δθ p,l , the position P p,l of the radar in the coordinate system in the slow time dimension is calculated p,l =(X p,lY p,l represents the position of the radar in the pth frame and the lth chirp, the number of positions is N*N c , the expression of position P p,l is:

[0046]

[0047] Based on the radar antenna spacing and the disc radius r, the position P p,l,k of the radar antenna in the coordinate system is calculated p,l,k =(X p,l,k ,Y p,l,k ); wherein λ represents the wavelength of the radar transmitted signal, the position P p,l,k =(X p,l,k ,Y p,l,k ) represents the position of the radar in the pth frame, the lth chirp and the kth antenna, the expression of position P p,l is:

[0048]

[0049] Specifically, Figure 3 is the local super-resolution image generation model of the turntable radar, at this time a new rectangular coordinate system is established with the center of the synthesized virtual aperture as the origin. The virtual aperture center represents the direction of the radar at the moment, we only detect the targets in the field of view of the radar in this direction, therefore we first need to calculate the specific position of all antennas in the virtual aperture in the newly established coordinate system. When calculating the position of the antenna, we need to calculate the angle of the radar in each chirp, i.e. the slow time dimension, relative to the initial position of the radar, here we use θ p,l to represent the angle of the radar in the pth frame and the lth chirp in the virtual aperture, θ0 represents the included angle between the initial direction of the radar and the current direction of the radar, the number of calculated angles is N*N c , the specific calculation relationship is as follows:

[0050]

[0051] Subsequently, the angle difference Δθ p,l of each radar position in the slow time dimension relative to the center of the virtual aperture is calculated, which represents the angle difference of the radar in the pth frame and the lth chirp relative to the radar at the center of the virtual aperture, the expression is as follows:

[0052]

[0053] Based on the angle difference Δθ p,l , the position P p,l of the radar in the coordinate system in the slow time dimension is calculated p,l =(X p,l); wherein, λ represents the wavelength of radar transmitting signal, position P p,l = (X p,l , Y p,l ) represents the position of radar at the pth frame and the lth chirp, the number of positions is N*N c , and the expression of position P p,l is as follows:

[0054]

[0055] Since the single-position radar has four antennas, and the radar is perpendicular to the center of the disc, the antenna numbers are marked from right to left as 1, 2, 3, and 4. The position of the No. 1 antenna coincides with the radar position at this time, and the positions of the No. 2, No. 3, and No. 4 antennas are also calculated in combination with the antenna spacing and the disc radius. The position of the radar at the pth frame, the lth chirp, and the kth antenna is represented by P p,l,k = (X p,l,k , Y p,l,k ), and the calculation expression of the antenna position is as follows:

[0056]

[0057] In the embodiments of the present application, the effective virtual antennas are obtained by screening all radar antennas based on a preset step length, including: starting from the first antenna as a starting point, and traversing and screening the radar antennas with a step length of The antenna index array of the screened antennas is Idx = [Idx1, Idx2, …, IdxG]; the screened antennas are marked according to the slow time dimension, and the effective virtual antennas are obtained based on the antenna index array; wherein, the effective virtual antennas are marked as Rx1, Rx2, Rx3, …, RxG, and the corresponding positions are represented as P Rx1 = (X Rx1 , Y Rx1 ), P Rx2 = (X Rx2 , Y Rx2 ), …, P RxG = (X RxG , Y RxG ).

[0058] Specifically, through the above calculation, the specific positions of N*N c *4 antennas in the virtual aperture can be obtained, and these antennas are arranged in a column according to the slow time dimension and marked as 1, 2, 3, …, (N*N c *4)-1, N*N c *4 antennas, and the corresponding antenna positions are P 1,1 , P 1,2 , P 1,3 , …, Since there are multiple radar antennas in the virtual aperture, the antenna positions will be too close at different slow times, resulting in very similar signals received by the antennas and very small or almost zero phase difference, so the array cannot extract the direction information of the target through the phase difference, and thus the antennas need to be selected to prevent the distance between antennas from exceeding and to increase the angle estimation ambiguity and the observation of the target by the antennas, we select the antennas with a step size of The specific process is as follows: the antennas start from antenna 1, find an antenna with a distance of from antenna 1, record the antenna number and position, then start from the antenna, find an antenna with a distance of from the antenna, and so on, until all N*N c *4 antennas are traversed. Suppose the selected antennas are G, the corresponding antenna numbers are Idx=[Idx1,Idx2,…,IdxG], mark the G antennas in the slow time dimension as Rx1,Rx2,Rx3,…,RxG, and find the specific positions of the antennas according to Idx, and suppose the corresponding positions are P Rx1 =(X Rx1 ,Y Rx1 ), P Rx2 =(X Rx2 ,Y Rx2 ),…,P RxG =(X RxG ,Y RxG ), the selected antennas will be used for angle estimation in the subsequent steps.

