Hierarchical spectrum monitoring data processing method and device based on cloud side end

Abstract

The invention discloses a hierarchical spectrum monitoring data processing method and device based on a cloud side end, and belongs to the technical field of electromagnetic spectrum management and control, and the method comprises the steps: S1, task issuing: distributing tasks to levels and nodes; s2, data processing: configuring different data processing algorithms for the end, the edge and the cloud to perform multi-level distributed calculation; after a task is received, after monitoring data collected by the multi-station end node is subjected to distributed end-level processing, fused data and key conclusion data are transmitted to corresponding edges, and then a plurality of edge-level distributed key processing results are transmitted to a corresponding cloud for analysis; the cloud node analyzes and processes spectrum data by using computing resources and algorithms, the edge nodes and the end nodes undertake partial computing and share data processing results at the same level, and the plurality of end nodes share the data processing results at the same level in a small range and autonomously and cooperatively process the same spectrum monitoring task; and S3, issuing a result. According to the invention, the overall utilization efficiency of resources, the complex spectrum environment adaptability and the processing capability are improved.

Description

Technical Field

[0001] The present invention relates to the field of electromagnetic spectrum management and control technology, and more specifically, to a cloud-edge-based hierarchical spectrum monitoring data processing method and device. Background Art

[0002] Although traditional spectrum monitoring methods have transitioned from individual monitoring to grid-based monitoring using multiple geographically distributed sites, the monitoring data processing architecture and methods still have limitations. The multi-site collaborative processing capabilities are poor, making it difficult to efficiently and accurately monitor large-scale spectrum activity in real time. These shortcomings are primarily reflected in the following two aspects: (1) The existing hierarchical spectrum monitoring data fusion processing methods are mostly centralized. The monitoring center node needs to collect and process the original data of each sub-node. The system transmission efficiency is low and the real-time performance is low.

[0003] (2) When faced with a complex and changing spectrum environment, the existing centralized method has limited fusion node processing resources and lacks flexibility in computing resources and algorithm configuration, making it difficult to effectively utilize resources to achieve rapid adaptation and accurate processing of large-scale, highly complex monitoring data. Summary of the Invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a hierarchical spectrum monitoring data processing method and device based on cloud-edge-end, so as to complete large-scale real-time multi-station collaborative spectrum monitoring data processing in real time and efficiently. The method realizes distributed computing based on the cloud-edge-end architecture, reduces the large amount of original data transmission between the cloud and the edge and the end by sinking the processing capability, optimizes the use of network bandwidth, reduces the delay of data in the transmission process, and improves the real-time monitoring and response capability of the system; realizes flexible configuration and management of computing resources through multi-level and multi-node distributed computing, allocates them to appropriate levels and nodes according to the nature of the task, flexibly configures and adjusts the processing algorithm of different levels and nodes, and improves the overall resource utilization efficiency, adaptability to complex spectrum environments and processing capabilities.

[0005] The object of the present invention is achieved through the following solutions: A cloud-edge-device-based hierarchical spectrum monitoring data processing method includes: S1, Task Distribution: Assign tasks to layers and nodes. Cloud and edge nodes serve as the initial task distribution points based on user needs. Cloud-level and edge-level spectrum monitoring tasks are distributed hierarchically from cloud to edge and then to end, or from edge to end. S2, Data Processing: Different data processing algorithms are configured on the end, edge, and cloud to perform multi-level distributed computing. After receiving a task, the monitoring data collected by multiple end nodes undergoes distributed end-level processing, and the fused data and key conclusion data are transmitted to the corresponding edge. Multiple edge-level distributed key processing results are then transmitted to the corresponding cloud for analysis. The cloud node uses computing resources and algorithms to analyze and process the spectrum data, while the edge and end nodes undertake part of the computation and share the data processing results at the same level. Multiple end nodes share data at the same level within a small range and autonomously and collaboratively process the same spectrum monitoring task. S3, result distribution: The processing results of cloud-level and edge-level tasks are distributed step by step from cloud to edge and then to end, or from edge to end.

