Echo map prediction method and system based on prior data, terminal and storage medium
By combining radio frequency and scattering center models to construct an echo model and using U-Net network to train and predict echo maps, the problem that mechanistic models in existing technologies cannot accurately predict echo map features is solved, and high-precision echo map prediction is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-01
- Publication Date
- 2026-06-23
Smart Images
Figure CN122260265A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of echo data prediction technology, and in particular to an echo map prediction method, system, terminal, and computer-readable storage medium based on prior data. Background Technology
[0002] Radar echo maps are the core visualization achievement of radar detection technology. They intuitively present the spatial distribution, intensity, and motion characteristics of targets by processing and mapping electromagnetic wave reflection signals.
[0003] However, radar echo maps contain a large amount of complex information. Predicting echo maps based on scene information is a complex task for radar. Mechanism-based models simplify the internal components of radar and ignore the influence of irrational factors. Although this can reduce the complexity of modeling, it cannot accurately reproduce the details of frequency offset, spectrum spread and amplitude changes inside the radar radio frequency circuit. Compared with real test data, there is a gap in the presentation of subtle signal features.
[0004] Therefore, existing technologies still need to be improved and developed. Summary of the Invention
[0005] The main objective of this invention is to provide an echo map prediction method, system, terminal, and computer-readable storage medium based on prior data, aiming to solve the problem that existing mechanism-based radar models cannot accurately predict all features in echo maps.
[0006] To achieve the above objectives, the present invention provides an echo map prediction method based on prior data, the method comprising the following steps: An echo model consisting of a radio frequency model and a scattering center model is obtained, multiple parameters of the training vehicle are defined, and the reflected signal of the training vehicle is received using the echo model based on all the parameters. Multiple signals are acquired using the reflected signals, and after filtering and processing all the signals, they are sampled to obtain the original data; Perform a multi-dimensional fast Fourier transform on the original data to output the echo map of the reflected signal; The U-Net network is trained using the echo map, and the trained U-Net network is used to predict the target vehicle, outputting the target echo map of the target vehicle.
[0007] Optionally, in the echo map prediction method based on prior data, the radio frequency model includes: a signal source, a power divider, a transmitting antenna, a receiving antenna, a mixer, and an analog-to-digital conversion module.
[0008] Optionally, the echo map prediction method based on prior data, wherein obtaining the echo model composed of a radio frequency model and a scattering center model, defining multiple parameters of the training vehicle, and using the echo model to receive the reflected signal of the training vehicle according to all the parameters, specifically includes: Obtain the radio frequency model and the scattering center model, and construct the echo model using the radio frequency model and the scattering center model; Define the radial velocity of the training vehicle, the power and frequency of the transmitted signal, the relative displacement value, and the signal delay; Based on radial velocity, power and frequency of transmitted signal, relative displacement value and signal delay, the echo model receives the reflected signal of the training vehicle through the receiving antenna; The reflected signal is obtained by reflecting a signal transmitted by the radar to the training vehicle.
[0009] Optionally, the echo map prediction method based on prior data, wherein defining the radial velocity of the training vehicle, the power and frequency of the transmitted signal, the relative displacement value, and the signal delay specifically includes: The radial velocity of the training vehicle is obtained, and the transmission frequency of the transmitted signal is defined using the signal source; Calculate the reflection frequency of the reflected signal based on the transmission frequency and the radial velocity: ; in, Indicates the reflection frequency. Indicates the transmission frequency. Indicates radial velocity, Indicates the wavelength of the transmitted signal; The reflected power of the reflected signal is calculated using the reflected frequency; The element spacing is obtained using an array measurement device, and the relative displacement between different receiving channels is calculated based on the offset angle between the array measurement device and the signal source. ; in, Indicates the relative displacement value. d Indicates the spacing between array elements. Indicates the offset angle; Calculate the signal delay of the reflected signal based on the distance between the echo model and the training vehicle: ; in, Indicates signal delay. Indicates distance, c Indicates the speed of signal propagation.
