Network operation management method and system, and electronic equipment

Abstract

The embodiment of the invention relates to a network operation management method and system and electronic equipment, and relates to the technical field of operation management, and the method comprises the steps: obtaining to-be-processed network data, and carrying out the spatial domain feature extraction of the to-be-processed network data, and obtaining corresponding spatial domain features; performing time domain feature extraction on to-be-processed network data to obtain a time domain feature corresponding to the to-be-processed network data; and fusing the time domain feature and the spatial domain feature to obtain a fused feature, performing multi-scale feature extraction on the fused feature, and determining a network traffic prediction value corresponding to the to-be-processed network data, so that an operation management system performs operation management on a target network based on the network traffic prediction value. According to the invention, the accuracy of network traffic prediction can be improved.

Description

Background Art

[0002] With the continuous development of technologies such as 5G, Internet of Things and cloud computing, and the emergence of an endless stream of new telecommunications network services, network load is increasing and scale is expanding. The network will face difficulties in operation and management and a surge in traffic.

[0003] In the related technologies, it is impossible to realize the operational management of the digital twin network based on the operation system, and the accuracy of the predicted network traffic is low, so it has limitations and cannot achieve precise management of the network. Summary of the invention

[0004] The purpose of the present disclosure is to provide a network operation management method, a network operation management system and an electronic device, thereby overcoming the problem of limitations of network operation management caused by the limitations and defects of related technologies at least to a certain extent.

[0005] According to a first aspect of the present disclosure, a network operation management method is provided, comprising: obtaining network data to be processed, and performing spatial domain feature extraction on the network data to be processed to obtain corresponding spatial domain features; performing time domain feature extraction on the network data to be processed to obtain time domain features corresponding to the network data to be processed; fusing the time domain features and the spatial domain features to obtain fused features, and performing multi-scale feature extraction on the fused features to determine a network traffic prediction value corresponding to the network data to be processed, so that an operation management system performs operation management on a target network based on the network traffic prediction value.

[0006] In an exemplary embodiment of the present disclosure, the extracting spatial domain features of the network data to be processed to obtain corresponding spatial domain features includes: obtaining a forward hidden vector of the network data to be processed, and determining a backward hidden vector of the network data to be processed; fusing the forward hidden vector and the backward hidden vector to obtain the spatial domain features.

[0007] In an exemplary embodiment of the present disclosure, the fusing of the forward hidden vector and the backward hidden vector to obtain the spatial domain feature includes: concatenating and combining the forward hidden vector and the backward hidden vector to determine the spatial domain vector.

[0008] In an exemplary embodiment of the present disclosure, extracting time domain features from the network data to be processed to obtain the time domain features includes: converting the network data to be processed into the frequency domain, and performing a convolution operation on the network data to be processed in the frequency domain to obtain the time domain features.

[0009] In an exemplary embodiment of the present disclosure, the convolution operation is performed on the network data to be processed in the frequency domain to obtain the time domain features, including: determining the network data to be processed in the current layer in the graph convolutional network, and performing a convolution operation on the network data to be processed of each node in the current layer to obtain a convolution result; updating the network data to be processed in the next layer according to the convolution result, and continuing to perform a convolution operation on the network data to be processed of each node in the next layer to obtain a convolution result to obtain the time domain features.

[0010] In an exemplary embodiment of the present disclosure, the multi-scale feature extraction is performed on the fused features to determine the network traffic prediction value corresponding to the network data to be processed, including: performing multi-scale feature extraction on the fused features respectively to obtain multi-scale features at different levels; performing a convolution operation on the multi-scale features to determine the network traffic prediction value.

[0011] In an exemplary embodiment of the present disclosure, the convolution operation is performed on the multi-scale features to determine the network traffic prediction value, including: performing a convolution operation on the multi-scale features based on a jump connection module, and processing the multi-scale convolution results through an activation function to obtain the network traffic prediction value.

[0012] In an exemplary embodiment of the present disclosure, the operational management of the target network based on the network traffic prediction value includes: managing the network status of the target network based on the network traffic prediction value.

[0013] According to a second aspect of the present disclosure, a network operation and management system is provided, including: an external system and a target network; an operation and management system, used to extract spatial domain features of the network data to be processed to obtain corresponding airspace features; extract time domain features of the network data to be processed to obtain time domain features; fuse the time domain features and the airspace features to obtain fused features, and perform multi-scale feature extraction on the fused features to determine a network traffic prediction value corresponding to the network data to be processed, so as to perform operation and management on the target network based on the network traffic prediction value.