[0006] Furthermore, in step S1, tasks are assigned to levels and nodes, and cloud and edge nodes serve as initial task issuance points to issue tasks based on user needs. Cloud-level and edge-level spectrum monitoring tasks are issued hierarchically from cloud to edge and then to end, or from edge to end, including the following sub-steps: Step S11, cloud-level task issuance: start the cloud-level spectrum monitoring task at the cloud node, and issue the task to the corresponding edge node, which is then issued by the edge node to the corresponding end node, triggering the data processing process; Step S12, edge-level task issuance: start the edge-level spectrum monitoring task at the edge node, and issue the task to the corresponding end node, triggering the data processing flow.

[0007] Furthermore, in step S2, the end, edge, and cloud are configured with different data processing algorithms to perform multi-level distributed computing. After receiving the task, the monitoring data collected by the multi-station end nodes is processed at the distributed end level, and the fused data and key conclusion data are transmitted to the corresponding edge. Then, multiple edge-level distributed key processing results are transmitted to the corresponding cloud for analysis. The cloud node uses computing resources and algorithms to analyze and process the spectrum data, and the edge and end nodes undertake part of the calculation and share the data processing results at the same level. Multiple end nodes share at the same level within a small range and autonomously and collaboratively process the same spectrum monitoring task, including the following sub-steps: Step S21, end-level data collection: After receiving the task, the end node calls the spectrum monitoring device and continuously collects the specified spectrum data of the specified frequency band within the specified time according to the virtual interface of the standard protocol, and uniformly stores and manages the spectrum monitoring data; Step S22, end-level data processing: Perform end-level comprehensive processing on the collected data to eliminate the negative impact of signal fluctuations. At the same time, use algorithms that can respond in a timely manner to perform data analysis to support subsequent processing of edge-level tasks; Step S23, end-level collaborative processing: Under set conditions, multiple end nodes within a small area are grouped together to share end-level processing results and collaboratively complete spectrum monitoring tasks. Step S24, end-level result upload: Based on the region where the end node is located, the end-level data distributed processing results of the cloud and edge-level tasks and the autonomous collaborative processing results of the end node are reported to the edge node in the corresponding region; Step S25, edge-level data processing: The edge node receives and comprehensively processes the distributed end-level processing results reported by end nodes in multiple different regions, and stores them in the database to update and manage the processing results. At the same time, a cross-region processing algorithm is configured to perform data analysis to complete the edge-level task and support the subsequent processing of the cloud-level task. Step S26, edge-level result upload: reporting the edge-level distributed processing results of the cloud-level task to the cloud node; Step S27, edge-level result sharing: sharing edge-level processing results among edge nodes after obtaining permission; Step S28, cloud-level data processing: The cloud node receives and comprehensively processes the edge-level processing results reported by edge nodes in multiple different regions, and stores them in a database to update and manage the monitoring data processing results. It also uses algorithms that rely on a large amount of historical data to conduct in-depth analysis and processing of complex spectrum data to complete cloud-level tasks. Step S29, cloud-level result saving: saving the signal recognition results and signal location results within the cloud node area.

[0008] Furthermore, in step S21, the designated spectrum data of the designated frequency band includes background noise and occupancy.

[0009] Furthermore, in step S22, the end-level comprehensive processing includes an averaging method and a merging method.

[0010] Furthermore, the use of an algorithm capable of timely response specifically includes using a trained neural network model to identify the signal and save the result.

[0011] Furthermore, in step S25, the cross-region processing algorithm specifically includes a target tracking algorithm.

[0012] Furthermore, in step S28, the algorithm that relies on a large amount of historical data specifically includes a computing resource and occupancy prediction algorithm.

[0013] Furthermore, in step S3, the processing results of the cloud-level and edge-level tasks are distributed step by step from the cloud to the edge and then to the end, or from the edge to the end, including the following sub-steps: Step S31, cloud-level result distribution: the cloud-level task processing results stored in the cloud node are distributed to each corresponding edge node, and then distributed to each corresponding end node via the edge node; Step S32, edge-level result distribution: the edge-level task processing result stored in the edge node is distributed to each corresponding end node.