[0059] Step 103, obtaining target distance information based on the collected radar echo signal.

[0060] In the embodiments of the present application, obtaining target distance information based on the collected radar echo signal includes: pre-processing the collected radar echo signal to obtain a first discrete echo signal; wherein the first discrete echo signal is represented as Y[n,m,k], n is the fast time dimension, representing the nth sampling point, m is the slow time dimension, representing the mth linear frequency modulation continuous wave signal, and k is the antenna dimension, representing the kth channel received signal; windowing the first discrete echo signal in the fast time dimension to obtain a second discrete echo signal; performing fast Fourier transform on the second discrete echo signal to obtain a third discrete echo signal; wherein the third discrete echo signal is a range-slow time-antenna dimension signal, represented as d∈[1,N d ], N d is the number of range sampling points; performing target range detection on the third discrete echo signal through a constant false alarm rate algorithm to obtain target distance information.

[0061] Specifically, the radar sends a linear frequency modulation continuous wave signal through the transmitter during data collection, the signal propagates in space, is emitted by an object and is received by the radar receiver, then after signal amplification, mixing and ADC sampling, a discrete echo signal is obtained, and a single frame of radar signal can be represented as Y[n,m,k], where n is the fast time dimension, representing the nth sampling point, m is the slow time dimension, representing the mth linear frequency modulation continuous wave signal, and k is the antenna dimension, representing the received signal of the kth channel. To obtain the distance, the Y[n,m,k] signal needs to be windowed in the fast time dimension to reduce spectral leakage and make the spectral analysis more accurate, and then a fast Fourier transform is performed in this dimension to obtain a cubic data of distance-slow time dimension-antenna dimension where d∈[1,N d ], N d is the number of distance sampling points.

[0062] Since the local super-resolution image of the radar is established in the coordinate system at the moment when the virtual aperture center, we focus on the center frame, that is The target of the frame needs to be first detected in the fast time dimension of the radar echo data of the first frame to obtain the target distance image, then non-coherent accumulation is performed in the antenna dimension to enhance the signal-to-noise ratio of the target, and then the constant false alarm rate (CFAR) algorithm is used to detect the target distance image to obtain the potential target distance unit and record it.

[0063] Step 104, constructing a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antenna, and generating a local super-resolution image based on the rotating super-resolution dictionary; wherein the rotating super-resolution dictionary contains target angle information.

[0064] Based on the effective virtual antenna spatial distribution and target distance information obtained by screening, a rotating super-resolution dictionary containing target angle information is constructed. The dictionary establishes a theoretical response model of the target signal reaching each antenna under different angles by presetting the angle range where the target may exist, forms an angle-signal mapping relationship covering the preset viewing angle, and the construction of the dictionary integrates the spatial position of the antenna array and the target distance parameter. Then, for the target distance unit detected in the radar echo signal, the signal features corresponding to the distance are extracted and matched with the theoretical signal templates in the rotating super-resolution dictionary for matching analysis, and by evaluating the similarity between the echo signal and each angle template, an energy spectrum diagram of the target angle distribution is generated. The peak position of the energy spectrum corresponds to the actual azimuth angle of the target, and its resolution breaks through the physical aperture limitation of the traditional array through the super-resolution design of the dictionary. Based on the above processing, the angle distribution of each distance unit is integrated to finally generate a local super-resolution image.

[0065] In this embodiment, a rotating super-resolution dictionary is constructed based on target distance information and the position information of the effective virtual antenna, including: calculating the phase difference between the target and the effective virtual antenna and between the target and the center of the virtual aperture; wherein, the phase difference matrix composed of the phase differences is represented as:

[0066]

[0067] Where azi represents the target azimuth angle, and R represents the distance from the target to the radar;

[0068] A rotating super-resolution dictionary is constructed based on the azimuth angle and the phase difference matrix; wherein, the rotating super-resolution dictionary is represented as:

[0069] Ψ=[Δφ(Azi(1)0,Δφ(Azi(2)0…,Δφ(Azi(N A )0],

[0070] The azimuth range of the detection is Azi, and the number of detected angles is N. A There are 1, and the dictionary size is G*N. A .