[0014] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute any one of the above-mentioned network operation management methods by executing the executable instructions.

[0015] In the technical solution provided in the embodiment of the present disclosure, on the one hand, by extracting the features of the time domain and space domain of the processed network data, the spatial domain features and the time domain features are obtained, and the network traffic prediction value is obtained by further extracting multi-scale features based on the spatial domain features and the time domain features. Through multi-dimensional features, the traffic prediction can be comprehensively performed, thereby improving the accuracy of the network traffic prediction. On the other hand, due to the improved accuracy of the traffic prediction, the operation management system can perform operation management and services for the target network based on the predicted network traffic, which can improve the reliability and stability of the target network. Based on the operation management system, the target network can be fully autonomously operated and managed, which can help operators make operation and maintenance decisions quickly and accurately, improve scheduling efficiency, and improve network availability.

[0016] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings herein are incorporated into the specification and constitute a part of the specification, illustrate embodiments consistent with the present disclosure, and together with the specification are used to explain the principles of the present disclosure. Obviously, the accompanying drawings described below are only some embodiments of the present disclosure, and for ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0018] Figure 1 A flowchart of a network operation management method in an embodiment of the present disclosure is schematically shown.

[0019] Figure 2 The structural diagram of the digital twin network operation and management system in the embodiment of the present disclosure is schematically shown.

[0020] Figure 3 The structural diagram of the network traffic prediction model in the embodiment of the present disclosure is schematically shown.

[0021] Figure 4 The structural diagram of the spatial domain feature extraction module in the embodiment of the present disclosure is schematically shown.

[0022] Figure 5 The structural diagram of the multi-scale convolution module of an embodiment of the present disclosure is schematically shown.

[0023] Figure 6 The flowchart of the overall interaction in the embodiment of the present disclosure is schematically shown.

[0024] Figure 7 The block diagram of the network operation management system in the embodiment of the present disclosure is schematically shown.

[0025] Figure 8The block diagram of the electronic device according to the embodiment of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0026] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as being limited to the examples set forth herein; on the contrary, these embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concepts of the example embodiments are fully conveyed to those skilled in the art. The described features, structures, or characteristics may be combined in one or more embodiments in any suitable manner. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced while omitting one or more of the specific details, or other methods, components, devices, steps, etc. may be adopted. In other cases, known technical solutions are not shown or described in detail to avoid obscuring various aspects of the present disclosure.

[0027] In addition, the accompanying drawings are only schematic illustrations of the present disclosure and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and thus their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0028] In an embodiment of the present disclosure, a network operation management method is provided to facilitate operation management services for a target network based on an operation management system, so that the target network can provide safe, stable and continuous services and improve the reliability level.

[0029] Next, refer to Figure 1 The network operation management method in the embodiment of the present disclosure is specifically described as shown in FIG.

[0030] In step S110, the network data to be processed is obtained, and spatial domain features are extracted from the network data to be processed to obtain corresponding spatial domain features.

[0031] In the embodiments of the present disclosure, in order to improve the performance of the network, the target network corresponding to the operation management system can be operated and managed by the operation management system to manage the functions of the target network. The operation management system can be any type of system that can provide services and capability management, and the target network can be any type of network corresponding to the operation management system, as long as it can be managed by the operation management system. In the embodiments of the present disclosure, the operation management system is a digital twin network operation management system, and the target network is a digital twin network.

[0032] In order to improve the service between the digital twin network operation and management system and the digital twin network, the digital twin network operation and management system can provide end-to-end network traffic prediction function, so as to realize the management of the digital twin network. This includes using the digital twin network management system to realize the fault prediction and service deployment of the existing network, so that the information communication network can realize completely autonomous operation and management, making the entire network run more efficiently.

[0033] The digital twin network operation and management system can be connected to external systems and digital twin networks. A digital twin network refers to a network system that creates virtual twins of physical network entities in a digital way and can interact and map with physical network entities as real-time as possible. A digital twin network can include a data sharing warehouse, service mapping model management, and network twin management.

