Urban operation state monitoring method based on end-edge-cloud cooperative processing

Abstract

The application discloses a kind of city running state monitoring methods based on end side cloud collaborative processing, it is related to the technical field of cooperative perception and information fusion, this method includes: the basic operation data of edge server is collected;The computing requirement of equipment terminal is obtained, and the mapping relationship of equipment terminal and edge server is established based on computing requirement and basic operation data;Under the preset node, the overflow data set of equipment terminal is read, and intelligent identification is carried out, to determine the data processing requirement set;Match associated edge server, and judge whether to meet data processing requirement set;If not satisfied, computing requirement deviation is calculated and transferred to cloud server for data processing.Solve the technical problems that the existing city running state monitoring exists data acquisition difficulty and the accuracy, reliability is not high, data processing capacity is limited, and then lead to running state monitoring result is not accurate, real-time, reliable.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of cooperative perception and information fusion technology, in particular to a city operation state monitoring method based on end-edge-cloud cooperative processing. BACKGROUND

[0002] In recent years, with the continuous acceleration of urbanization process, the scale and complexity of the city are increasing, which brings great challenges to the operation and management of the city. In order to realize the intelligent and efficient operation and management of the city, the monitoring of the city operation state becomes an important task, which is to monitor and analyze various data information in the city in real time, and then understand the operation state of the city, plan the city, manage the traffic, etc. However, the traditional city operation state monitoring has limited ability to acquire various data in the city and process data, and has high data transmission delay, which cannot monitor the operation state of the city in real time and accurately.

[0003] Therefore, in the current city operation state monitoring related technology, there are technical problems of difficult data acquisition, low accuracy and reliability, limited data processing capacity, and thus the operation state monitoring result is not accurate, real-time and reliable. SUMMARY

[0004] The present application provides a city operation state monitoring method based on end-edge-cloud cooperative processing, which uses perception devices, edge servers and cloud servers to cooperate, allocates computing tasks, automatically stratifies data, and makes autonomous decisions, solves the technical problems of difficult data acquisition, low accuracy and reliability, limited data processing capacity, and thus the operation state monitoring result is not accurate, real-time and reliable in the existing city operation state monitoring, and achieves the technical effects of real-time and accurate data acquisition and efficient data processing, and improves the accuracy and reliability of the city operation state monitoring result.

[0005] The present application provides a city operation state monitoring method based on end-edge-cloud cooperative processing, which includes: collecting the basic operation data of the edge server, the basic operation data including position coordinates, data processing amount and confidence; acquiring the computing demand of the device terminal, establishing the mapping relationship between the device terminal and the edge server based on the computing demand and the basic operation data; reading the overflow data set of the device terminal at the preset node, and intelligently identifying the overflow data set to determine the data processing demand set; matching the associated edge server based on the mapping relationship, and judging whether the associated edge server meets the data processing demand set; if not, the computing demand deviation is calculated and transferred to the cloud server for data processing, wherein the edge server, the device terminal and the cloud server are in communication connection.

[0006] In a possible implementation, the mapping relationship between the device terminal and the edge server is established based on the computing requirement and the basic running data, and the following processing is performed: the computing requirement of the device terminal is obtained, and the computing requirement includes a predicted data volume and a predicted response speed; the basic running data is mapped and divided with the predicted data volume and the predicted response speed as constraints, and the mapping relationship between the device terminal and the edge server is established.

[0007] In a possible implementation, the computing requirement of the device terminal is obtained, and the following processing is performed: a historical computing requirement set of the device terminal in a preset time window is collected; the historical computing requirement set is subjected to outlier rejection and missing value supplement to obtain a standard historical computing requirement set; historical data volume and historical response speed are extracted based on the standard historical computing requirement set, and the predicted data volume and the predicted response speed are obtained through mean value calculation.