[0010] Optionally, the echo map prediction method based on prior data, wherein the step of acquiring multiple signals using the reflected signal, filtering and processing all the signals before sampling to obtain the original data, specifically includes: The power divider divides the reflected signal into a current frame signal and an radio frequency signal, and sends the current frame signal and the radio frequency signal to the mixer. The mixer is used to subtract the frequency of the current frame signal from that of the radio frequency signal received by the receiving antenna to obtain the intermediate frequency signal; After sampling the intermediate frequency signal using the analog-to-digital conversion module, the original data is obtained.
[0011] Optionally, in the echo map prediction method based on prior data, the echo map includes: a distance-velocity map and a distance-angle map; The step of performing a multi-dimensional fast Fourier transform on the original data to output the echo map of the reflected signal specifically includes: In the fast time dimension of the original data, a fast Fourier transform is performed on each pulse and each receiving antenna to obtain the distance information in the original data; In the slow time dimension of the original data, a fast Fourier transform is performed on each range cell and each receiving antenna to obtain the range-velocity map in the original data; In the receiving antenna dimension of the original data, a fast Fourier transform is performed on each range-velocity unit in the range-velocity map to obtain the angle dimension information in the original data, and the angle dimension information is sliced to obtain the range-angle map.
[0012] Optionally, the echo map prediction method based on prior data, wherein training the U-Net network using the echo map and predicting the target vehicle using the trained U-Net network to output the target echo map of the target vehicle specifically includes: The range-velocity map and the range-angle map are input into the constructed U-Net network, and the real echo map output by the real radar is input into the U-Net network for fitting in order to train the U-Net network; The target reflection signal reflected by the target vehicle is input into a trained U-Net network for prediction, and the target echo map of the target vehicle is output.
[0013] Furthermore, to achieve the above objectives, the present invention also provides an echo map prediction system based on prior data, wherein the echo map prediction system based on prior data includes: The signal acquisition module is used to acquire an echo model composed of a radio frequency model and a scattering center model, define multiple parameters of the training vehicle, and use the echo model to receive the reflected signal of the training vehicle according to all the parameters. The data acquisition module is used to acquire multiple signals using the reflected signal, and to sample all the signals after filtering and processing them to obtain the original data; The data processing module is used to perform multi-dimensional fast Fourier transform on the raw data and output the echo map of the reflected signal. The model training module is used to train the U-Net network using the echo map, and to use the trained U-Net network to predict the target vehicle and output the target echo map of the target vehicle.
[0014] Furthermore, to achieve the above objectives, the present invention also provides a terminal, wherein the terminal includes: a memory, a processor, and an echo map prediction program based on prior data stored in the memory and executable on the processor, wherein when the echo map prediction program based on prior data is executed by the processor, it implements the steps of the echo map prediction method based on prior data as described above.
[0015] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores an echo map prediction program based on prior data, and when the echo map prediction program based on prior data is executed by a processor, it implements the steps of the echo map prediction method based on prior data as described above.
[0016] In this invention, an echo model composed of a radio frequency model and a scattering center model is obtained. Multiple parameters of a training vehicle are defined, and the echo model is used to receive the reflected signals of the training vehicle based on all the parameters. Various signals are acquired using the reflected signals, and after filtering and processing, all signals are sampled to obtain raw data. A multi-dimensional fast Fourier transform is performed on the raw data to output the echo map of the reflected signals. The echo map is used to train a U-Net network (convolutional neural network), and the trained network is used to predict the target vehicle, outputting the target echo map of the target vehicle. This invention leverages the advantages of a mechanistic-based signal-level radar model as a prior model in terms of real-time performance, high generalization ability, and accuracy in predicting key features. It generates the backbone features in the echo map based on scene information, and then uses a data-driven model to quickly adjust the detailed parts of the echo map, compensating for the shortcomings of the prior model in predicting detailed features and significantly improving the prediction accuracy of the echo map. Attached Figure Description
[0017] Figure 1 This is a flowchart of a preferred embodiment of the echo map prediction method based on prior data of the present invention; Figure 2 This is a schematic diagram of a radar model of a preferred embodiment of the echo map prediction method based on prior data of the present invention; Figure 3 This is a schematic diagram of the data-driven model of a preferred embodiment of the echo map prediction method based on prior data of the present invention; Figure 4 This is an echo map of a preferred embodiment of the echo map prediction method based on prior data of the present invention; Figure 5 This is a structural diagram of a preferred embodiment of the echo map prediction system based on prior data of the present invention; Figure 6 This is a structural diagram of a preferred embodiment of the terminal of the present invention. Detailed Implementation
[0018] To make the objectives, technical solutions, and advantages of this invention clearer and more explicit, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
[0019] The echo map prediction method based on prior data described in the preferred embodiment of the present invention, such as... Figure 1 As shown, the echo map prediction method based on prior data includes the following steps: Step S10: Obtain the echo model composed of the radio frequency model and the scattering center model, define multiple parameters of the training vehicle, and use the echo model to receive the reflected signal of the training vehicle according to all the parameters.