[0071] Specifically, the target distance is known at this point, the antenna position selected in the virtual aperture is known, and in the newly created coordinate system, the location of the virtual aperture center is known. Using the target azimuth as the origin, if the target's azimuth is known, we can calculate the target's exact location, and thus calculate the phase difference between the target and the selected G antennas and between the target and the central aperture. Let the target azimuth be azi, and R represent the distance from the target to the radar. Then the phase difference matrix can be expressed as:

[0072]

[0073] Therefore, we can combine the phase difference matrices corresponding to multiple azimuth angles to construct a rotating super-resolution dictionary. Let the range of detected azimuth angles be Azi, and the number of detected angles be N. A The constructed rotational super-resolution dictionary is .

[0074] Ψ=[Δφ(Azi(1)),Δφ(Azi(2))…,Δφ(Azi(N A ))]

[0075] The dictionary size is G*N A The dictionary records the phase difference between antennas when the target is in different positions. The phase difference contains angle information. Based on this, the angle can be recovered using the corresponding echo data of the antenna.

[0076] In the embodiment of the present application, the local super-resolution image is generated based on the rotating super-resolution dictionary, which comprises: obtaining second echo data under a corresponding distance unit from the radar echo signal based on target distance information; arranging the second echo data in the slow time dimension to obtain third echo data; extracting the third echo data based on an antenna index array to obtain fourth echo data; wherein the fourth echo data is represented as obtaining a target angle spectrum based on the fourth echo data; wherein the target angle spectrum is represented as: wherein abs is a modulus operation, P is a target power spectrum, and the length is N A , and θ is a target angle; and generating a local super-resolution image based on the target angle spectrum.

[0077] Specifically, in the entire virtual aperture, the data size captured by the radar is N r *N c *4*N, the distance of the target has been calculated in the foregoing, so the data size of the radar under a specific distance unit is N*N c *4, and similar to the calculation of the antenna position in the foregoing, N*N c *4 echo data of the radar under the specific distance unit is taken out and then arranged in the slow time dimension, so as to correspond to N*N c *4 antennas in the virtual aperture, and the arranged echo data is denoted as y, which has a size of (N*N c *4)×1, then the data is extracted by using the index array Idx of the selected antenna in the foregoing to obtain echo data for angle estimation .

[0078]

[0079] Combined with the constructed rotating super-resolution dictionary, the radar model can be represented as:

[0080]

[0081] wherein is a signal vector, and n is a noise signal vector. Then, the signals received by multiple antenna units in the virtual aperture are weighted and phase-aligned, i.e., coherent integration, to obtain a target angle spectrum as follows:

[0082]

[0083] wherein abs is a modulus operation, P is a target power spectrum, and the length is N ABased on this, the power values ​​at various angles at that distance can be obtained. Finally, each potential target range cell is traversed, and a corresponding rotating super-resolution dictionary is constructed according to the above method. The corresponding echo data is extracted, the target angle spectrum is calculated, and the power values ​​corresponding to different angles of all target range cells are displayed on the imaging plane to obtain a local radar super-resolution image in one direction.

[0084] Step 105: Perform coordinate transformation and data stitching on each local super-resolution image to generate a global super-resolution image.

[0085] Since each local image corresponds to an observation viewpoint at a different position on the turntable, the origin of its coordinate system and the angular reference are different. Therefore, it is necessary to perform coordinate transformation on each local image to obtain data in the same coordinate system, and then stitch them together to finally generate a global super-resolution image.

[0086] In this embodiment, coordinate transformation and data stitching are performed on each local super-resolution image to generate a global super-resolution image, including: calculating the angle difference between the corresponding coordinate system in each local super-resolution image and the stationary indoor coordinate system; wherein, the angle difference of the i-th local super-resolution image... Represented as:

[0087]

[0088] The radar generates images frame by frame, starting from the Nth frame, with the time difference between two images being the radar frame interval T. f The angle of rotation of the disk is T. f w; Coordinate transformation is performed on each local super-resolution image based on the angle difference; where the coordinate transformation of the i-th local super-resolution image is expressed as:

[0089]

[0090] Among them, (x i ′ ,y i ′ (x) represents the position of the target in the i-th image within the Cartesian coordinate system at the center of the virtual aperture. i ,y i ) represents the position of the target in the i-th image in a static indoor Cartesian coordinate system; the data of each local super-resolution image after coordinate transformation is stitched together to generate a global super-resolution image.