[0014] A cloud-edge-based hierarchical spectrum monitoring data processing device includes a processor and a memory, wherein a computer program is stored in the memory. When the computer program is loaded by the processor, it executes the method described in any one of the above items.

[0015] The beneficial effects of the present invention include: (1) By sinking the processing power to the edge and end, only the key transmission processing is uploaded step by step, reducing the large amount of original data transmission between the cloud and the edge and end, optimizing the use of network bandwidth, reducing the delay of data during transmission, and improving the real-time monitoring and response capabilities of the system.

[0016] (2) Flexible configuration and management of computing resources are achieved through multi-level and multi-node distributed computing. Tasks are assigned to appropriate levels and nodes according to their nature. Different levels and nodes are flexibly configured and the processing algorithms are adjusted to improve the overall resource utilization efficiency, adaptability to large-scale complex spectrum environments, and processing capabilities.

[0017] Based on the above two aspects, the method of the present invention can efficiently and accurately complete large-scale real-time multi-station collaborative spectrum monitoring data processing, helping users to grasp the spectrum activity in a large range, thereby supporting spectrum management and control related tasks such as long-distance and cross-regional signal activity analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 This is a schematic diagram of hierarchical spectrum monitoring data processing based on cloud-edge-device architecture; Figure 2 This is a flow chart of a hierarchical spectrum monitoring data processing method based on a cloud-edge-device architecture according to an embodiment of the present invention; Figure 3 Flowchart for signal type recognition based on neural network model. DETAILED DESCRIPTION

[0020] All features disclosed in all embodiments in this specification, or steps in all methods or processes implicitly disclosed, except for mutually exclusive features and / or steps, can be combined and / or expanded or replaced in any manner.

[0021] In a specific embodiment, Figure 1As shown, the cloud-edge-end architecture in the present invention is composed of three levels: cloud, edge, and end. Based on this architecture, the three cross-level processes of task assignment, data processing, and result distribution are completed in sequence, thereby realizing hierarchical spectrum monitoring data processing. In the inventive concept, on the one hand, the spectrum data is preliminarily processed and screened by the end and edge, and only the fused data and key conclusive data are uploaded to the cloud, reducing the data transmission volume and transmission pressure, and effectively avoiding the problem of limited data transmission. At the same time, the real-time processing capabilities of the end and edge can respond to some needs in a timely manner, improving the timeliness of processing. On the other hand, the cloud can use powerful computing resources and advanced algorithms to conduct in-depth analysis and processing of complex spectrum data. By configuring different processing algorithms on the end and edge in a targeted manner, it can perceive the changes in the spectrum environment in real time and provide timely feedback to the cloud, thereby achieving better adaptation and processing of the complex spectrum environment, while improving the overall resource utilization efficiency.

[0022] Specifically, a cloud-edge-based hierarchical spectrum monitoring data processing method is provided, comprising the following steps: Step S1, task issuance: Assign the task to the appropriate level and node based on its nature. Both cloud and edge nodes can serve as the initial task issuance points to issue tasks based on user needs. Cloud-level and edge-level spectrum monitoring tasks are issued hierarchically from cloud to edge and then to end, or from edge to end, to achieve flexible configuration and management of computing resources.

[0023] Step S2, Data Processing: The end, edge, and cloud flexibly configure different data processing algorithms to perform multi-level distributed computing. After receiving the task, the monitoring data collected by multiple end nodes undergoes distributed end-level processing, and the fused data and key conclusion data are transmitted to the corresponding edge. The multiple edge-level distributed key processing results are then transmitted to the corresponding cloud for in-depth analysis, effectively reducing the amount of original data transmission and transmission pressure, and improving processing timeliness. Cloud nodes use powerful computing resources and advanced algorithms to conduct in-depth analysis and processing of complex spectrum data. Edge and end nodes can undertake partial computing and share data comprehensive processing results at the same level. Multiple end nodes can also share at the same level within a small range and autonomously and collaboratively process the same spectrum monitoring task, improving overall resource utilization efficiency, adaptability to complex spectrum environments, and the system's real-time processing capabilities.

