Temporary source positioning method, device and equipment based on multi-beam imaging and storage medium

Abstract

The invention provides a transient source positioning method, device and equipment based on multi-beam imaging and a storage medium, and relates to the technical field of radio astronomy observation and data processing. The method comprises the following steps: acquiring voltage data acquired by each receiving station of a radio array; constructing a beam weight tensor according to a plurality of preset target directions and the geometric structure of the radio array; and carrying out complex number tensor multiplication on the voltage data and the beam weight tensor, and synthesizing digital beams in one-to-one correspondence with all target directions in parallel. Obtaining a beam intensity sequence according to the digital beam, and mapping the beam intensity sequence to generate a two-dimensional image; and carrying out detection processing on the two-dimensional image, and identifying and positioning a temporary source. The method is used for achieving the effect of improving the positioning efficiency of the transient source according to multi-beam imaging.

Description

Technical Field

[0001] This application relates to the field of radio astronomy observation and data processing technology, and more specifically, to a transient source localization method, apparatus, device, and storage medium based on multi-beam imaging. Background Technology

[0002] Currently, modern radio telescope arrays such as the Murchison Widefield Array (MWA), the Low Frequency Array (LOFAR), the MeerKAT radio telescope, and the future Square Kilometre Array (SKA) can all record complex voltage signals from multiple stations. These signals are typically acquired with high temporal resolution and can be used for rapid transient source research. For example, the voltage acquisition system of the MWA can acquire the raw voltage of the entire array on a sub-millisecond scale. However, current technologies still suffer from enormous computational demands and insufficient real-time performance in performing multibeamforming, short-time integration, and rapid imaging of voltage data.

[0003] In existing technologies, multi-beamforming under traditional CPU architectures typically employs a cyclic processing approach. This cyclic processing approach involves sequentially calculating each beam and each antenna channel (or each time sampling point) through nested multi-layered loops, ultimately synthesizing all beams. Then, based on the synthesized beams, transient sources are located.

[0004] However, in existing technologies, under scenarios with high beam count and high temporal resolution, the cyclic processing method, which requires serial calculation of each beam, struggles to maintain real-time speed, resulting in poor positioning efficiency. Furthermore, conventional radio interferometry imaging procedures require complete correlation, meshing, and Fourier transform processes, which are numerous and unsuitable for the rapid response requirements of fast transient sources. Summary of the Invention

[0005] The purpose of this application is to provide a transient source localization method, apparatus, device, and storage medium based on multibeam imaging, so as to solve the above-mentioned problems existing in the prior art and improve the localization efficiency of transient sources based on multibeam imaging.

[0006] Firstly, a transient source localization method based on multi-beam imaging is provided, which may include: Acquire voltage data collected by each receiving station of the radio array; Based on multiple preset target directions and the geometry of the radio array, a beam weight tensor is constructed; complex tensor multiplication is performed on the voltage data and the beam weight tensor to synthesize digital beams that correspond one-to-one with all the target directions in parallel. Based on the digital beam, a beam intensity sequence is obtained, and the beam intensity sequence is mapped to generate a two-dimensional image; the two-dimensional image is then processed to identify and locate transient sources.

[0007] Secondly, a transient source localization device based on multi-beam imaging is provided, the device comprising: The acquisition module is used to acquire voltage data collected by each receiving station of the radio array; The tensor module is used to construct a beam weight tensor based on multiple preset target directions and the geometry of the radio array; perform complex tensor multiplication operations on the voltage data and the beam weight tensor, and synthesize digital beams that correspond one-to-one with all the target directions in parallel. The identification module is used to obtain a beam intensity sequence based on the digital beam, and map the beam intensity sequence to generate a two-dimensional image; to perform detection processing on the two-dimensional image to identify and locate the radio transient source.

[0008] Thirdly, an electronic device is provided, which includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; When a processor executes a program stored in memory, it implements any of the steps described in the first aspect above.

[0009] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when executed by a processor, the computer program implements the steps of any of the methods described in the first aspect above.