[0091] Specifically, we calculate and select the antenna positions when generating the first image, and the subsequent partial image generation still uses these selected antennas and the corresponding antenna positions, without the need to recalculate the positions of the antennas in the virtual aperture, because the imaging plane is relative to the aperture center, and the relative position of the antennas in the virtual aperture does not change due to the rotation of the disc, and the only change is the angle of the imaging plane. Therefore, from the second image, we only need to obtain the corresponding N frames of radar echo data, then use the consistent antenna index Idx to extract the echo data, and then perform super-resolution angle estimation in the respective imaging plane to obtain the corresponding image. After obtaining M images, the 360° scanning of the environment is completed. Let be the angle difference between the rectangular coordinate system established by the i-th image and the static indoor coordinate system, which is calculated as follows:

[0092]

[0093] In addition, as the target deviates from the front of the radar, the radar angular resolution gradually decreases. Based on this characteristic, we can select a small angle range in front of the radar, and combine the rotatable characteristics of the turntable, so that the main lobe of the radar is always pointing at the target, thereby enhancing the angular resolution of the target twice,

[0094] Since the radar generates images frame by frame from the N-th frame, the time difference between two images is T f , the angle of disc rotation is T f w, so when calculating the angle spectrum corresponding to a certain distance, we only perform angle spectrum calculation on the angle range T f w in front of the radar to obtain a small-angle super-resolution image. Since there is an angle deviation between the imaging plane of the partial super-resolution image and the static indoor rectangular coordinate system, coordinate conversion is needed, and the coordinate conversion of the i-th image is as follows:

[0095]

[0096] where (x i ′ ,y i ′ ) is the position of the target in the virtual aperture center rectangular coordinate system in the i-th image, and (x i ,y i ) is the position of the target in the static indoor rectangular coordinate system in the i-th image. Data stitching is performed on the multiple radar super-resolution images after coordinate conversion to obtain a global radar super-resolution image. At this time, the target has consistent angular super-resolution in all directions, and there is no deviation due to the deviation of the target from the front of the radar.

[0097] From the above, the embodiment of the application first synthesizes a virtual aperture based on the collected radar data, establishes a coordinate system based on the virtual aperture, then calculates the position information of the radar antenna in the coordinate system, and filters all radar antennas based on a preset step length to obtain effective virtual antennas, then obtains target distance information based on the collected radar echo signal, constructs a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antennas, and generates a local super-resolution image based on the rotating super-resolution dictionary, and finally, performs coordinate conversion and data splicing on each local super-resolution image to generate a global super-resolution image. In the non-uniform sampling rotating MIMO super-resolution imaging method proposed by the application, the radar with a turntable can be dynamically installed, does not infringe on personal privacy during detection, is not sensitive to light, dust, smoke and temperature, and can work all day and all weather. Compared with the traditional angle estimation algorithm, the method uses a rotating MIMO radar to construct a virtual large aperture, regional imaging and global fusion, effectively improves the target angle resolution, avoids the angle resolution attenuation caused by the view angle deviation, can realize full-area dead-angle-free high-precision detection, and meets the panoramic perception demand.

[0098] It should be understood that the size of the serial number of each step in the embodiment does not mean the order of execution of the steps, and the execution order of each step should be determined according to its function and inherent logic, and should not constitute the only limitation on the implementation process of the embodiment of the application.

[0099] In summary, the detailed process of the non-uniform sampling rotating MIMO super-resolution imaging method according to the embodiment of the application can be referred to Figure 4 , specifically as follows:

[0100] Step 401, synthesizing a virtual aperture based on the collected radar data, and establishing a coordinate system based on the virtual aperture;

[0101] Step 402, calculating the angle of the radar relative to the initial position of the radar in the slow time dimension, and calculating the angle difference of each radar position relative to the center of the virtual aperture, and calculating the position of the radar in the coordinate system in the slow time dimension based on the angle difference;

[0102] Step 403, calculating the position of the radar antenna in the coordinate system based on the antenna spacing and the radius of the disc of the radar;

[0103] Step 404, filtering all radar antennas based on a preset step length to obtain effective virtual antennas;

[0104] Step 405, obtaining target distance information based on the collected radar echo signal;

[0105] Step 406, calculating the phase difference between the target to the effective virtual antenna and the target to the center of the virtual aperture, and constructing a rotating super-resolution dictionary based on the azimuth angle and the phase difference matrix;

[0106] In step 407, second echo data under a corresponding distance unit is obtained from the radar echo signal based on the target distance information, the second echo data is arranged in a slow time dimension to obtain third echo data, and the third echo data is extracted based on an antenna index array to obtain fourth echo data.

[0107] In step 408, a target angle spectrum is obtained based on the fourth echo data, and a local super-resolution image is generated based on the target angle spectrum.