[0008] In a possible implementation, the overflow data set is intelligently identified to determine the data processing requirement set, and the following processing is further performed: a network structure of a confidence recognition channel is constructed based on a BP neural network; a sample training set is obtained, and the confidence recognition channel is subjected to supervised learning through the sample training set to obtain a confidence recognition channel in a convergent state; the overflow data set is identified through the confidence recognition channel to output a confidence requirement set; an overflow data volume and a requirement response speed of the overflow data set are read, and the data processing requirement set is obtained in combination with the confidence requirement set.

[0009] In a possible implementation, the associated edge server is determined to be whether to meet the data processing requirement set, and the following processing is further performed: a first data processing requirement in the data processing requirement set is selected, and the first data processing requirement includes a first confidence requirement, a first overflow data volume, and a first requirement response speed; a first device terminal corresponding to the first data processing requirement is read, and a first edge server set is obtained based on the mapping relationship; the first edge server set is filtered to determine a matched edge server set with the first confidence requirement as a constraint; whether the matched edge server in the matched edge server set meets the first overflow data volume and the first requirement response speed is sequentially determined; if neither of them meets the first overflow data volume and the first requirement response speed, it is determined that the associated edge server does not meet the data processing requirement set.

[0010] In a possible implementation, before the sequentially judging whether the matched edge server in the matched edge server set meets the first overflow data quantity and the first demand response speed, the following processing is further performed: reading a remaining data processing space and a real-time response speed of the matched edge server; judging the first overflow data quantity and the first demand response speed based on the remaining data processing space and the real-time response speed.

[0011] In a possible implementation, if the matched edge server does not meet the demand, the cloud server is calculated for a demand deviation and is transferred to the cloud server for data processing, and the following processing is further performed: obtaining a preset confidence of the cloud server; if the preset confidence does not meet a confidence demand, the matched edge server is used for waiting processing on the overflow data.

[0012] The application further provides a city operation state monitoring system based on end-edge-cloud collaborative processing, comprising:

[0013] A basic operation data acquisition module is configured to acquire basic operation data of an edge server, wherein the basic operation data comprises position coordinates, data processing quantity and confidence;

[0014] A terminal server mapping relationship establishment module is configured to acquire a computing demand of a device terminal, and establish a mapping relationship between the device terminal and the edge server based on the computing demand and the basic operation data;

[0015] An overflow data set intelligent identification module is configured to read an overflow data set of the device terminal at a preset node, and intelligently identify the overflow data set to determine a data processing demand set;

[0016] An associated edge server matching module is configured to match an associated edge server based on the mapping relationship, and judge whether the associated edge server meets the data processing demand set;

[0017] A cloud server data processing module is configured to, if the associated edge server does not meet the demand, calculate a demand deviation and transfer to the cloud server for data processing, wherein the edge server, the device terminal and the cloud server are in communication connection.

[0018] The urban operation state monitoring method based on edge-cloud collaborative processing provided in the application collects basic operation data of an edge server, wherein the basic operation data includes position coordinates, data processing amount and confidence level; obtains the computing requirement of a device terminal, and establishes a mapping relationship between the device terminal and the edge server based on the computing requirement and the basic operation data; reads the overflow data set of the device terminal under a preset node, and intelligently identifies the overflow data set to determine a data processing requirement set; matches an associated edge server based on the mapping relationship, and judges whether the associated edge server meets the data processing requirement set; if not, the computing requirement deviation is calculated and transferred to a cloud server for data processing, wherein the edge server, the device terminal and the cloud server are in communication connection, which solves the technical problems that the existing urban operation state monitoring has difficulty in data acquisition, and the accuracy and reliability are not high, and the data processing capability is limited, thereby resulting in that the operation state monitoring result is not accurate, real-time and reliable, achieves real-time and accurate data acquisition and efficient data processing, and achieves the technical effect of improving the accuracy and reliability of the urban operation state monitoring result. BRIEF DESCRIPTION OF DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings of the embodiments of the present disclosure will be briefly introduced below. The flowcharts are used to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the foregoing or the following operations are not necessarily performed in sequence. On the contrary, according to the needs, various steps can be processed in reverse order or at the same time. At the same time, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0020] Figure 1 The urban operation state monitoring method based on edge-cloud collaborative processing provided in the embodiments of the present application is shown in the flowchart.