[0010] This application provides a method, apparatus, device, and storage medium for locating transient sources based on multi-beam imaging. It acquires voltage data collected by each receiving station of a radio array. Based on multiple preset target directions and the geometry of the radio array, a beam weight tensor is constructed. Complex tensor multiplication is performed on the voltage data and the beam weight tensor to synthesize digital beams corresponding one-to-one with all target directions. Based on the digital beams, a beam intensity sequence is obtained, and the beam intensity sequence is mapped to generate a two-dimensional image. The two-dimensional image is then processed to identify and locate the transient source. In this scheme, multi-station voltage data collected by the radio array is used as input to construct a beam weight tensor, and the multi-beam formation process is transformed into tensor complex matrix multiplication, thereby achieving simultaneous synthesis of more than one hundred beams. Subsequently, the beam intensity sequence is acquired, and a two-dimensional image is directly generated using the beam intensity sequence. By detecting the two-dimensional image, transient source location can be completed in milliseconds. Therefore, by using parallel tensor computation, the synthesis computation of over 100 beams is changed from a serial loop to a single parallel scheduling, which significantly improves the efficiency of multi-beam forming, imaging, and transient source positioning. It is suitable for the detection of rapid events such as transient celestial body monitoring, radio bursts, meteor tails, and ionospheric disturbances, and has the advantages of strong real-time performance, high versatility, and no need to modify the array front-end hardware. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating a transient source localization method based on multi-beam imaging provided in this application embodiment; Figure 2 A flowchart illustrating a transient source localization method based on multi-beam imaging provided in this application embodiment; Figure 3 A flowchart illustrating a transient source localization method based on multibeam imaging provided in this application; Figure 4 A schematic diagram of a transient source localization device based on multi-beam imaging provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application. Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by those skilled in the art. The words "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are only used to distinguish different components. The words "comprising" or "including," etc., mean that the element or object preceding the word covers the element or object listed after the word and its equivalents, but do not exclude other elements or objects. The words "connected," "coupled," or "connected," etc., are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up," "down," "left," "right," etc., are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0014] The transient source localization method based on multi-beam imaging provided in this application can be applied to electronic devices, GPUs in electronic devices, terminal devices, transient source localization devices or apparatuses based on multi-beam imaging, or other apparatuses or apparatuses capable of executing this embodiment, without limitation. Optionally, GPUs are the necessary hardware foundation for realizing parallel synthesis of over 100 beams and millisecond-level fast imaging. The combination with "tensor computation" specifies the concrete way to utilize GPUs for computation, which is the standard mode for modern deep learning frameworks (such as PyTorch and TensorFlow) to leverage GPU performance. This embodiment describes the execution entity as the GPU. Optionally, this application is particularly applicable to real-time or near-real-time detection and localization of fast radio transient sources.

[0015] The terminal can be a user equipment (UE) such as a mobile phone, smartphone, laptop computer, digital broadcast receiver, personal digital assistant (PDA), or tablet computer (PAD), handheld device, in-vehicle device, wearable device, computing device, or other processing device connected to a wireless modem, mobile station (MS), or mobile terminal. This terminal has the ability to communicate with one or more core networks via a radio access network (RAN).

[0016] The preferred embodiments of this application are described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit this application. Furthermore, the embodiments and features in the embodiments of this application can be combined with each other without conflict.

[0017] Figure 1 This is a flowchart illustrating a transient source localization method based on multi-beam imaging, provided as an embodiment of this application. Figure 1 As shown, the method may include: Step S101: Obtain voltage data collected by each receiving station of the radio array.

[0018] For example, voltage data collected by each receiving station of the radio array is acquired. For instance, the voltage data is complex voltage data.

[0019] Step S102: Construct a beam weight tensor based on multiple preset target directions and the geometry of the radio array; perform complex tensor multiplication on the voltage data and the beam weight tensor, and synthesize digital beams that correspond one-to-one with all target directions in parallel.