[0108] In step 409, an angle difference between a corresponding coordinate system in each local super-resolution image and a stationary indoor coordinate system is calculated, and each local super-resolution image is coordinate-converted based on the angle difference.

[0109] In step 410, each local super-resolution image after coordinate conversion is data-stitched to generate a global super-resolution image.

[0110] For more detailed processes of each of steps 401-410, refer to the related part of the foregoing description, which will not be repeated here.

[0111] Please refer to Figure 5 , Figure 5 A non-uniform sampling rotating MIMO super-resolution imaging system is provided in the embodiments of the present application. The system can be used to implement the non-uniform sampling rotating MIMO super-resolution imaging method involved in the embodiments of the present application. The non-uniform sampling rotating MIMO super-resolution imaging system 500 mainly includes:

[0112] An aperture synthesis module 501 is configured to synthesize a virtual aperture based on the collected radar data, and establish a coordinate system based on the virtual aperture.

[0113] An antenna screening module 502 is configured to calculate position information of radar antennas in the coordinate system, and screen all radar antennas based on a preset step length to obtain effective virtual antennas.

[0114] A target acquisition module 503 is configured to acquire target distance information based on the collected radar echo signal.

[0115] A first generation module 504 is configured to construct a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antennas, and generate a local super-resolution image based on the rotating super-resolution dictionary. The rotating super-resolution dictionary contains target angle information.

[0116] A second generation module 505 is configured to coordinate-convert and data-stitch each local super-resolution image to generate a global super-resolution image.

[0117] In some embodiments of the present embodiment, when the aperture synthesis module 501 performs the function of synthesizing a virtual aperture based on the collected radar data, and establishing a coordinate system based on the virtual aperture, it is configured to: control the radar turntable with a disc radius of r to rotate at an angular velocity of ω, and acquire N frames of first echo data through the radar at a frame interval T f , where each frame of data contains N c chirp signals with a time interval of T c ; synthesize a virtual aperture based on the N frames of first echo data; take the starting time of the first chirp signal in the first frame of data as the center of the virtual aperture; and establish a coordinate system with the center of the virtual aperture as the origin, the horizontal direction of the radar at that time as the positive direction of the x-axis, and the vertical direction of the radar at that time as the positive direction of the y-axis.

[0118] In some embodiments of the present embodiment, when the antenna screening module 502 performs the function of calculating the position information of the radar antenna in the coordinate system, it is configured to: calculate the angle θ p,l of the radar in the slow time dimension relative to the initial position of the radar; where the angle θ p,l represents the angle of the radar in the pth frame of the lth chirp in the virtual aperture, θ0 represents the included angle between the initial direction of the radar and the current direction of the radar, the number of calculated angles is N*N c , and the expression of the angle θ p,l is:

[0119]

[0120] Based on the angle θ p,l , the angle difference Δθ p,l of each radar position relative to the center of the virtual aperture is calculated; where the expression of the angle difference Δθ p,l is:

[0121]

[0122] Based on the angle difference Δθ p,l , the position P p,l =(X p,l , Y p,l ) of the radar in the coordinate system in the slow time dimension is calculated; where the position P p,l =(X p,l , Y p,l ) represents the position of the radar at the pth frame of the lth chirp, the number of positions is N*N c , and the expression of the position p p,l is:

[0123] ​

[0124] Antenna spacing based on radar and disc radius r to calculate the position P of the radar antenna in the coordinate system p,l,k =(X p,l,k ,Y p,l,k ); wherein λ represents the wavelength of the radar transmitted signal, the position P p,l,k =(X p,l,k ,Y p,l,k ) represents the position of the radar at the pth frame, the lth chirp and the kth antenna, and the expression of the position P p,l,k

[0125]

[0126] Further, in some embodiments of the present embodiment, the antenna screening module 502, when performing the function of screening all radar antennas based on a preset step to obtain effective virtual antennas, is configured to: starting from the first antenna as a starting point, traversing and screening the radar antennas with a step of ; wherein the antenna index array of the screened antennas is Idx=[Idx1, Idx2, …, IdxG]; the screened antennas are marked according to the slow time dimension, and the effective virtual antennas are obtained based on the antenna index array; wherein the effective virtual antennas are marked as Rx1, Rx2, Rx3, …, RxG, and the corresponding positions are represented as P Rx1 =(X Rx1 ,Y Rx1 ), P Rx2 =(X Rx2 ,Y Rx2 ), …, P RxG =(X RxG ,Y RxG ).