[0021] Figure 2 The urban operation state monitoring system structure diagram based on edge-cloud collaborative processing provided in the embodiments of the present application is shown in the structure diagram. DETAILED DESCRIPTION

[0022] The above description is only a summary of the technical solutions of the present application. In order to more clearly understand the technical means of the present application, the embodiments of the present application can be implemented according to the content of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the following specific embodiments of the present application are described.

[0023] In order to make the purposes, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings, and the described embodiments should not be regarded as limitations to the present application. All other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0024] In the following description, "some embodiments" are referred to, which describe a subset of all possible embodiments, but it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict, and the term "first\second" referred to is only to distinguish similar objects, and does not represent a specific order of the objects. The terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or server including a series of steps or units does not have to be limited to those steps or units clearly listed, but can include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as understood by those skilled in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0025] The embodiments of the present application provide a city running state monitoring method based on end-side cloud collaborative processing, as shown in Figure 1 The method comprises the following steps:

[0026] City running state monitoring is a process for real-time monitoring, collecting and analyzing various elements and indicators in the city, understanding the city running state, discovering problems and optimizing city running management, covering data information such as traffic flow, congestion condition, air quality, noise monitoring, energy use, etc. City running state monitoring relies on a large number of sensors, monitoring equipment and data collection devices. The goal of city running state monitoring is to realize the intelligentization and sustainable development of the city, discover problems in time and take measures to solve traffic congestion, environmental pollution, resource waste, etc. It also provides a scientific basis for city planning management, environmental protection, etc. to optimize the operation efficiency of the city.

[0027] The "end-edge-cloud collaboration" fusion processing technology is a fine distributed computing model. Through the collaborative work among the sensing devices, edge servers and cloud servers, the computing tasks are processed and distributed, the acquired data is automatically layered and judged, and the information processing nodes are autonomously decided. The technology classifies and stores the original data according to the confidence level of the data, and according to the confidence level, the data with different information amounts are transmitted to the cloud for further processing. In the cloud, the system performs secondary identification according to the uploaded event confidence, uses different algorithms for different confidence data, actively retrieves the original effective data as needed, and realizes the collaborative optimization and automatic deployment of large, medium and small models. This kind of automatic layered processing significantly reduces the storage and transmission requirements of data and the computing pressure of the cloud, thereby improving the system efficiency. The technology can dynamically adjust the computing tasks and resources, so that the entire system can more flexibly cope with various application requirements and environmental changes, actively analyze the quality and value of data, and through the hierarchical processing of data confidence, the accuracy and reliability of data processing can be improved, the robustness and adaptability of the system can be enhanced, so that the system can enhance the stability while ensuring high accuracy without relying on a large amount of bandwidth and server cluster.

[0028] In step S100, the basic running data of the edge server is collected, including position coordinates, data processing amount and confidence. Specifically, the edge server is a server close to the user device for storing and processing data, which is deployed close to the user terminal device or data source. The basic running data of the edge server is collected, including position coordinates, data processing amount and confidence. The position coordinates refer to the geographic position information of the edge server, such as latitude and longitude coordinate data, etc. The position coordinates of the edge server can be used to determine the geographic position of the edge server, and to make position-related analysis and decision, such as network resource optimization, service scheduling, etc. The data processing amount refers to the amount of data processed by the edge server, which is an important indicator for evaluating the processing capacity of the edge server, and can be used to monitor the load and performance of the server, such as the number of data processed per second or per minute, data transmission width, storage capacity, etc. The confidence refers to the reliability of the data processed by the edge server, reflecting the trust degree of the edge server to the data, which can include the accuracy of data processing, the credibility of data source, the stability of data transmission, etc. Through the confidence, the reliability of the data processed by the edge server can be evaluated to ensure the credibility, accuracy and integrity of the data. The confidence can be measured and calculated through source verification, comparison of redundant data, etc.