[0034] Digital twin network operation and management system: Based on the operation and management services of the digital twin network, it mainly manages the hardware, software, data, users, interfaces and other objects that support the digital twin network. The management object of the digital twin network operation and management system is the digital twin network, that is, the digital twin network operation and management system is used to operate and manage the digital twin network and provide operation and management services to it to maintain the healthy, stable and continuous operation of the digital twin network. The digital twin network operation and management system mainly provides full life cycle operation and management services for the digital twin network from multiple dimensions such as network status, hardware, software, data, service capabilities, users, security and maturity, and provides digital twin network operation and management capability services, capability orchestration and capability exposure functions.

[0035] The main functions of the digital twin network operation management system include the following aspects:

[0036] 1. Service requirements for digital twin network operation and management capabilities. For example, the basic elements that the basic objects of digital twin network operation and management rely on are managed, including but not limited to network status, hardware, software, service capabilities, data, users, security, maturity, etc., and atomic capabilities are extracted to form a set of functional atomic capabilities for digital twin network operation and management, providing basic service capabilities for the full life cycle management of digital twin networks, including network status, hardware, software, service capabilities, etc.

[0037] 2. Requirements for orchestration of digital twin network operation and management capabilities: Orchestrate the basic operation and management capabilities of the digital twin network, such as the network, hardware, software, data, services, and security elements that the digital twin network relies on for operation and management, and provide end-to-end services. Provide management for various capabilities, such as atomic capabilities and end-to-end capabilities, and manage the capabilities after business orchestration by modeling, and provide end-to-end service capabilities.

[0038] 3. Requirements for opening up digital twin network operation and management capabilities: Analyze and maintain digital twin network operation and management intentions from external systems, and match effective end-to-end services. Provide direct external opening of service capabilities to support autonomous orchestration of external applications. Requirements for capability opening: Analyze and maintain digital twin network operation and management intentions from external systems, and provide external opening capabilities.

[0039] refer to Figure 2 As shown in the structural diagram, the structure of the digital twin network operation and management system may include a digital twin network operation and management capability opening layer, a digital twin network operation and management capability orchestration layer, a digital twin network operation and management capability service layer, and system management.

[0040] Among them, the digital twin network operation and management capability open layer includes digital twin network operation and management intention analysis and digital twin network operation and management applications. The digital twin network operation and management capability open layer is used to interact with external systems. For example, it can analyze or parse user intentions or other request instructions input by external systems, etc., and output the processing results obtained through the digital twin network operation and management capability orchestration layer and the digital twin network operation and management capability service layer to the open system.

[0041] The digital twin network operation management capability orchestration layer includes the digital twin network operation management orchestrator and the digital twin network operation management capability knowledge base, which are used to re-arrange capabilities. Capability orchestration refers to the arrangement of all capabilities, for example, multiple capabilities can be combined and orchestrated to achieve unified management of capabilities.

[0042] The digital twin network operation and management capability service layer is used to manage the digital twin network. Specifically, the digital twin network operation and management capability service layer can manage one or more of the network status management, hardware management, software management, data auditing, service capability management, user management, security monitoring, and maturity management of the digital twin network. The network status refers to the network traffic, network speed, etc.

[0043] System management includes permission management, log management and security management, which are used to manage the permissions, logs, security and other dimensions of the digital twin network operation and management capability open layer, digital twin network operation and management capability orchestration layer and digital twin network operation and management capability service layer.

[0044] In order to solve the technical problems existing in the related technologies, the embodiments of the present disclosure may provide a network traffic prediction model in the network status management in the digital twin network operation and management capability service layer, thereby realizing end-to-end network traffic prediction function through the network traffic prediction model, and realizing high-precision prediction of traffic data, and further ensuring that the digital twin network can provide safe, stable and continuous services based on high-precision traffic data, thereby improving the reliability of the digital twin network.

[0045] In some embodiments, the network data to be processed can describe the network status at a preset time, and the preset time can be any time, for example, a certain moment, a certain day, or a certain time period, which is set according to actual needs. The network data to be processed can be time series data of the target network. For example, the network data to be processed can be the network data of the target network on September 16th, which is A; the network data of the target network on September 17th, which is B. Based on this, a traffic prediction algorithm based on a digital twin network operation and management system can be constructed, and the spatiotemporal features between the data in the input time series can be adaptively extracted, so as to achieve high-precision prediction of traffic data. Exemplarily, the prediction of traffic data can be achieved through a network traffic prediction model.