[0127] In some embodiments of the present embodiment, the target acquisition module 503, when performing the function of acquiring target distance information based on the collected radar echo signal, is configured to: pre-process the collected radar echo signal to obtain a first discrete echo signal; wherein the first discrete echo signal is represented as Y[n, m, k], n is the fast time dimension, representing the nth sampling point, m is the slow time dimension, representing the mth linear frequency modulation continuous wave signal, and k is the antenna dimension, representing the received signal of the kth channel; performing windowing processing on the first discrete echo signal in the fast time dimension to obtain a second discrete echo signal; performing fast Fourier transform on the second discrete echo signal to obtain a third discrete echo signal; wherein the third discrete echo signal is a distance-slow time dimension-antenna dimension signal, represented as d∈[1,N d ] and N d ​The number of distances from the sampling point; the target distance information is obtained by performing target distance detection on the third discrete echo signal through a constant false alarm rate algorithm.

[0128] In some embodiments of the present application, the first generation module 504, when performing the function of constructing a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antenna, is configured to: calculate the phase difference between the target and the effective virtual antenna and the target and the virtual aperture center; wherein the phase difference matrix composed of the phase difference is represented as:

[0129]

[0130] Wherein, azi is the target azimuth angle, and R represents the distance from the target to the radar.

[0131] The rotating super-resolution dictionary is constructed based on the azimuth angle and the phase difference matrix; wherein the rotating super-resolution dictionary is represented as:

[0132] Ψ = [ Δφ ( Azi ( 1 ) ), Δφ ( Azi ( 2 ) ), …, Δφ ( Azi ( N A ) ) ],

[0133] Wherein, the detected azimuth angle range is Azi, the number of detected angles is N A , and the dictionary size is G*N A .

[0134] In some embodiments of the present application, the first generation module 504, when performing the function of generating a local super-resolution image based on the rotating super-resolution dictionary, is configured to: obtain second echo data under the corresponding distance unit from the radar echo signal based on the target distance information; arrange the second echo data in the slow time dimension to obtain third echo data; extract the third echo data based on the antenna index array to obtain fourth echo data; wherein the fourth echo data is represented as Obtain the target angle spectrum based on the fourth echo data; wherein the target angle spectrum is represented as: Wherein, abs is a modulo operation, P is a target power spectrum, the length is N A , and θ is a target angle; generate a local super-resolution image based on the target angle spectrum.

[0135] In some embodiments of the present application, the second generation module 505, when performing the function of coordinate conversion and data splicing on each local super-resolution image to generate a global super-resolution image, is configured to: calculate the angle difference between the corresponding coordinate system in each local super-resolution image and the static indoor coordinate system; wherein the angle difference of the i-th local super-resolution image is represented as:

[0136] ​

[0137] Wherein, the radar generates images frame by frame from the Nth frame, the time difference between two images is the radar frame interval T f , the angle of the disc rotation is T f w;Based on the angle difference, the coordinate conversion of each local super-resolution image is carried out;Wherein, the coordinate conversion of the ith local super-resolution image is represented as:

[0138]

[0139] Wherein, (x i ′ ,y i ′ ) is the position of the target in the virtual aperture center rectangular coordinate system in the ith image, (x i ,y i ) is the position of the target in the static indoor rectangular coordinate system in the ith image;The data splicing of each local super-resolution image after coordinate conversion is carried out, and the global super-resolution image is generated.

[0140] In detail, each module in the non-uniform sampling rotating MIMO super-resolution imaging system 500 provided by the embodiment of the application adopts the same technical means as the non-uniform sampling rotating MIMO super-resolution imaging method in the above Figure 1 , and can produce the same technical effects, which will not be described here.

[0141] Please refer to Figure 6 , Figure 6 The module block diagram of the radar provided by the embodiment of the application.

[0142] As Figure 6 shown, the application also provides a radar, which can be used to implement the non-uniform sampling rotating MIMO super-resolution imaging method in the foregoing embodiments, comprising a memory 601 and at least one processor 602;Wherein, the memory 601 is used to store at least one program, and when the at least one program is executed by the at least one processor 602, the at least one processor 602 executes the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiment of the application.

[0143] Please refer to Figure 7 , Figure 7 The module block diagram of the computer readable storage medium provided by the embodiment of the application.

[0144] As Figure 7As shown, the application also provides a computer readable storage medium 700, which stores executable instructions 710, and the executable instructions 710 are executed to perform the non-uniform sampling rotating MIMO super-resolution imaging method provided by the embodiments of the application.