[0029] After obtaining the basic running data of the edge server, step S200 is performed to obtain the computing requirement of the device terminal, and a mapping relationship between the device terminal and the edge server is established based on the computing requirement and the basic running data. The device terminal refers to various terminal devices used by users for data processing, communication and interaction, such as desktop computers, laptops, smartphones, tablets, smart home devices, sensor devices and vehicle-mounted intelligent terminal devices, etc. The computing requirement refers to the computing capability and resource requirement of the device terminal in a specific application scenario, which can include computing capability, network requirement, memory requirement, etc. The computing capability of the device terminal is used to judge the capability of the device terminal to execute tasks, such as the type and speed of the processor (CPU) of the device terminal, the number of cores, the frequency, etc. The network requirement refers to the requirement of the device terminal on the bandwidth delay and stability when connecting to the network, including the type of network used (such as Wi-Fi, 4G / 5G). The memory requirement refers to the requirement of the device terminal on the memory (RAM) capacity, such as the size of the memory, the hard disk space and other hardware specifications. Tasks that process a large amount of data usually require higher memory capacity to ensure good performance and response speed. The mapping relationship between the device terminal and the edge server is established based on the computing requirement and the basic running data, which means that the device terminal and the edge server are matched and connected according to the computing requirement of the device terminal and the basic running data of the edge server, and an appropriate corresponding mapping relationship is established to realize more efficient computing resource allocation and data processing. Specifically, the mapping relationship can include computing resource allocation, data processing strategy and data storage management, etc. The computing resource allocation refers to establishing the corresponding relationship between the device terminal and the edge server according to the computing requirement of the device terminal and the running data characteristics of the edge server, and reasonably allocating resources, such as matching the device terminal with high computing requirement with the edge server with good performance. The data processing strategy refers to allocating the terminal data to the edge server for processing according to the type, amount and processing requirement of the data, such as allocating large-scale data processing tasks, complex algorithms or real-time sensor data processing tasks to the edge server. The data storage management refers to storing data in a suitable location according to the storage requirement of the device terminal running data and the storage capacity of the edge server to realize efficient management and access of data. By establishing the mapping relationship between the device terminal and the edge server, the computing resources of the edge server can be utilized to the maximum extent, the computing efficiency and response speed can be improved, the load of the device terminal can be reduced, and large-scale real-time data processing can be realized.

[0030] In a possible implementation, the step S200 of establishing the mapping relationship between the device terminal and the edge server based on the computing requirement and the basic running data further includes the step S210 of obtaining the computing requirement of the device terminal, the computing requirement including a predicted data volume and a predicted response speed. Specifically, the predicted data volume refers to the data volume generated, received, and processed by the device terminal according to the application scenario and data processing requirement of the device terminal; and the predicted response speed refers to the expected response speed of the device terminal, such as task execution time and computing delay, according to the application requirement of the device terminal and the complexity of data processing and computing task. The step S220 of mapping and dividing the basic running data to meet the predicted data volume and the predicted response speed as constraints is further included to establish the mapping relationship between the device terminal and the edge server. Through the steps S210-S220, the predicted data volume and the predicted response speed of the device terminal are obtained, the basic running data of the edge server is mapped and divided, and finally the mapping relationship between the device terminal and the edge server is established, realizing the network interconnection of "end-edge", and realizing efficient processing of data and ensuring the accuracy of data.

[0031] In a possible implementation, the step S210 of obtaining the predicted data volume and the predicted response speed of the device terminal further includes the step S211 of collecting a historical computing requirement set of the device terminal in a preset time window. Specifically, the computing requirement of the device terminal is collected and recorded in a preset time range, and the historical computing requirement set refers to a collection of historical computing requirement data of the device terminal. The step S212 of removing outliers and supplementing missing values from the historical computing requirement set to obtain a standard historical computing requirement set is further included. Specifically, the historical computing requirement set is subjected to outlier removal and missing value supplementation, the outlier removal refers to removing the detected outliers from the historical computing requirement set, for example, using outlier detection to identify outliers; and the missing value supplementation refers to supplementing the historical computing requirement set when the proportion of missing values is large, for example, using mean interpolation to supplement missing values. The step S213 of extracting historical data volume and historical response speed based on the standard historical computing requirement set and performing mean calculation to obtain the predicted data volume and the predicted response speed is further included. Through the steps S211-S213, the historical computing requirement data of the device terminal is obtained, and the computing requirement of the device terminal is predicted based on the historical computing requirement to obtain the predicted data volume and the predicted response speed, which facilitates the subsequent establishment of the mapping relationship between the device terminal and the edge server and realizes efficient processing of data.