[0046] Figure 3 The structure diagram of the network traffic prediction model is schematically shown in Figure 3 As shown in , the network traffic prediction model may include a time domain feature extraction module, a space domain feature extraction module and a time-space domain fusion module.

[0047] First, the spatial domain feature extraction module is explained. Traditional machine learning models (average autoregression model, Kalman filter, etc.) are more difficult to model nonlinear spatiotemporal data. In order to solve the technical problems existing in traditional machine learning models, based on the structure of the above-mentioned network traffic prediction model, spatial domain features can be extracted from the processed network data to obtain spatial domain features. Exemplarily, feature extraction can be performed on the processed network data through the spatial domain feature extraction module. Specifically, the spatial domain feature extraction module can be a Bi-LSTM (Bi-directional Long Short-Term Memory, bi-phase long short-term memory network), which can not only solve the problem of long-term dependence of time series but also perform parallel computing, and never capture multi-dimensional time series feature correlation information.

[0048] refer to Figure 3 As shown in , the Bi-LSTM module consists of three forward modules and three backward modules, which can use the information before and after the time series to make predictions. Figure 4 As shown in , for the forward module, the network data to be processed can be input in sequence, and each forward module LSTM f Get the forward hidden vector h corresponding to each forward module f0 ,h f1 ,h f2 , thus obtaining the forward hidden vector {h f0 ,h f1 ,h f2 For the backward module, the network data to be processed can be input in reverse order, and each backward module LSTM b Get the backward hidden vector h corresponding to each backward module b0 ,h b1 ,h b2 , and further obtain the backward hidden vector {h b0 ,h b1 ,h b2}. The forward hidden vector and the backward hidden vector can be further fused to obtain a fused vector. Here, the fusion of the forward hidden vector and the backward hidden vector can be understood as a splicing combination, that is, splicing and combining the same data in the network data to be processed, and the following vector {[h f0 ,h b0 ],[h f1 ,h b1 ],[h f2 ,h b2 ]}, the concatenated forward hidden vector and backward hidden vector can be further fused, and the fused vector η = {h0, h1, h2} is used as the spatial domain feature output by the spatial domain feature extraction module.

[0049] In step S320, time domain features are extracted from the network data to be processed to obtain time domain features corresponding to the network data to be processed.

[0050] In addition to obtaining spatial domain features, in order to more accurately predict spatiotemporal data features, the time domain feature extraction module in the network traffic prediction model can be used to extract features from the network data to be processed. In related technologies, the model only uses standard data in Euclidean space. For data in non-Euclidean space, a graph convolution network (GCN) can be used to process it. The graph convolution network converts the network data to be processed from the spatial domain to the spectral domain for processing based on graph spectrum theory and convolution theorem, and performs convolution operations in the frequency domain to obtain time domain features.

[0051] The graph convolution network can include a spectrum-based graph convolution, which maps nodes to the frequency domain space through Fourier transform, realizes the convolution in the time domain by multiplying in the frequency domain space, and finally maps the features after the product back to the time domain space. The spectrum-based graph convolution mainly uses the graph Fourier transform to realize convolution. In addition, the graph convolution network can also include a graph convolution based on the spatial domain, the core of which is to aggregate the information of neighbor nodes. For example, the hidden states of all directly connected neighbor nodes can be summed to update the hidden state of the current node.

[0052] In the implementation of the present disclosure, it is assumed that the network data to be processed is processed into graph data, in which there are N nodes, each node has its own node features, and each node has its own vector representation. The features of these nodes can be set to form an N×D-dimensional matrix X, and then the relationship between each node will also form an N×N-dimensional matrix A, also called an adjacency matrix.

[0053] In some embodiments, when using a graph convolutional network to extract time domain features, the vector of the current layer of the network data to be processed in the graph convolutional network can be determined, and the vector of each node in the current layer can be convolved to obtain a convolution result; the vector of the next layer is updated according to the convolution result, and the vector of each node in the next layer is continued to be convolved to obtain a convolution result to obtain time domain features.