[0020] In the embodiments disclosed in this invention, such as Figure 2 As shown, a signal-level radar model based on a mechanism architecture is first used as a priori model. Taking advantage of its advantages in real-time performance, high generalization ability and main feature prediction accuracy, the main features in the echo map are generated based on the scene information. The priori model is constructed from a radio frequency model and a scattering center model. The radio frequency model includes: signal source, power divider, transmitting antenna, receiving antenna, mixer and analog-to-digital conversion module.
[0021] Specifically, an radio frequency model and a scattering center model are obtained, and an echo model is constructed using the radio frequency model and the scattering center model; Define the radial velocity of the training vehicle, the power and frequency of the transmitted signal, the relative displacement value, and the signal delay; Based on radial velocity, power and frequency of transmitted signal, relative displacement value and signal delay, the echo model receives the reflected signal of the training vehicle through the receiving antenna; The reflected signal is obtained by reflecting a signal transmitted by the radar to the training vehicle.
[0022] Furthermore, the radial velocity of the training vehicle is obtained, and the transmission frequency of the transmitted signal is defined using the signal source; Calculate the reflection frequency of the reflected signal based on the transmission frequency and the radial velocity: ; in, Indicates the reflection frequency. Indicates the transmission frequency. Indicates radial velocity, Indicates the wavelength of the transmitted signal; The reflected power of the reflected signal is calculated using the reflected frequency; The element spacing is obtained using an array measurement device, and the relative displacement between different receiving channels is calculated based on the offset angle between the array measurement device and the signal source. ; in, Indicates the relative displacement value. d Indicates the spacing between array elements. Indicates the offset angle; Calculate the signal delay of the reflected signal based on the distance between the echo model and the training vehicle: ; in, Indicates signal delay. Indicates distance, c Indicates the speed of signal propagation.
[0023] In the embodiments disclosed in this invention, the target vehicle is equivalent to a point target at the center of the vehicle using the scattering center method. This process ignores the multipath effect of the radar signal and excludes the influence of other factors in the scene (such as ground reflection and fixed objects), assuming that the radar only receives the signal reflected back from the equivalent point of the target. The features of the prior model ensure the accuracy of the backbone features of the echo map, avoid deviation from the true value, and provide a reliable prediction basis for the data-driven model.
[0024] Step S20: Use the reflected signal to acquire multiple signals, and after filtering and processing all the signals, sample them to obtain the original data.
[0025] Specifically, the power divider divides the reflected signal into a current frame signal and a radio frequency signal, and sends the current frame signal and the radio frequency signal to the mixer; The mixer is used to subtract the frequency of the current frame signal from that of the radio frequency signal received by the receiving antenna to obtain the intermediate frequency signal; After sampling the intermediate frequency signal using the analog-to-digital conversion module, the original data is obtained.
[0026] The signal source is used to set parameters such as the frequency of the transmitted signal. The power divider divides the signal into multiple parts. One part is transmitted to the transmitting antenna (RF signal), and another part is transmitted to the mixer. This part is also called the current frame signal. This process can simulate the signal processing process of real radar and has a higher degree of realism.
[0027] Furthermore, the mixer processes the signal of the current frame and the radio frequency signal received by the receiving antenna, subtracting the frequency parameter of the radio frequency signal from the frequency parameter of the current frame signal. The resulting signal after frequency subtraction is the intermediate frequency signal, which is then output. Finally, the analog-to-digital converter module is used for sampling to obtain the raw data. This process simulates the actual sampling process, sampling the signal according to a preset sampling rate, which can be set as needed.