[0020] For example, the 'digital beam' in this invention refers to a beam with a specific spatial orientation formed through digital signal processing, with each beam corresponding to a preset target direction. The preset target directions are flexible pre-inputs, and their specific sources can be varied. For example, the target directions may come from astronomical observation plans, all-sky uniform grid scanning, or prompts from other auxiliary equipment (such as alarms from optical telescopes). For instance, the preset target directions can be determined according to actual observation needs, and the acquisition methods include, but are not limited to: Scientific observation plan: One or more fixed directions pre-defined according to specific astronomical observation needs (such as monitoring specific celestial objects or sky regions). All-sky gridded scan: Generating a uniformly oriented grid with a certain angular resolution (e.g., 1 degree) for the entire visible sky covering the array. External event trigger: Responding to the initial coordinates of transient sources provided by other observational instruments (such as optical telescopes or high-energy satellites), setting those coordinates and their surrounding area as the target direction. Data-driven discovery: Identifying regions of interest through preliminary analysis of historical data or low-frequency wide-field-of-view data.

[0021] In this step, beam weight vectors are calculated based on multiple preset target directions and the geometry of the radio array. These beam weight vectors are then combined into a beam weight tensor. This beam weight tensor and complex voltage data are input into a deep learning framework and executed on the GPU as complex tensor matrix multiplication. This allows for the simultaneous formation of multiple digital beams in a single scheduling operation, significantly improving processing efficiency.

[0022] Therefore, by achieving "full grid coverage" or "full target set coverage" through "single scheduling", the entire set of preset beams can be output at once, realizing complete parallelization without leaving any beams that need to be processed serially, thus ensuring positioning efficiency.

[0023] Step S103: Obtain the beam intensity sequence based on the digital beam and map the beam intensity sequence to generate a two-dimensional image; perform detection processing on the two-dimensional image to identify and locate transient sources.

[0024] For example, a short-time integral is performed on the digital beam to obtain a beam intensity sequence with a preset time resolution. For instance, the preset time resolution can be on the millisecond scale, and there is no limitation on the preset time resolution. Then, a pre-constructed spatial back-projection operator is used to map the multi-beam intensities in the beam intensity sequence to a two-dimensional image; where the two-dimensional image can be a two-dimensional sky image. This back-projection operator is determined by the geometry of the radio array and the target orientation, and is executed in batches in a tensor-based manner on the GPU, thereby quickly converting the multi-beam results into a two-dimensional image. Finally, threshold detection or statistical significance analysis is directly performed on the two-dimensional image to identify and locate radio transient sources that suddenly increase in brightness or appear abruptly. These transient sources include radio transient sources, etc.

[0025] Therefore, the entire process, from beamforming, integration, imaging to saliency detection, is completed within the GPU, avoiding data backhaul delays and large amounts of data transfer between the GPU and CPU. This enables end-to-end millisecond-level processing, meets the real-time triggering requirements for fast transient events, improves the triggering efficiency of transient sources, and further enhances the response speed.

[0026] The method provided in this application acquires voltage data collected by each receiving station of a radio array. Based on multiple preset target directions and the geometry of the radio array, a beam weight tensor is constructed. Complex tensor multiplication is performed on the voltage data and the beam weight tensor to synthesize digital beams corresponding one-to-one with all target directions in parallel. Based on the digital beams, a beam intensity sequence is obtained, and the beam intensity sequence is mapped to generate a two-dimensional image. The two-dimensional image is then processed to identify and locate transient sources. In this scheme, multi-station voltage data collected by the radio array is used as input to construct a beam weight tensor, and the multi-beamforming process is transformed into tensor complex matrix multiplication, thereby achieving simultaneous synthesis of more than one hundred beams. Subsequently, the beam intensity sequence is acquired, and a two-dimensional image is directly generated using the beam intensity sequence. By detecting the two-dimensional image, transient source location can be completed in milliseconds. Therefore, by using parallel tensor computation, the synthesis computation of over 100 beams is changed from a serial loop to a single parallel scheduling, which significantly improves the efficiency of multi-beam forming, imaging, and transient source positioning. It is suitable for the detection of rapid events such as transient celestial body monitoring, radio bursts, meteor tails, and ionospheric disturbances, and has the advantages of strong real-time performance, high versatility, and no need to modify the array front-end hardware.