[0024] Step S3: Result distribution: The processing results of cloud-level and edge-level tasks are distributed step by step from cloud to edge and then to end, or from edge to end.

[0025] More specifically, step S1 includes the following sub-steps: Step S11, cloud-level task issuance: start the cloud-level spectrum monitoring task at the cloud node, and issue the task to the corresponding edge node, which is then issued by the edge node to the corresponding end node, triggering the data processing process; Step S12, edge-level task issuance: start the edge-level spectrum monitoring task at the edge node, and issue the task to the corresponding end node, triggering the data processing flow.

[0026] More specifically, step S2 includes the following sub-steps: Step S21, end-level data collection: After receiving the task, the end node calls the spectrum monitoring device and continuously collects the specified spectrum data of the specified frequency band within the specified time according to the virtual interface of the standard protocol, such as background noise and occupancy, and uniformly stores and manages the spectrum monitoring data; Step S22, end-level data processing: Perform end-level comprehensive processing on the collected data, including methods such as averaging and merging to eliminate the negative impact of signal fluctuations. At the same time, configure signal recognition and other algorithms that can respond in a timely manner to perform data analysis to support subsequent processing of edge-level tasks; Step S23, end-level collaborative processing: Under certain conditions, multiple end nodes within a small area can be grouped together to share end-level processing results and collaboratively complete spectrum monitoring tasks. Step S24, end-level result upload: Based on the region where the end node is located, the end-level data distributed processing results of the cloud / edge-level tasks and the autonomous collaborative processing results of the end node are reported to the edge node in the corresponding region; Step S25, edge-level data processing: The edge node receives and comprehensively processes the distributed end-level processing results reported by end nodes in multiple different regions, and stores them in the database to update and manage the processing results. At the same time, cross-region processing algorithms such as target tracking are configured to perform data analysis to complete edge-level tasks and support subsequent processing of cloud-level tasks. Step S26, edge-level result upload: reporting the edge-level distributed processing results of the cloud-level task to the cloud node; Step S27, edge-level result sharing: sharing edge-level processing results among edge nodes after obtaining permission; Step S28, cloud-level data processing: The cloud node receives and comprehensively processes the edge-level processing results reported by edge nodes in multiple different regions, and stores them in the database to update and manage the monitoring data processing results. It also uses powerful computing resources and occupancy prediction algorithms that rely on a large amount of historical data to conduct in-depth analysis and processing of complex spectrum data to complete cloud-level tasks.

[0027] Step S29, cloud-level result saving: all signal recognition results and signal location results within the cloud node area are saved.

[0028] More specifically, step S3 includes the following sub-steps: Step S31, cloud-level result distribution: the cloud-level task processing results stored in the cloud node are distributed to each corresponding edge node, and then distributed to each corresponding end node via the edge node; Step S32, edge-level result distribution: the edge-level task processing result stored in the edge node is distributed to each corresponding end node.

[0029] Example 1 In other embodiments, a hierarchical spectrum monitoring data processing method based on cloud-edge is provided, such as Figure 2 In this embodiment, a cloud-level spectrum monitoring data processing task and an edge-level spectrum monitoring data processing task are simultaneously initiated, and an inter-end node collaborative data processing task is triggered. The following steps specifically illustrate how this embodiment method, based on the three cross-level processes of task assignment, data processing, and result delivery, simultaneously completes multiple spectrum monitoring data processing tasks.