[0028] Step S30: Perform a multi-dimensional fast Fourier transform on the original data and output the echo map of the reflected signal.
[0029] The echo map includes: a distance-velocity map and a distance-angle map; after obtaining the main features of the echo map using a prior model, such as Figure 3 As shown, data-driven models can be used to quickly adjust the details of the echo map.
[0030] Specifically, in the fast time dimension of the original data, a fast Fourier transform is performed on each pulse and each receiving antenna to obtain the distance information in the original data; In the slow time dimension of the original data, a fast Fourier transform is performed on each range cell and each receiving antenna to obtain the range-velocity map in the original data; In the receiving antenna dimension of the original data, a fast Fourier transform is performed on each range-velocity unit in the range-velocity map to obtain the angle dimension information in the original data, and the angle dimension information is sliced to obtain the range-angle map.
[0031] The model receives a three-dimensional matrix of raw data: [fast time sampling points * slow time pulse count * number of receiving antennas]. Three Fast Fourier Transforms (FFTs) are performed on each of the three dimensions of the raw data to obtain the radar echo map: For ranging, a first FFT is performed along the fast time dimension for each pulse and each receiving antenna to obtain range information; for velocity, a second FFT is performed along the slow time dimension (slow time pulse sequence) for each range cell and each receiving antenna to obtain the range-velocity map; and for angle measurement, a third FFT is performed along the receiving antenna dimension for each range-velocity cell to obtain the angle information. The range-angle map can then be obtained by slicing the data.
[0032] Step S40: Train the U-Net network using the echo map, and use the trained U-Net network to predict the target vehicle, outputting the target echo map of the target vehicle.
[0033] Among them, such as Figure 4 As shown, the data-driven model, leveraging its high accuracy and robustness in simple mapping problems, uses the echo map generated by the prior model as input to predict results that are closer to the true echo map. In this way, the data-driven model compensates for the shortcomings of the prior model in predicting detailed features, thereby significantly improving the prediction accuracy of the echo map.
[0034] Specifically, the range-velocity map and the range-angle map are input into the constructed U-Net network, and the real echo map output by the real radar is input into the U-Net network for fitting in order to train the U-Net network; The target reflection signal reflected by the target vehicle is input into a trained U-Net network for prediction, and the target echo map of the target vehicle is output.
[0035] In another embodiment of this invention, the echo model calculates the amplitude, frequency, and time delay of the target echo signal according to a formula. All components in the mechanism-based radar RF model are ideal components, with an operating frequency of 77 GHz, a drawdown of 400 MHz, and a period of 20 μs. An omnidirectional antenna is used for both transmission and reception. The mixer model calculates the amplitude, frequency, and phase information of the intermediate frequency signal by mixing the local oscillator signal with the RF signal according to the received signal delay. Then, the calculated signal is processed by a Fast Fourier Transform to obtain the echo map, where... Figure 4 Figure (a) shows a distance-angle plot, as shown in Figure (a). Figure 4 Figure (b) shows the distance-velocity plot.
[0036] Furthermore, a radar prior model is constructed using the modeling method disclosed in this invention, and the U-Net network is trained. The training parameters include the number of training epochs (40), the number of training samples per batch (24), the learning rate (0.05), and the weight decay (...). The parameters are: momentum (0.9), number of input channels (2), number of output channels (2), whether to use bilinear interpolation (no), optimizer (RMSprop, Root Mean Square Propagation), and loss function (MSELoss, Mean Squared Error Loss). The echo map predicted by the U-Net model trained based on these parameters has significant improvements in the prediction of detailed features compared with the echo map of the ideal mechanism model, and has improved in the target distribution area.
[0037] This invention leverages the advantages of a mechanistic-based signal-level radar model as a prior model in terms of real-time performance, high generalization ability, and accuracy in predicting key features. It generates the backbone features in the echo map based on scene information, and then uses a data-driven model to quickly adjust the details of the echo map, thus compensating for the shortcomings of the prior model in predicting detailed features and significantly improving the prediction accuracy of the echo map.