[0032] After the mapping relationship between the device terminal and the edge server is established, step S300 is performed, and the overflow data set of the device terminal is read under a preset node, and the overflow data set is intelligently identified to determine a data processing requirement set. Specifically, the overflow data set is read from the device terminal. These overflow data may be data lost or not timely processed in the transmission process of the device terminal. The overflow data is identified and processed using intelligent identification technology, for example, the type of the overflow data is intelligently identified, the integrity of the overflow data is analyzed, and abnormal values possibly existing in the overflow data are identified. Based on the intelligent identification result of the overflow data set, the data processing requirement set is determined.

[0033] In a possible manner, step S300 intelligently identifies the overflow data set to determine the data processing requirement set, and further includes step S310 of constructing a network structure of a confidence recognition channel based on a BP neural network. Specifically, a channel for recognizing confidence is constructed by using a BP neural network, and the network structure and parameters of the recognition channel are designed and trained based on the BP neural network. In the network structure of the confidence recognition channel, the input layer receives incoming data, the hidden layer is responsible for processing data and feature extraction, and the output layer generates the final confidence recognition prediction result. Step S320 includes obtaining a sample training set and performing supervised learning on the confidence recognition channel through the sample training set to obtain a confidence recognition channel that tends to be in a convergent state. Step S330 includes identifying the overflow data set through the confidence recognition channel to output a confidence requirement set. Specifically, the overflow data set is input into the constructed confidence recognition channel for intelligent identification, the confidence prediction value of each sample data is calculated, and the confidence requirement set is output. Step S340 includes reading the overflow data amount and demand response speed of the overflow data set, and obtaining the data processing requirement set in combination with the confidence requirement set. Steps S310 to S340 perform intelligent identification of the confidence of the sample data through the constructed confidence recognition channel, and finally obtain the data processing requirement set in combination with the overflow data amount and demand response speed of the overflow data set.

[0034] After the data processing requirement set is obtained, step S400 is performed to match the associated edge server based on the mapping relationship and determine whether the associated edge server meets the data processing requirement set. According to the requirements in the data processing requirement set, the data processing requirement set is matched with the available edge server, and it is determined whether the edge server meets the requirements in the data processing requirement set. Specifically, according to the pre-defined mapping relationship, the data processing requirement set is matched with the edge server, the associated edge server is determined according to the matching result, and the associated edge server is evaluated to determine whether it meets the data processing requirement set, so as to ensure that the edge server can meet the data processing speed and resource requirement in the data processing requirement.

[0035] In a possible implementation, the step S400 of judging whether the associated edge server meets the data processing requirement set further includes the following steps. Step S410, a first data processing requirement in the data processing requirement set is selected, wherein the first data processing requirement includes a first confidence requirement, a first overflow data volume, and a first demand response speed. Step S420, a first device terminal corresponding to the first data processing requirement is read, and a first edge server set is obtained based on the mapping relationship matching. Specifically, the first device terminal is found in the preset device terminal according to the device terminal information corresponding to the first data processing requirement, and all edge servers associated with the first device terminal are found by matching the first device terminal according to the mapping relationship between the device terminal and the edge server, that is, the first edge server set. Step S430, the first edge server set is filtered to determine a matching edge server set, with the first confidence requirement as a constraint. Specifically, the edge servers in the first edge server set that do not meet the confidence requirement are filtered out according to the first confidence requirement, and the edge servers that meet the confidence requirement are combined to form a new edge server set, which is referred to as the matching edge server set. Step S440, whether the matching edge server in the matching edge server set meets the first overflow data volume and the first demand response speed is judged in sequence. The edge servers in the matching edge server set are judged whether they meet the first overflow data volume and the first demand response speed. Step S450, if neither of them meets the first overflow data volume and the first demand response speed, it is determined that the associated edge server does not meet the data processing requirement set.