[0030] Step S1, task issuance: Assign tasks to layers and nodes. Cloud and edge nodes, acting as initial task issuance points based on user needs, issue tasks hierarchically, from cloud to edge and then to end, or vice versa. In this embodiment, cloud nodes issue cloud-level tasks to acquire all monitoring signal types within their coverage area, step by step from cloud to edge and then to end. Simultaneously, edge nodes issue edge-level tasks to their corresponding end nodes to acquire all spectrum data within their coverage area. Step S2, data processing: The end, edge, and cloud configure different data processing algorithms to perform multi-level distributed computing; after receiving the task, the monitoring data collected by the multi-station end nodes undergoes distributed end-level processing, and the fused data and key conclusion data are transmitted to the corresponding edge, and then the multiple edge-level distributed key processing results are transmitted to the corresponding cloud for analysis; the cloud node uses computing resources and algorithms to analyze and process the spectrum data, and the edge and end nodes undertake part of the calculation and share the data processing results at the same level. Multiple end nodes share at the same level within a small range and autonomously and collaboratively process the same spectrum monitoring task. In this embodiment, after the end node receives the cloud and edge-level tasks, it processes, uploads, and shares the two task-related monitoring data collected by the end node step by step to complete the edge-level and cloud-level tasks. In addition, after multiple end nodes within a small area meet the trigger conditions, they automatically transmit the end-level processing results to each other to collaboratively complete the signal positioning task, and upload the signal positioning results to the corresponding edge and cloud nodes step by step.

[0031] Step S3, result distribution: the processing results of cloud and edge level tasks are distributed step by step from cloud to edge and then to end, or from edge to end. In this embodiment, the comprehensive processing results of all monitoring signal types within the coverage area of ​​cloud node and edge node are distributed. The comprehensive processing results of all spectrum data within the range are sent down step by step from cloud to edge to end, and from edge to end.

[0032] Example 2 On the basis of Example 1, step S1 includes the following sub-steps: Step S11, Cloud-Level Task Issuance: A cloud-level spectrum monitoring task is initiated at the cloud node and issued to the corresponding edge node, which then issues the task to the corresponding end node, triggering the data processing flow. In this embodiment, the cloud node initiates a cloud-level task to acquire all monitoring signal types within the cloud node's coverage area and issues a task to each corresponding edge node to acquire all monitoring signal types within the edge node's coverage area. The edge node then issues a task to each corresponding end node to acquire all signals within the end node's coverage area and identify their types, thereby triggering the data processing flow.

[0033] Step S12: edge-level task issuance: start the edge-level spectrum monitoring task at the edge node and issue the task to the corresponding end node, triggering the data processing process. Start the edge-level task of obtaining all spectrum data within the coverage area of ​​the edge node, and issue the task of obtaining all spectrum data within the range of the end node to the corresponding end node, thereby triggering the data processing flow.

[0034] Example 3 On the basis of Example 1, step S2 includes the following sub-steps: Step S21: End Node Data Collection: After receiving the task, the end node invokes the spectrum monitoring device and, using a virtual interface using a standard protocol, continuously collects spectrum data for a specified frequency band within a specified timeframe. The spectrum monitoring data is then centrally stored and managed. In this embodiment, the end node receives the task using the monitoring device to collect various spectrum monitoring data in real time, stores it in memory for data accumulation, and centrally stores and manages it using standardized interfaces, protocols, and virtualization technology.

[0035] Step S22, End-Level Data Processing: The collected data undergoes end-level comprehensive processing to eliminate the negative impact of signal fluctuations. Data analysis is performed using timely responsive algorithms to support subsequent edge-level tasks. In this embodiment, signal types are identified based on the signal data in memory using methods such as neural networks. The results are accumulated in memory and fused when the accumulated data collected by the monitoring device reaches a threshold.

[0036] Step S23, End-Level Collaborative Processing: Under predefined conditions, multiple end nodes within a small area are grouped together, sharing end-level processing results and collaboratively completing the spectrum monitoring task. In this embodiment, after obtaining appropriate permissions and meeting certain triggering conditions, multiple end nodes within the small area automatically exchange end-level processed data and collaboratively complete the positioning task.

[0037] Step S24: End-level result storage and upload: Based on the end node's region, the distributed processing results of end-level data for cloud- and edge-level tasks, as well as the autonomous collaborative processing results of the end node, are reported to the edge node in the corresponding region. In this embodiment, all identification results, spectrum data, and positioning information within the range of each end node are stored and reported to the corresponding edge node.

[0038] Step S25, Edge-Level Data Processing: The edge node receives and comprehensively processes the distributed end-level processing results reported by end nodes in multiple different regions, storing them in a database for update management. It also deploys a cross-regional processing algorithm for data analysis to complete edge-level tasks and support subsequent processing of cloud-level tasks. In this embodiment, the signal recognition results, spectrum data, and signal locations reported by multiple end nodes are acquired, stored, and further fused. After fusion, the edge-level task initiated by the edge node is completed.