[0027] Figure 2 A flowchart illustrating a transient source localization method based on multi-beam imaging provided in this application is shown below. Figure 2 As shown, in this embodiment... Figure 1 Based on the embodiments, the method is described in detail below, and the method includes: Step S201: Obtain voltage data collected by each receiving station of the radio array.

[0028] In one example, after step S201, the method further includes: performing time alignment processing on the voltage data to form complex voltage data in tensor form by time slices.

[0029] For example, Figure 3 A flowchart illustrating a transient source localization method based on multi-beam imaging provided in this application is shown below. Figure 3 As shown, voltage data collected by each receiving station of the radio array is first acquired. Then, the raw voltage data collected by each receiving station of the radio array is uniformly time-aligned, and the voltage data is constructed into a "tensor" data structure suitable for parallel computing according to time slices.

[0030] Step S202: Construct the beam weight tensor based on the preset multiple target directions and the geometry of the radio array.

[0031] For example, such as Figure 3As shown, based on multiple preset target directions and the geometry of the radio array, the beam weight vector corresponding to each target direction is calculated. The beam weight vectors (including amplitude and phase compensation) corresponding to all target directions are combined to construct a beam weight tensor, i.e., a beam weight matrix. The beam weight tensor is a beam weight matrix that includes multiple beam weight vectors, each beam weight vector corresponding to a single target direction. The specific internal arrangement of the beam weight tensor (column-first or row-first) is not limited.

[0032] Step S203: Perform complex tensor multiplication on the voltage data and beam weight tensor, and synthesize digital beams that correspond one-to-one with all target directions in parallel.

[0033] In one example, S203 includes: calling the batch matrix multiplication function in the deep learning framework to perform complex tensor multiplication on the voltage data and the beam weight tensor, and synthesizing digital beams that correspond one-to-one with all target directions in parallel.

[0034] For example, such as Figure 3 As shown, in specific implementation, the complex tensor multiplication operation is achieved as follows: Complex voltage data organized by time slices is constructed as an input tensor, with dimensions including time, receiving station, and frequency channel. On a graphics processing unit (GPU), the input tensor and beam weight tensor are multiplied using batch matrix multiplication functions in deep learning frameworks (such as PyTorch and TensorFlow) to synthesize digital beams corresponding one-to-one with all target directions in parallel.

[0035] This operation is encapsulated as a single parallel computing kernel within the GPU. Leveraging the GPU's many-core architecture, it performs synchronous calculations on different target directions, time slices, and frequency channels, thus completing the synthesis of all beams in a single scheduling operation and directly outputting a single output tensor containing beam signals from all target directions. This process completely avoids the traditional iterative calculation mode for each direction, achieving a paradigm shift from "vector-by-vector computation" to "global tensor transformation," which is crucial for achieving millisecond-level synthesis of hundreds or more beams.

[0036] Step S204: Obtain the beam intensity sequence based on the digital beam.

[0037] In one example, S204 includes: performing short-time integration on the digital beam to obtain a beam intensity sequence with a preset time resolution.

[0038] For example, such as Figure 3 As shown, a short-time integral is performed on the digital beam to obtain a beam intensity sequence with a preset time resolution. For example, the preset time resolution can be on the millisecond scale, and there is no limitation on the preset time resolution.

[0039] Step S205: Map the beam intensity sequence to generate a two-dimensional image.

[0040] In one example, S205 includes: mapping the beam intensity sequence to a two-dimensional image using a pre-built spatial backprojection operator; wherein the spatial backprojection operator is determined by the target orientation and the geometry of the radio array.