[0038] Furthermore, such as Figure 5 As shown, based on the above-described echo map prediction method based on prior data, the present invention also provides an echo map prediction system based on prior data, wherein the echo map prediction system based on prior data includes: The signal acquisition module 51 is used to acquire an echo model composed of a radio frequency model and a scattering center model, define multiple parameters of the training vehicle, and use the echo model to receive the reflected signal of the training vehicle according to all the parameters. Data acquisition module 52 is used to acquire multiple signals using the reflected signal, and to sample all the signals after filtering and processing to obtain raw data; The data processing module 53 is used to perform multi-dimensional fast Fourier transform on the raw data and output the echo map of the reflected signal. The model training module 54 is used to train the U-Net network using the echo map, and to predict the target vehicle using the trained U-Net network, and output the target echo map of the target vehicle.
[0039] Furthermore, such as Figure 6 As shown, based on the above-mentioned echo map prediction method and system based on prior data, the present invention also provides a terminal, which includes a processor 10, a memory 20 and a display 30. Figure 6Only some of the terminal components are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.
[0040] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory. In other embodiments, the memory 20 may be an external storage device of the terminal, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, the memory 20 may include both internal and external storage devices. The memory 20 is used to store application software and various types of data installed on the terminal, such as the program code installed on the terminal. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores an echo map prediction program 40 based on prior data, which can be executed by the processor 10 to implement the echo map prediction method based on prior data in this application.
[0041] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in the memory 20 or process data, such as executing the echo map prediction method based on prior data.
[0042] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the terminal and to display a visual user interface. The components of the terminal communicate with each other via a system bus.
[0043] In one embodiment, when the processor 10 executes the echo map prediction program 40 based on prior data in the memory 20, the following steps are performed: An echo model consisting of a radio frequency model and a scattering center model is obtained, multiple parameters of the training vehicle are defined, and the reflected signal of the training vehicle is received using the echo model based on all the parameters. Multiple signals are acquired using the reflected signals, and after filtering and processing all the signals, they are sampled to obtain the original data; Perform a multi-dimensional fast Fourier transform on the original data to output the echo map of the reflected signal; The U-Net network is trained using the echo map, and the trained U-Net network is used to predict the target vehicle, outputting the target echo map of the target vehicle.
[0044] The radio frequency model includes: a signal source, a power divider, a transmitting antenna, a receiving antenna, a mixer, and an analog-to-digital conversion module.
[0045] The step of acquiring an echo model composed of a radio frequency model and a scattering center model, defining multiple parameters of the training vehicle, and using the echo model to receive the reflected signal of the training vehicle according to all the parameters specifically includes: Obtain the radio frequency model and the scattering center model, and construct the echo model using the radio frequency model and the scattering center model; Define the radial velocity of the training vehicle, the power and frequency of the transmitted signal, the relative displacement value, and the signal delay; Based on radial velocity, power and frequency of transmitted signal, relative displacement value and signal delay, the echo model receives the reflected signal of the training vehicle through the receiving antenna; The reflected signal is obtained by reflecting a signal transmitted by the radar to the training vehicle.
[0046] Specifically, the definition of the radial velocity of the training vehicle, the power and frequency of the transmitted signal, the relative displacement value, and the signal delay includes: The radial velocity of the training vehicle is obtained, and the transmission frequency of the transmitted signal is defined using the signal source; Calculate the reflection frequency of the reflected signal based on the transmission frequency and the radial velocity: ; in, Indicates the reflection frequency. Indicates the transmission frequency. Indicates radial velocity, Indicates the wavelength of the transmitted signal; The reflected power of the reflected signal is calculated using the reflected frequency; The element spacing is obtained using an array measurement device, and the relative displacement between different receiving channels is calculated based on the offset angle between the array measurement device and the signal source. ; in, Indicates the relative displacement value. d Indicates the spacing between array elements. Indicates the offset angle; Calculate the signal delay of the reflected signal based on the distance between the echo model and the training vehicle: ; in, Indicates signal delay. Indicates distance, c Indicates the speed of signal propagation.