[0039] Step S26, edge-level result saving and uploading: reporting the edge-level distributed processing results of the cloud-level task to the cloud node. In this embodiment, all signal recognition results and signal locations within the area of ​​each edge node are saved and reported to the corresponding cloud node.

[0040] Step S27, edge-level result sharing: sharing edge-level processing results between edge nodes after obtaining the corresponding authority. In this embodiment, after obtaining the corresponding authority, edge nodes transmit edge-level processing results to each other.

[0041] Step S28, Cloud-Level Data Processing: The cloud node receives and comprehensively processes edge-level processing results reported by edge nodes in multiple different regions, storing them in a database to update and manage the monitoring data processing results. It also uses algorithms that rely on extensive historical data to conduct in-depth analysis and processing of complex spectrum data to complete cloud-level tasks. In this embodiment, edge-level signal identification results and signal location data reported by multiple edge nodes are acquired, stored, and fused. After fusion, the cloud-level task initiated by the cloud node is completed.

[0042] Step S29, cloud-level result saving: all signal recognition results and signal location results within the cloud node area are saved. In this embodiment, the two cloud-level processing result lists are integrated and saved as a cloud signal history fusion table in the database, thereby completing the cloud-level task.

[0043] In this embodiment, as a further optional implementation, step S22 includes the following sub-steps: Step S221, signal recognition: Figure 3As shown in the figure, a neural network is pre-trained using historical spectrum monitoring data sets and then deployed on each end node. During real-time monitoring, after pre-processing the data stored in the memory, the trained network model is used to identify the signal type and the results are stored in the memory. Step S222, recognition result fusion: when the accumulated time of the signal recognition results of each signal is greater than a threshold, the data is fused to form a comprehensive recognition result; Step S223, spectrum data fusion: when the accumulated time length of the spectrum data collected by the monitoring device is greater than a threshold, the spectrum data is fused.

[0044] In step S23, the sub-steps are included: Step S231, inter-end transmission: The inter-end node collaborative positioning task is triggered based on certain conditions. Multiple end nodes that meet the conditions within a small area are automatically grouped together. Each node sends its processed data to other nodes in the group and simultaneously receives data sent by other end nodes. Step S232, multi-station collaborative positioning: based on the information transmission between the end nodes, collaboratively complete the signal positioning task and obtain the positioning result.

[0045] In step S24, the sub-steps are included: Step S241, result saving: integrating the end-level processing results (signal recognition processing results, spectrum data processing results) and the collaborative processing results (positioning processing results) and saving them to the end signal history table; Step S242, result reporting: set the reporting frequency according to the network status between the end and the edge and the computing pressure of the edge node, and send the end-level processing results and collaborative processing results to the edge node corresponding to each end node.

[0046] In step S25, the sub-steps are included: Step S251, parsing and storing: parsing the signal recognition processing results, spectrum data processing results, and signal positioning processing results sent by the end node, and storing them in the edge signal history table in sequence; Step S252, spatial fusion: read the edge signal history table stored in the database, and perform edge-level fusion processing on multiple end-level signal recognition processing results, multiple end-level spectrum data processing results, and multiple end-level positioning processing results.

[0047] In step S26, the sub-steps are included: Step S261, result saving: Integrate the edge-level processing result list and save it as an edge signal history fusion table in the database, thereby completing the edge-level task; Step S262, result reporting: Set the reporting frequency based on the edge-cloud network status and cloud node computing pressure, and send the edge-level signal recognition result list and edge-level positioning result list to the corresponding cloud node to support subsequent cloud-level processing.

[0048] In step S28, the sub-steps are included: Step S281, parsing and storing: parsing the edge-level signal recognition result list and edge-level positioning result list reported by each edge node, and storing them in the database cloud signal history table; Step S282, spatial fusion: read the cloud signal history table stored in the database, and perform cloud-level fusion processing on the edge-level signal recognition processing results and the edge-level positioning processing results respectively.