[0041] For example, such as Figure 3 As shown, based on the target direction and the geometry of the radio array, a spatial backprojection operator is calculated. Then, using tensor operations, the beam intensity sequence is directly and rapidly mapped into a two-dimensional image (i.e., ...). Figure 3 (A two-dimensional fast image in the image), which directly reflects the distribution of sky brightness.

[0042] Therefore, it can bypass the complex interferometric imaging process and achieve a fast and direct conversion from beam to image, which greatly improves the imaging speed.

[0043] Step S206: Perform detection processing on the two-dimensional image to identify and locate the transient source.

[0044] In one example, S206 includes: performing threshold detection or statistical significance analysis on a two-dimensional image to identify and locate radio transient sources.

[0045] For example, such as Figure 3 As shown, threshold detection or statistical significance analysis is performed on the two-dimensional image to identify and locate radio transient sources that suddenly increase in brightness or appear abruptly.

[0046] Step S207: Based on the identified and located radio transient sources, record the event information of the radio transient sources and generate an alarm.

[0047] For example, such as Figure 3 As shown, once a possible transient source (such as a fast radio burst) is identified in a two-dimensional image through "saliency detection", the event information (such as time, location, intensity, etc.) will be automatically recorded, and an alarm will be generated (such as issuing a notification, storing it in the database, triggering other devices, etc.) to promptly remind relevant personnel.

[0048] The method provided in this application acquires voltage data collected by each receiving station of a radio array. Based on multiple preset target directions and the geometry of the radio array, a beam weight tensor is constructed. Complex tensor multiplication is performed on the voltage data and the beam weight tensor to synthesize digital beams corresponding one-to-one with all target directions in parallel. A beam intensity sequence is obtained from the digital beams. The beam intensity sequence is mapped to generate a two-dimensional image. Detection processing is performed on the two-dimensional image to identify and locate transient sources. Based on the identified and located radio transient sources, event information of the radio transient sources is recorded, and alarms are generated. Therefore, by unifying the array response matrix, beamforming process, short-time integration, and spatial back-projection process into tensor operations, and utilizing the parallel computing capabilities of the GPU deep learning framework, simultaneous synthesis of over 100 beams and millisecond-level imaging are achieved. In this method, multi-beamforming is directly accomplished through a single tensor matrix multiplication, avoiding repetitive cyclic calculations. The imaging process is batch-executed in the GPU using back-projection operators based on array geometry, thereby achieving rapid two-dimensional image generation. Furthermore, transient source triggering can be achieved through saliency detection on the GPU side. The solution provided by this invention can significantly improve processing efficiency, enabling real-time or near-real-time detection of fast events without changing the array hardware system, and has good engineering application value. Moreover, the method relies on tensor quantization operations, exhibiting good scalability and the ability to achieve higher performance in future larger arrays.

[0049] Corresponding to the above method, embodiments of this application also provide a transient source localization device based on multi-beam imaging, such as... Figure 4 As shown, the device includes: Acquisition module 41 is used to acquire voltage data collected by each receiving station of the radio array; Tensor module 42 is used to construct a beam weight tensor based on multiple preset target directions and the geometry of the radio array; perform complex tensor multiplication operations on the voltage data and the beam weight tensor, and synthesize digital beams that correspond one-to-one with all the target directions in parallel. The identification module 43 is used to obtain a beam intensity sequence based on the digital beam, and map the beam intensity sequence to generate a two-dimensional image; to perform detection processing on the two-dimensional image to identify and locate the radio transient source.

[0050] The functions of each functional unit in the transient source localization device based on multi-beam imaging provided in the above embodiments of this application can be implemented through the above methods and steps. Therefore, the specific working process and beneficial effects of each unit in the transient source localization device based on multi-beam imaging provided in the embodiments of this application will not be repeated here.

[0051] This application also provides an electronic device, such as... Figure 5As shown, it includes a processor 510, a communication interface 520, a memory 530, and a communication bus 540, wherein the processor 510, the communication interface 520, and the memory 530 communicate with each other through the communication bus 540.

[0052] Memory 530 is used to store computer programs; The processor 510 performs the above steps when executing the program stored in the memory 530.