[0047] Specifically, the process of acquiring multiple signals using the reflected signal, filtering and processing all the signals, and then sampling to obtain the original data includes: The power divider divides the reflected signal into a current frame signal and an radio frequency signal, and sends the current frame signal and the radio frequency signal to the mixer. The mixer is used to subtract the frequency of the current frame signal from that of the radio frequency signal received by the receiving antenna to obtain the intermediate frequency signal; After sampling the intermediate frequency signal using the analog-to-digital conversion module, the original data is obtained.
[0048] The echo map includes: a distance-velocity map and a distance-angle map; The step of performing a multi-dimensional fast Fourier transform on the original data to output the echo map of the reflected signal specifically includes: In the fast time dimension of the original data, a fast Fourier transform is performed on each pulse and each receiving antenna to obtain the distance information in the original data; In the slow time dimension of the original data, a fast Fourier transform is performed on each range cell and each receiving antenna to obtain the range-velocity map in the original data; In the receiving antenna dimension of the original data, a fast Fourier transform is performed on each range-velocity unit in the range-velocity map to obtain the angle dimension information in the original data, and the angle dimension information is sliced to obtain the range-angle map.
[0049] Specifically, the step of training the U-Net network using the echo map and then using the trained U-Net network to predict the target vehicle and output the target echo map of the target vehicle includes: The range-velocity map and the range-angle map are input into the constructed U-Net network, and the real echo map output by the real radar is input into the U-Net network for fitting in order to train the U-Net network; The target reflection signal reflected by the target vehicle is input into a trained U-Net network for prediction, and the target echo map of the target vehicle is output.
[0050] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores an echo map prediction program based on prior data, and the echo map prediction program based on prior data, when executed by a processor, implements the steps of the echo map prediction method based on prior data as described above.
[0051] In summary, this invention provides an echo map prediction method and related equipment based on prior data. The method includes: acquiring an echo model composed of a radio frequency model and a scattering center model; defining multiple parameters of a training vehicle; receiving the reflected signal of the training vehicle using the echo model according to all the parameters; acquiring multiple signals using the reflected signal, and sampling after filtering and processing all the signals to obtain raw data; performing a multi-dimensional fast Fourier transform on the raw data to output the echo map of the reflected signal; training a U-Net network using the echo map, and using the trained U-Net network to predict the target vehicle, outputting the target echo map of the target vehicle. This invention utilizes the advantages of a mechanistic architecture-based signal-level radar model as a prior model in terms of real-time performance, high generalization ability, and accuracy in predicting key features. It generates the backbone features in the echo map based on scene information, and then uses a data-driven model to quickly adjust the detailed parts of the echo map, compensating for the shortcomings of the prior model in predicting detailed features and significantly improving the prediction accuracy of the echo map.
[0052] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.
[0053] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.). The program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The computer-readable storage medium can be a memory, magnetic disk, optical disk, etc.
[0054] It should be understood that the application of the present invention is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.
Claims
1. A method for predicting echo maps based on prior data, characterized in that, The echo map prediction method based on prior data includes: An echo model consisting of a radio frequency model and a scattering center model is obtained, multiple parameters of the training vehicle are defined, and the reflected signal of the training vehicle is received using the echo model based on all the parameters. Multiple signals are acquired using the reflected signals, and after filtering and processing all the signals, they are sampled to obtain the original data; Perform a multi-dimensional fast Fourier transform on the original data to output the echo map of the reflected signal; The U-Net network is trained using the echo map, and the trained U-Net network is used to predict the target vehicle, outputting the target echo map of the target vehicle.
2. The echo map prediction method based on prior data according to claim 1, characterized in that, The radio frequency model includes: a signal source, a power divider, a transmitting antenna, a receiving antenna, a mixer, and an analog-to-digital conversion module.
3. The echo map prediction method based on prior data according to claim 2, characterized in that, The process of acquiring an echo model composed of a radio frequency model and a scattering center model, defining multiple parameters of the training vehicle, and using the echo model to receive the reflected signal of the training vehicle based on all the parameters specifically includes: Obtain the radio frequency model and the scattering center model, and construct the echo model using the radio frequency model and the scattering center model; Define the radial velocity of the training vehicle, the power and frequency of the transmitted signal, the relative displacement value, and the signal delay; Based on radial velocity, power and frequency of transmitted signal, relative displacement value and signal delay, the echo model receives the reflected signal of the training vehicle through the receiving antenna; The reflected signal is obtained by reflecting a signal transmitted by the radar to the training vehicle.