[0049] Example 4 On the basis of Example 1, step S3 includes the following sub-steps: Step S31: Cloud-level result distribution: The comprehensive processing results of all monitored signal types within the cloud node coverage area are distributed to the corresponding edge node, and then distributed to the corresponding end node via the edge node. The edge and end nodes receive and analyze the cloud-level task processing results and store them in their respective databases. Step S32, edge-level result distribution: the comprehensive processing results of all spectrum data within the edge node coverage area are distributed to the corresponding end node, and the end node receives and analyzes the edge-level task processing results and stores them in the database.

[0050] In this embodiment, as a further optional implementation, step S31 includes the following sub-steps: Step S311, cloud-edge result delivery: The delivery frequency of cloud-level task results is set according to the network status between cloud and edge. The cloud node that starts the cloud-level task delivers the cloud-level task results stored in the database to each edge node, that is, the comprehensive processing results of all signal types monitored within the coverage area of ​​the cloud node. Each edge node receives and parses the results sent by the cloud node and saves them in the database. Step S311, edge-end result delivery: The delivery frequency of cloud-level task results is set according to the network status between the edge and the end. The edge node that receives the result sends the cloud-level task result to each corresponding end node. The end node receives and parses the result and saves it in the database.

[0051] In this embodiment, as a further optional implementation, step S31 includes the following sub-steps: Edge-to-end result delivery: The delivery frequency of edge-level task results is set according to the network status between the edges. The edge node that starts the edge-level task sends the edge-level task results stored in the database to each corresponding end node, that is, the comprehensive processing results of all spectrum data within the edge node coverage area. Each end node receives and parses the results and saves them in the database.

[0052] As a second aspect of the present invention, a cloud-edge-based hierarchical spectrum monitoring data processing device is also provided, including a processor and a memory, wherein a computer program is stored in the memory. When the computer program is loaded by the processor, it executes a method as described in any one of the above embodiments.

[0053] The units involved in the embodiments of the present invention may be implemented in software or hardware, and the units described may also be provided in a processor. In some cases, the names of these units do not limit the units themselves.

[0054] According to one aspect of an embodiment of the present invention, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations described above.

[0055] As another aspect, embodiments of the present invention further provide a computer-readable medium, which may be included in the electronic device described in the above embodiments, or may exist independently and not incorporated into the electronic device. The computer-readable medium carries one or more programs, and when executed by the electronic device, the electronic device implements the methods described in the above embodiments.

Claims

1. A cloud-edge-based hierarchical spectrum monitoring data processing method, characterized in that: include: S1, Task Distribution: Assign tasks to layers and nodes. Cloud and edge nodes serve as the initial task distribution points based on user needs. Cloud-level and edge-level spectrum monitoring tasks are distributed hierarchically from cloud to edge and then to end, or from edge to end. S2, Data Processing: Different data processing algorithms are configured on the end, edge, and cloud to perform multi-level distributed computing. After receiving a task, the monitoring data collected by multiple end nodes undergoes distributed end-level processing, and the fused data and key conclusion data are transmitted to the corresponding edge. Multiple edge-level distributed key processing results are then transmitted to the corresponding cloud for analysis. The cloud node uses computing resources and algorithms to analyze and process the spectrum data, while the edge and end nodes undertake part of the computation and share the data processing results at the same level. Multiple end nodes share data at the same level within a small range and autonomously and collaboratively process the same spectrum monitoring task. S3, result distribution: The processing results of cloud-level and edge-level tasks are distributed step by step from cloud to edge and then to end, or from edge to end.

2. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 1 is characterized in that: In step S1, tasks are assigned to layers and nodes. Cloud and edge nodes serve as initial task dispatching points based on user needs. Cloud-level and edge-level spectrum monitoring tasks are dispatched hierarchically from cloud to edge and then to end, or from edge to end. This includes the following sub-steps: Step S11, cloud-level task issuance: start the cloud-level spectrum monitoring task at the cloud node, and issue the task to the corresponding edge node, which is then issued by the edge node to the corresponding end node, triggering the data processing process; Step S12, edge-level task issuance: start the edge-level spectrum monitoring task at the edge node, and issue the task to the corresponding end node, triggering the data processing flow.

3. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 1 is characterized in that: In step S2, the end, edge, and cloud are configured with different data processing algorithms to perform multi-level distributed computing. After receiving the task, the monitoring data collected by multiple end nodes is processed at the distributed end level, and the fused data and key conclusion data are transmitted to the corresponding edge. Then, multiple edge-level distributed key processing results are transmitted to the corresponding cloud for analysis. The cloud node uses computing resources and algorithms to analyze and process the spectrum data, while the edge and end nodes undertake part of the calculation and share the data processing results at the same level. Multiple end nodes share at the same level within a small range and autonomously and collaboratively process the same spectrum monitoring task, including the following sub-steps: Step S21, end-level data collection: After receiving the task, the end node calls the spectrum monitoring device and continuously collects the specified spectrum data of the specified frequency band within the specified time according to the virtual interface of the standard protocol, and uniformly stores and manages the spectrum monitoring data; Step S22, end-level data processing: Perform end-level comprehensive processing on the collected data to eliminate the negative impact of signal fluctuations. At the same time, use algorithms that can respond in a timely manner to perform data analysis to support subsequent processing of edge-level tasks; Step S23, end-level collaborative processing: Under set conditions, multiple end nodes within a small area are grouped together to share end-level processing results and collaboratively complete spectrum monitoring tasks. Step S24, end-level result upload: Based on the region where the end node is located, the end-level data distributed processing results of the cloud and edge-level tasks and the autonomous collaborative processing results of the end node are reported to the edge node in the corresponding region; Step S25, edge-level data processing: The edge node receives and comprehensively processes the distributed end-level processing results reported by end nodes in multiple different regions, and stores them in the database to update and manage the processing results. At the same time, a cross-region processing algorithm is configured to perform data analysis to complete the edge-level task and support the subsequent processing of the cloud-level task. Step S26, edge-level result upload: reporting the edge-level distributed processing results of the cloud-level task to the cloud node; Step S27, edge-level result sharing: sharing edge-level processing results among edge nodes after obtaining permission; Step S28, cloud-level data processing: The cloud node receives and comprehensively processes the edge-level processing results reported by edge nodes in multiple different regions, and stores them in a database to update and manage the monitoring data processing results. It also uses algorithms that rely on a large amount of historical data to conduct in-depth analysis and processing of complex spectrum data to complete cloud-level tasks. Step S29, cloud-level result saving: saving the signal recognition results and signal location results within the cloud node area.

4. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 3 is characterized in that: In step S21, the designated spectrum data of the designated frequency band includes background noise and occupancy.

5. The cloud-edge-based hierarchical spectrum monitoring data processing method according to claim 3 is characterized in that: In step S22, the end-level comprehensive processing includes an averaging method and a merging method.

6. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 3, characterized in that: The use of an algorithm capable of timely response specifically includes using a trained neural network model to identify the signal and save the result.

7. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 3, characterized in that: In step S25, the cross-region processing algorithm specifically includes a target tracking algorithm.

8. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 3, characterized in that: In step S28, the algorithm that relies on a large amount of historical data specifically includes a computing resource and occupancy prediction algorithm.

9. The cloud-edge-end based hierarchical spectrum monitoring data processing method according to claim 1, characterized in that: In step S3, the processing results of the cloud-level and edge-level tasks are distributed step by step from the cloud to the edge and then to the end, or from the edge to the end, including the following sub-steps: Step S31, cloud-level result distribution: the cloud-level task processing results stored in the cloud node are distributed to each corresponding edge node, and then distributed to each corresponding end node via the edge node; Step S32, edge-level result distribution: the edge-level task processing result stored in the edge node is distributed to each corresponding end node.

10. A hierarchical spectrum monitoring data processing device based on cloud-edge, characterized in that: The method comprises a processor and a memory, wherein a computer program is stored in the memory, and when the computer program is loaded by the processor, the method according to any one of claims 1 to 9 is executed.

Images