[0036] In a possible implementation, the step S440 further includes a step S436 of reading the residual data processing space and real-time response speed of the matching edge server. Specifically, information of the matching edge server set is acquired, and the residual data processing space of each matching edge server is acquired, for example, the storage capacity and current usage of the edge server are inquired; test data is sent to the edge server to evaluate the real-time response speed of the matching edge server. The step S437 of judging the first overflow data stream and the first demand response speed based on the residual data processing space and the real-time response speed is further included. Specifically, the residual data processing space is compared with the size of the first overflow data stream. If the residual data processing space is greater than or equal to the size of the first overflow data stream, it indicates that the edge server has enough space to process the data stream; the real-time response speed is compared with the first demand response speed. If the real-time response speed can meet the requirement of the first demand response speed, it indicates that the edge server can respond to the processing request in time.

[0037] Next, the step S500 of calculating a demand deviation and transferring data to a cloud server for data processing is performed if the condition is not met, wherein the edge server, the device terminal and the cloud server are in communication connection. Specifically, when the associated edge server does not meet the data processing demand set, the demand deviation is calculated, and the demand in the data processing demand set is compared with the resource capability of the associated edge server to calculate the difference in resources or the degree of demand exceeding the server capability, for example, the demand deviation is calculated according to the processing speed, resource requirement in the data processing demand set and the computing capability, storage capacity of the associated edge server, and the data is transferred to the cloud server for data processing through network transmission to ensure that the data can be processed in time by the high-performance cloud server. By calculating the demand deviation and transferring the data to the cloud server for processing, the characteristics of the edge server and the cloud server can be fully utilized to provide suitable processing resources to meet the data processing demand set and perform data processing.

[0038] In a possible implementation, the step S500 further includes a step S510 of acquiring a preset confidence of the cloud server. The preset confidence is a confidence degree set based on the capability of the cloud server to perform a task. The step S520 of waiting for processing of the overflow data by the matching edge server if the preset confidence does not meet the confidence requirement is further included. Specifically, when the preset confidence of the cloud server cannot meet the confidence requirement, the overflow data in the matching edge server is in a waiting state and is not processed temporarily until a cloud server that meets the requirement is found, and then the data is transferred to the cloud server for data processing.

[0039] In the foregoing, reference is made to Figure 1The method for monitoring the urban operation state based on the end-edge-cloud collaborative processing according to the embodiment of the present application is described in detail. Next, the method for monitoring the urban operation state based on the end-edge-cloud collaborative processing according to the embodiment of the present application will be described in detail with reference to the accompanying drawings. Figure 2 The system for monitoring the urban operation state based on the end-edge-cloud collaborative processing according to the embodiment of the present application is described.

[0040] The system for monitoring the urban operation state based on the end-edge-cloud collaborative processing according to the embodiment of the present application is used to solve the technical problem that the existing urban operation state monitoring has the difficulty in data acquisition and the low accuracy and reliability, and the limited data processing capacity, thereby resulting in the inaccurate, real-time and unreliable operation state monitoring result, so as to achieve the technical effect of real-time and accurate data acquisition and efficient data processing, and thereby improving the accuracy and reliability of the urban operation state monitoring result. The system for monitoring the urban operation state based on the end-edge-cloud collaborative processing comprises a basic operation data acquisition module 10, a terminal server mapping relationship establishment module 20, an overflow data set intelligent identification module 30, an associated edge server matching module 40 and a cloud server data processing module 50.

[0041] The basic operation data acquisition module 10 is used to acquire the basic operation data of the edge server, and the basic operation data comprises the position coordinates, the data processing capacity and the confidence degree.

[0042] The terminal server mapping relationship establishment module 20 is used to acquire the computing requirement of the device terminal, and establish the mapping relationship between the device terminal and the edge server based on the computing requirement and the basic operation data.