[0053] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0054] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0055] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0056] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0057] The implementation methods and beneficial effects of the various components of the electronic device in the above embodiments for solving the problem can be found in [reference needed]. Figure 1 The steps in the illustrated embodiments are used to implement the electronic device. Therefore, the specific working process and beneficial effects of the electronic device provided in this application will not be repeated here.

[0058] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores instructions that, when executed on a computer, cause the computer to perform any of the transient source localization methods based on multi-beam imaging described in the above embodiments.

[0059] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the transient source localization methods based on multi-beam imaging described in the above embodiments.

[0060] Those skilled in the art will understand that the embodiments in this application can be provided as methods, systems, or computer program products. Therefore, the embodiments in this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the embodiments in this application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This application describes embodiments of methods, apparatus (systems), and computer program products according to embodiments of this application with reference to flowchart illustrations and / or block diagrams. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0064] Although preferred embodiments have been described in this application, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of this application.

[0065] Obviously, those skilled in the art can make various modifications and variations to the embodiments of this application without departing from the spirit and scope of the embodiments of this application. Therefore, if these modifications and variations to the embodiments of this application fall within the scope of the claims in this application and their equivalents, then this application also intends to include these modifications and variations.

Claims

1. A transient source localization method based on multi-beam imaging, characterized in that, include: Acquire voltage data collected by each receiving station of the radio array; Based on multiple preset target directions and the geometry of the radio array, a beam weight tensor is constructed; The voltage data and the beam weight tensor are multiplied by a complex tensor to synthesize digital beams that correspond one-to-one with all the target directions in parallel. Based on the digital beam, a beam intensity sequence is obtained, and the beam intensity sequence is mapped to generate a two-dimensional image; the two-dimensional image is then processed to identify and locate transient sources.

2. The method as described in claim 1, characterized in that, Performing complex tensor multiplication operations on the voltage data and the beam weight tensor, and synthesizing digital beams corresponding one-to-one with all the target directions in parallel, includes: The batch matrix multiplication function in the deep learning framework is invoked to perform complex tensor multiplication on the voltage data and the beam weight tensor, and to synthesize digital beams that correspond one-to-one with all the target directions in parallel.

3. The method as described in claim 1, characterized in that, The detection and processing of the two-dimensional image to identify and locate the radio transient source includes: Threshold detection or statistical significance analysis is performed on the two-dimensional image to identify and locate the radio transient source.

4. The method as described in claim 1, characterized in that, Based on the digital beam, a beam intensity sequence is obtained, including: The digital beam is subjected to short-time integration to obtain a beam intensity sequence with a preset time resolution.

5. The method as described in claim 1, characterized in that, The process of mapping the beam intensity sequence to generate a two-dimensional image includes: A pre-constructed spatial back-projection operator is used to map the beam intensity sequence to a two-dimensional image; wherein the spatial back-projection operator is determined by the target direction and the geometry of the radio array.

6. The method according to any one of claims 1-5, characterized in that, After acquiring the voltage data collected by each receiving station of the radio array, the method further includes: The voltage data is time-aligned and composed into complex voltage data in tensor form according to time slices.

7. The method according to any one of claims 1-5, characterized in that, The method further includes: Based on the identified and located radio transient sources, the event information of the radio transient sources is recorded, and an alarm is generated.

8. A transient source localization device based on multi-beam imaging, characterized in that, include: The acquisition module is used to acquire voltage data collected by each receiving station of the radio array; The tensor module is used to construct a beam weight tensor based on multiple preset target directions and the geometry of the radio array; perform complex tensor multiplication operations on the voltage data and the beam weight tensor, and synthesize digital beams that correspond one-to-one with all the target directions in parallel. The identification module is used to obtain a beam intensity sequence based on the digital beam, and map the beam intensity sequence to generate a two-dimensional image; to perform detection processing on the two-dimensional image to identify and locate the radio transient source.

9. An electronic device, characterized in that, The electronic device includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images