[0054] refer to Figure 3As shown in , the graph convolution network can include multiple layers. The input of the graph convolution network is the entire graph. In the current layer GCN Layer 1, the weighted average of the adjacent nodes of each node is performed to achieve a convolution operation on each node, and the convolution result is used to update the vector of the node in the next layer. The convolution result is then passed through an activation function such as ReLU to complete the processing of the first layer. Next, after another convolution layer GCN Layer 2, the weighted sum of the adjacent nodes of each node in the second layer is performed to perform a convolution operation on each node, and the convolution result is passed through a layer of activation function; and the output convolution result is used to update the vector of the node in the next layer. Repeat the above steps until the number of convolution layers reaches the expected depth, stop the convolution process, and activate the convolution result output by the last layer using the activation function, and obtain the output of the graph convolution network as the time domain feature output by the time domain feature extraction module.

[0055] The graph convolutional network is a multi-layer GCN based on the hierarchical propagation rule. The propagation method between layers is as follows: for each layer of nodes, the vector of the next layer of nodes can be obtained by logically processing the vector of the current layer of nodes, the adjacency matrix, the diagonal matrix, the linear transformation matrix of the current layer, and the activation function. The vector of each node in the next layer can be updated according to the rule expressed by the following formula (1):

[0056]

[0057] in, is a diagonal matrix, H k represents the vector of the kth layer, W k represents the linear transformation matrix of the kth layer, σ is the activation function, is the adjacency matrix and I is the identity matrix.

[0058] In step S130, the time domain features and the spatial domain features are fused to obtain fused features, and multi-scale feature extraction is performed on the fused features to determine the network traffic prediction value corresponding to the network data to be processed, so that the operation management system can perform operation management on the target network based on the network traffic prediction value.

[0059] In the embodiments of the present disclosure, in order to avoid the limitations caused by time domain features and simple spatial domain feature analysis in related technologies, time domain features and spatial domain features can be fused to extract time dependency and spatial dependency in parallel, capture the relationship between local and fused features, effectively fuse time and space features, and improve the prediction accuracy of traffic data.

[0060] In the disclosed embodiment, after the time domain feature extraction module and the space domain feature extraction module, the time domain features extracted by the time domain feature extraction module and the space domain features extracted by the space domain feature extraction module can be fused through the time-space fusion module to obtain a fusion feature including the time domain features and the space domain features. Specifically, the time domain features and the space domain features can be activated respectively through an activation function, and the activated time domain features and the space domain features can be fused to obtain the fusion feature. The fusion here can be splicing.

[0061] Further, refer to Figure 3 As shown in , the spatiotemporal fusion module may include a multi-scale skip connection module. The schematic diagram of the multi-scale convolution module structure is as follows Figure 5 As shown. Based on this, the fused features obtained by fusion can be input into the multi-scale jump connection module in the spatiotemporal fusion module to perform multi-scale feature extraction, thereby determining the output result. Specifically, multi-scale feature extraction can be achieved through convolution kernels of different scales. The convolution kernels of different scales can be convolution kernels of three scales: 1×1, 5×5, and 7×7. Of course, they can also be convolution kernels of other scales, which are not specifically limited here.

[0062] Since the receptive field of the convolution operation is limited, there are certain limitations in extracting time series data using a fixed number of convolution kernels and a fixed convolution scale. The use of a multi-scale convolution structure can extract richer and more comprehensive feature information, thereby improving the prediction accuracy of network traffic based on comprehensive feature information. In the disclosed embodiment, the fusion features of the input multi-scale jump connection module are respectively extracted using convolution kernels of three scales, 1×1, 5×5, and 7×7, to extract features of data at different levels, so that the multi-scale jump connection module extracts hierarchical information between multiple perception domains in parallel, that is, splicing in parallel, and all branch networks are connected in a channel manner, thereby extracting fusion feature information more comprehensively.

[0063] refer to Figure 5As shown in , based on the multi-scale jump connection module, the input fusion features can be extracted using a multi-scale convolution kernel to obtain multi-scale features; further, the extracted multi-scale features can be convolved and activated to obtain convolution results, and the convolution results can be subjected to sliding convolution operations to obtain sliding convolution results. Specifically, after multiple convolutions, the multi-scale features are sequentially input into a network with the same multi-path residual block structure to learn the features of the time series data. Each residual block contains a jump connection structure, and each jump connection structure uses a combination of a global residual self-learning method and a local residual self-learning method to share weights, thereby improving the accuracy of feature extraction. Next, the sliding convolution result can be processed using an activation function to obtain an intermediate convolution result, and the intermediate convolution result can be further processed using an activation function to obtain the network traffic prediction value output by the network traffic prediction model.