[0043] The overflow data set intelligent identification module 30 is used to read the overflow data set of the device terminal under the preset node, and intelligently identify the overflow data set to determine the data processing requirement set.

[0044] The associated edge server matching module 40 is used to match the associated edge server based on the mapping relationship, and judge whether the associated edge server meets the data processing requirement set.

[0045] The cloud server data processing module 50 is used to calculate the deviation of the computing requirement and flow to the cloud server for data processing if the associated edge server does not meet the data processing requirement set, wherein the edge server, the device terminal and the cloud server are in communication connection.

[0046] In the following, the specific configuration of the terminal server mapping relationship establishing module 20 will be described in detail. As described above, the mapping relationship between the device terminal and the edge server is established based on the computing requirement and the basic running data, and the terminal server mapping relationship establishing module 20 can further include: obtaining the computing requirement of the device terminal, the computing requirement including the predicted data volume and the predicted response speed; mapping and dividing the basic running data to establish the mapping relationship between the device terminal and the edge server, with the constraint of satisfying the predicted data volume and the predicted response speed.

[0047] In the following, the specific configuration of the terminal server mapping relationship establishing module 20 will be described in detail. The terminal server mapping relationship establishing module 20 further includes: collecting a set of historical computing requirements of the device terminal within a preset time window; performing outlier rejection and missing value supplementation on the set of historical computing requirements to obtain a set of standard historical computing requirements; extracting historical data volume and historical response speed based on the set of standard historical computing requirements, and performing mean value calculation to obtain the predicted data volume and the predicted response speed.

[0048] In the following, the specific configuration of the overflow data set intelligent identification module 30 will be described in detail. As described above, the overflow data set is intelligently identified to determine the data processing requirement set, and the overflow data set intelligent identification module 30 can further include: constructing the network structure of the confidence recognition channel based on the BP neural network; obtaining a sample training set, and performing supervised learning on the confidence recognition channel through the sample training set to obtain a confidence recognition channel that tends to be in a convergent state; identifying the overflow data set through the confidence recognition channel respectively, and outputting a confidence requirement set; reading the overflow data volume and the demand response speed of the overflow data set, and combining the confidence requirement set to obtain the data processing requirement set.

[0049] In the following, the specific configuration of the associated edge server matching module 40 will be described in detail. As described above, the associated edge server is determined whether to satisfy the data processing requirement set, and the associated edge server matching module 40 further includes: selecting a first data processing requirement in the data processing requirement set, the first data processing requirement including a first confidence requirement, a first overflow data volume, and a first demand response speed; reading a first device terminal corresponding to the first data processing requirement, and obtaining a first edge server set based on the mapping relationship matching; filtering the first edge server set to determine a matching edge server set with the constraint of satisfying the first confidence requirement; sequentially determining whether the matching edge servers in the matching edge server set satisfy the first overflow data volume and the first demand response speed; if none of them satisfies, it is determined that the associated edge server does not satisfy the data processing requirement set.

[0050] Next, the specific configuration of the associated edge server matching module 40 will be described in detail. As described above, before the associated edge server matching module 40 sequentially judges whether the matching edge server in the matching edge server set meets the first overflow data volume and the first demand response speed, the associated edge server matching module 40 further comprises: reading the remaining data processing space and real-time response speed of the matching edge server; judging the first overflow data volume and the first demand response speed based on the remaining data processing space and the real-time response speed.

[0051] Next, the specific configuration of the cloud server data processing module 50 will be described in detail. As described above, if the demand deviation is not met, the cloud server data processing module 50 can further comprise: obtaining the pre-set confidence of the cloud server; if the pre-set confidence does not meet the confidence requirement, the overflow data is processed by the matching edge server.

[0052] The urban operation state monitoring system based on end-edge-cloud collaborative processing provided by the embodiments of the present application can execute the urban operation state monitoring method based on end-edge-cloud collaborative processing provided by any of the embodiments of the present application, and has the corresponding function modules and beneficial effects of the execution method.

[0053] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or server, and each unit and module included is only divided according to the functional logic, but is not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific name of each functional unit is only for easy mutual differentiation, and does not limit the protection scope of the present application.