[0145] The steps of a method or algorithm described in connection with the embodiments disclosed herein can be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module can reside in random access memory (RAM), flash memory, read-only memory (ROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0146] In the above embodiments, all or part of the embodiments can be implemented by software, hardware, firmware, or any combination thereof. When implemented by software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments are generated. The computer can be a general purpose computer, a special purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer readable storage medium or transferred from one computer readable storage medium to another computer readable storage medium, for example, the computer instructions can be transferred from one website, computer, server or data center to another website, computer, server or data center through wired (such as coaxial cable, optical fiber, digital subscriber line) or wireless (such as infrared, wireless, microwave, etc.) mode. The computer readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. containing one or more available media sets. The available media can be magnetic media (such as floppy disk, hard disk, magnetic tape), optical media (such as DVD), or semiconductor media (such as solid state disk), etc.

[0147] It should be noted that each of the embodiments in the present application is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts of each embodiment can be referred to each other. For product class embodiments, since they are similar to method class embodiments, the description is relatively simple, and the relevant parts can be referred to the part of the method class embodiment.

[0148] It is also important to note that the use of relational terms such as first and second, and the like, are used solely to distinguish one from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0149] The above description of disclosed embodiments provides enabling concepts for making or using the present content. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein can be applied to other embodiments without departing from the spirit or scope of the claimed content. Thus, the present content is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the claims, the principles and the novel features disclosed herein.

Claims

1. A non-uniform sampling rotational MIMO super-resolution imaging method, characterized in that, The method comprises the following steps: synthesizing a virtual aperture based on collected radar data, and establishing a coordinate system based on the virtual aperture; calculating position information of radar antennas in the coordinate system, and screening all the radar antennas based on a preset step length to obtain effective virtual antennas; obtaining target distance information based on collected radar echo signals; constructing a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antennas, and generating a local super-resolution image based on the rotating super-resolution dictionary; wherein the rotating super-resolution dictionary contains target angle information; performing coordinate conversion and data splicing on each local super-resolution image to generate a global super-resolution image.

2. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 1, wherein, The step of synthesizing a virtual aperture based on collected radar data, and establishing a coordinate system based on the virtual aperture comprises the following steps: A radar turntable with a disc radius of r is rotated at a constant angular velocity ω, and a radar emits a chirp signal every frame interval T f Frame data is transmitted; wherein each frame data contains N C chirp signals with a time interval of T C ​ obtaining N frames of first echo data through the radar, and synthesizing a virtual aperture based on the N frames of first echo data; The first The start time of the first chirp signal in the frame data is taken as the virtual aperture center; establishing a coordinate system with the center of the virtual aperture as the coordinate origin, the horizontal direction of the radar at this moment as the positive direction of the x-axis, and the vertical direction of the radar at this moment as the positive direction of the y-axis.

3. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 2, wherein, The step of calculating position information of radar antennas in the coordinate system comprises the following steps: calculating an angle θ of the radar relative to a radar initial position in a slow time dimension p,l ; wherein the angle θ p,l represents an angle of the radar in a virtual aperture in a pth frame and an lth chirp, θ0 represents an included angle between a radar initial orientation and a current orientation of the radar, a number of calculated angles is N*N c , and an expression of the angle θ p,l is: Based on angle θ p,l Calculate the angular difference Δθ between each radar position and the center of the virtual aperture. p,l ; where the angle difference Δθ p,l The expression is: based on the angle difference Δθ p,l calculating the position P of the radar in the coordinate system in the slow time dimension p,l = (X p,l , Y p,l ); wherein the position P p,l = (X p,l , Y p,l ) represents the position of the radar at the pth frame and the lth chirp, the position number is N*N c , and the expression of the position P p,l is: Based on the antenna spacing of the radar Calculate the position P of the radar antenna in the coordinate system based on the radius r of the disk. p,l,k =(X p,l,k ,Y p,l,k ); where λ represents the wavelength of the radar transmitted signal, bit P p,l,k = (X p,l,k , Y p,l,k ) denotes the position of the radar in the pth frame, the lth chirp, and the kth antenna, and the expression of the position P p,l,k is:

4. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 3, wherein, The step of screening all the radar antennas based on a preset step length to obtain effective virtual antennas comprises the following steps: From the first antenna as a starting point, with a step size of The radar antenna is traversed and screened to obtain a screened antenna and an antenna index array; wherein the number of screened antennas is G, and the expression of the antenna index array is Idx=[Idx1, Idx2, …, IdxG]. The antennas screened are marked according to a slow time dimension, and valid virtual antennas are obtained based on the antenna index array; wherein the valid virtual antennas are marked as Rx1, Rx2, Rx3, …, RxG, and the corresponding positions are represented as P Rx1 =(X Rx1 ,Y Rx1 ), P Rx2 =(X Rx2 ,Y Rx2 ), …, P RxG =(X RxG ,Y RxG ).

5. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 4, wherein, The step of obtaining target distance information based on collected radar echo signals comprises the following steps: preprocessing the collected radar echo signals to obtain first discrete echo signals; wherein the first discrete echo signals are represented as Y[n,m,k], n is a fast time dimension, representing the nth sampling point, m is a slow time dimension, representing the mth linear frequency modulation continuous wave signal, and k is an antenna dimension, representing the received signal of the kth channel; performing windowing processing on the first discrete echo signals in the fast time dimension to obtain second discrete echo signals; performing a fast Fourier transform on the second discrete echo signal to obtain a third discrete echo signal; wherein the third discrete echo signal is a range-slow-time-dimension-antenna-dimension signal, denoted as d∈[1,N d ], N d is the number of distance sampling points; performing target distance detection on the third discrete echo signals through a constant false alarm rate algorithm to obtain target distance information.

6. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 5, wherein, The step of constructing a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antennas comprises the following steps: calculating the phase difference between the target and the effective virtual antennas and the phase difference between the target and the center of the virtual aperture; wherein the phase difference matrix composed of the phase differences is represented as: wherein azi represents the target azimuth angle, and R represents the distance from the target to the radar; constructing a rotating super-resolution dictionary based on the azimuth angle and the phase difference matrix; wherein the rotating super-resolution dictionary is represented as: Ψ = [Δφ(Azi(1)), Δφ(Azi(2))..., Δφ(Azi(N A ))], Wherein, the azimuth range of the detection is Azi, the number of the detection angle is N A , the dictionary size is G*N A .

7. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 6, wherein, The step of generating a local super-resolution image based on the rotating super-resolution dictionary comprises the following steps: obtaining second echo data corresponding to the distance unit from the radar echo signals based on the target distance information; arranging the second echo data in the slow time dimension to obtain third echo data; extracting the third echo data based on the antenna index array to obtain fourth echo data; wherein the fourth echo data is represented as obtaining a target angle spectrum based on the fourth echo data; wherein the target angle spectrum is represented as: where abs is a modulo operation, P is a target power spectrum, and N is a length A and θ is a target angle. generating a local super-resolution image based on the target angle spectrum.

8. The non-uniform sampling rotational MIMO super-resolution imaging method of claim 7, wherein, The step of performing coordinate conversion and data splicing on each local super-resolution image to generate a global super-resolution image comprises the following steps: calculating an angle difference between a corresponding coordinate system in each of the partial super-resolution images and a stationary indoor coordinate system; wherein the angle difference of the i-th partial super-resolution image is is represented as: Wherein, the radar generates images frame by frame from the Nth frame, the time difference between two images is radar frame interval T f , the angle of the disc rotation is T f w; performing coordinate conversion on each local super-resolution image based on the angle difference; wherein the coordinate conversion of the ith local super-resolution image is represented as: wherein (x i ′ ,y i ′ ) is the position of the target in the virtual aperture center rectangular coordinate system in the i-th image, (x i ,y i ) is the position of the target in the static indoor rectangular coordinate system in the i-th image; The local super-resolution images after coordinate conversion are spliced to generate a global super-resolution image.

9. A non-uniformly sampled rotational MIMO super-resolution imaging system, characterized in that, The method comprises the steps of: The aperture synthesis module is configured to synthesize a virtual aperture based on the collected radar data, and establish a coordinate system based on the virtual aperture; The antenna screening module is configured to calculate the position information of the radar antennas in the coordinate system, and screen all the radar antennas based on a preset step to obtain effective virtual antennas; The target acquisition module is configured to acquire target distance information based on the collected radar echo signals; The first generation module is configured to construct a rotating super-resolution dictionary based on the target distance information and the position information of the effective virtual antennas, and generate local super-resolution images based on the rotating super-resolution dictionary; wherein the rotating super-resolution dictionary contains target angle information; The second generation module is configured to perform coordinate conversion and data splicing on each of the local super-resolution images to generate a global super-resolution image.

10. A radar, characterized by The computer program is executed by the processor to implement the steps of the non-uniform sampling rotating MIMO super-resolution imaging method in any one of claims 1 to 8. The computer program is executed by the processor to implement the steps of the non-uniform sampling rotating MIMO super-resolution imaging method in any one of claims 1 to 8. ​ 11. A computer readable storage medium having stored thereon a computer program, characterized in that ​

Images