4. The echo map prediction method based on prior data according to claim 3, characterized in that, The definition of the training vehicle's radial velocity, transmitted signal power and frequency, relative displacement value, and signal delay specifically includes: The radial velocity of the training vehicle is obtained, and the transmission frequency of the transmitted signal is defined using the signal source; Calculate the reflection frequency of the reflected signal based on the transmission frequency and the radial velocity: ； in, Indicates the reflection frequency. Indicates the transmission frequency. Indicates radial velocity, Indicates the wavelength of the transmitted signal; The reflected power of the reflected signal is calculated using the reflected frequency; The element spacing is obtained using an array measurement device, and the relative displacement between different receiving channels is calculated based on the offset angle between the array measurement device and the signal source. ； in, Indicates the relative displacement value. d Indicates the spacing between array elements. Indicates the offset angle; Calculate the signal delay of the reflected signal based on the distance between the echo model and the training vehicle: ； in, Indicates signal delay. Indicates distance, c Indicates the speed of signal propagation.
5. The echo map prediction method based on prior data according to claim 2, characterized in that, The process of acquiring multiple signals using the reflected signal, filtering and processing all the signals, and then sampling them to obtain the original data specifically includes: The power divider divides the reflected signal into a current frame signal and an radio frequency signal, and sends the current frame signal and the radio frequency signal to the mixer. The mixer is used to subtract the frequency of the current frame signal from that of the radio frequency signal received by the receiving antenna to obtain the intermediate frequency signal; After sampling the intermediate frequency signal using the analog-to-digital conversion module, the original data is obtained.
6. The echo map prediction method based on prior data according to claim 1, characterized in that, The echo graph includes: a distance-velocity graph and a distance-angle graph; The step of performing a multi-dimensional fast Fourier transform on the original data to output the echo map of the reflected signal specifically includes: In the fast time dimension of the original data, a fast Fourier transform is performed on each pulse and each receiving antenna to obtain the distance information in the original data; In the slow time dimension of the original data, a fast Fourier transform is performed on each range cell and each receiving antenna to obtain the range-velocity map in the original data; In the receiving antenna dimension of the original data, a fast Fourier transform is performed on each range-velocity unit in the range-velocity map to obtain the angle dimension information in the original data, and the angle dimension information is sliced to obtain the range-angle map.
7. The echo map prediction method based on prior data according to claim 6, characterized in that, The process of training the U-Net network using the echo image and then using the trained U-Net network to predict the target vehicle and output the target echo image of the target vehicle specifically includes: The range-velocity map and the range-angle map are input into the constructed U-Net network, and the real echo map output by the real radar is input into the U-Net network for fitting in order to train the U-Net network; The target reflection signal reflected by the target vehicle is input into a trained U-Net network for prediction, and the target echo map of the target vehicle is output.
8. An echo map prediction system based on prior data, characterized in that, The echo map prediction system based on prior data is used to implement the echo map prediction method based on prior data as described in any one of claims 1-7, wherein the echo map prediction system based on prior data includes: The signal acquisition module is used to acquire an echo model composed of a radio frequency model and a scattering center model, define multiple parameters of the training vehicle, and use the echo model to receive the reflected signal of the training vehicle according to all the parameters. The data acquisition module is used to acquire multiple signals using the reflected signal, and to sample all the signals after filtering and processing them to obtain the original data; The data processing module is used to perform multi-dimensional fast Fourier transform on the raw data and output the echo map of the reflected signal. The model training module is used to train the U-Net network using the echo map, and to use the trained U-Net network to predict the target vehicle and output the target echo map of the target vehicle.
9. A terminal, characterized in that, The terminal includes: a memory, a processor, and an echo map prediction program based on prior data stored in the memory and executable on the processor. When the echo map prediction program based on prior data is executed by the processor, it implements the steps of the echo map prediction method based on prior data as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores an echo map prediction program based on prior data, which, when executed by a processor, implements the steps of the echo map prediction method based on prior data as described in any one of claims 1-7.