[0054] The above specific embodiments do not constitute a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent replacement and improvement made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A city operation state monitoring method based on edge-cloud collaborative processing, characterized in that, The method comprises: Collecting basic operation data of the edge server, the basic operation data comprising position coordinates, data processing capacity and confidence level; Obtaining the computing requirement of the device terminal, and establishing a mapping relationship between the device terminal and the edge server based on the computing requirement and the basic operation data; Reading the overflow data set of the device terminal under a preset node, and intelligently identifying the overflow data set to determine a data processing requirement set; Matching the associated edge server based on the mapping relationship, and determining whether the associated edge server meets the data processing requirement set; If not, calculating the requirement deviation and transferring to the cloud server for data processing, wherein the edge server, the device terminal and the cloud server are in communication connection; The method further comprises: Collecting the historical computing requirement set of the device terminal within a preset time window; Performing outlier rejection and missing value supplementation on the historical computing requirement set to obtain a standard historical computing requirement set; Extracting historical data volume and historical response speed based on the standard historical computing requirement set, and performing mean value calculation to obtain the predicted data volume and the predicted response speed; The method further comprises: Based on the BP neural network, a network structure of a confidence level identification channel is constructed; A sample training set is obtained, and the confidence level identification channel is supervised to learn through the sample training set to obtain a confidence level identification channel that tends to be in a convergent state; The overflow data set is identified through the confidence level identification channel, and a confidence level requirement set is outputted; The overflow data volume and the requirement response speed of the overflow data set are read, and the data processing requirement set is obtained in combination with the confidence level requirement set. The method further comprises: A first data processing requirement in the data processing requirement set is selected, the first data processing requirement comprising a first confidence level requirement, a first overflow data volume and a first requirement response speed; A first device terminal corresponding to the first data processing requirement is read, and a first edge server set is obtained based on the mapping relationship; 2. The city operation state monitoring method based on edge-cloud collaborative processing according to claim 1, characterized in that, The first edge server set is filtered to determine a matching edge server set under the constraint of meeting the first confidence level requirement; It is determined in turn whether the matching edge servers in the matching edge server set meet the first overflow data volume and the first requirement response speed; If none of them meet the requirements, it is determined that the associated edge server does not meet the data processing requirement set. The method further comprises: ​ ​ 3.The urban running state monitoring method based on edge-cloud collaborative processing according to claim 2, characterized in that, ​ read the remaining data processing space and real-time response speed of the matching edge server; judge the first overflow data amount and the first demand response speed based on the remaining data processing space and the real-time response speed.

4. The city operation state monitoring method based on edge-cloud collaborative processing according to claim 1, characterized in that, if not, calculate the demand deviation and transfer to the cloud server for data processing, including: obtain the pre-set confidence of the cloud server; if the pre-set confidence does not meet the confidence requirement, wait for the matching edge server to process the overflow data.

5. The urban operation state monitoring system based on edge-cloud collaborative processing, characterized in that, The system is used to implement the city operation state monitoring method based on edge-cloud collaborative processing according to any one of claims 1-4, and the system comprises: a basic operation data acquisition module, the basic operation data acquisition module is used to acquire the basic operation data of the edge server, and the basic operation data includes position coordinates, data processing amount and confidence; a terminal server mapping relationship establishment module, the terminal server mapping relationship establishment module is used to acquire the computing demand of the device terminal, and establish the mapping relationship between the device terminal and the edge server based on the computing demand and the basic operation data; an overflow data set intelligent identification module, the overflow data set intelligent identification module is used to read the overflow data set of the device terminal under the pre-set node, and intelligently identify the overflow data set to determine the data processing demand set; an associated edge server matching module, the associated edge server matching module is used to match the associated edge server based on the mapping relationship, and judge whether the associated edge server meets the data processing demand set; a cloud server data processing module, the cloud server data processing module is used to calculate the demand deviation and transfer to the cloud server for data processing if not, wherein the edge server, the device terminal and the cloud server are in communication connection.

Images