[0064] In the disclosed embodiments, multi-scale convolution is performed on the fused features obtained by fusing the time domain features and the frequency domain features through convolution kernels of different scales. Since the time domain features and the spatial domain features are fused, the fused features can be extracted from multiple dimensions, thereby improving the accuracy and comprehensiveness of the features. In addition, convolution can be performed based on multi-scale convolution kernels, thereby increasing the comprehensiveness of the extracted features. By combining the time and spatial domain features, the traffic data features are extracted in multiple dimensions, thereby improving the accuracy of the traffic prediction algorithm, thereby improving the accuracy of the network traffic prediction model in predicting the network traffic.

[0065] In the disclosed embodiment, after obtaining the network traffic prediction value, an operation management service may be provided to the target network corresponding to the operation management system based on the operation management system corresponding to the network traffic prediction value.

[0066] When the operation management system is a digital twin network operation management system and the target network is a digital twin network, the digital twin network operation management system can manage the network status of the digital twin network, and the network status may include but is not limited to one or more of the network connection status, network traffic size, network speed, traffic direction, and network connection duration. In addition, the digital twin network operation management system can also manage any one or more of the hardware, software, data auditing, service capabilities, users, security monitoring, and maturity of the digital twin network. Since the network traffic prediction model in the digital twin network operation management system can accurately predict network traffic, when accurately predicting network traffic, it can improve the management ability of the network status of the digital twin network. Based on accurate traffic prediction, the network status of the digital twin network can be accurately managed, thereby improving the reliability of the digital twin network and making the digital twin network operate efficiently. Since the network traffic can be accurately predicted, the problem can be discovered and handled before the predicted traffic overload is determined, which can improve the availability of the network, reduce the cost of network optimization, and improve the efficiency of operation management.

[0067] The operation and management system based on digital twin network that integrates Bi-LSTM and GCN network traffic prediction proposed in the embodiments of the present disclosure helps operators to make operation and maintenance decisions quickly and accurately, improve scheduling efficiency, support user business development, and achieve efficient operation and maintenance. It provides support for information and communication physical networks to achieve low-cost trial and error, intelligent decision-making, and high-efficiency innovation, and fills the lack of automation, intelligent systematic simulation, optimization, verification, and control capabilities for the realization of intelligent operation of information and communication networks.

[0068] Based on the structure of the digital twin network operation management system, the interaction process between the external system and the digital twin network can be realized based on the digital twin network operation management system. In some embodiments, Figure 6 The overall schematic diagram of the interaction process is shown in Figure 6 As shown in , it mainly includes the following steps:

[0069] Step S602: the external system receives the user's intention and outputs an expression of the user's intention.

[0070] Step S604: The digital twin network operation and management system analyzes the user's intention and sends the user's intention to the capability exposure layer of the digital twin network operation and management system. The specific process includes the following:

[0071] Step S606: Perform user intent analysis in the capability exposure layer of the digital twin network operation and management system. The digital twin operation and management capability exposure layer is responsible for interacting with external systems, parsing digital twin network operation and management intentions from external systems, and providing required digital twin network operation and management applications.

[0072] Step S608, determine whether the analysis is successful. If yes, go to step S5612; if not, go to step S610 and go to step S602.

[0073] Step S610: The capability orchestration layer manages the capability provisioning and manages the capabilities after business orchestration to provide end-to-end service capabilities.

[0074] Step S612: If the analysis is successful, the successfully analyzed user intent is shared with the capability service layer, so that the capability service layer can provide external capability services.

[0075] Step S614: Output the capability call request to the digital twin network.

[0076] Step S616, determine whether a matching request is found. If yes, go to step S618; if no, go to step S620.

[0077] Step S618: The capability call request is sent to the digital twin network so that the digital twin network obtains the capability call result and proceeds to step S612.

[0078] Step S620, return to the user.

[0079] In the disclosed embodiment, the digital twin network operation and management system can parse the user intent input by the external system, and input it into the digital twin network to obtain the corresponding capability call result, thereby realizing the interaction between the external system and the digital twin network operation and management system and the digital twin network. The digital twin network operation and management system provides operation and management services to maintain the healthy, stable and continuous operation of the digital twin network. The digital twin network operation and management system provides full life cycle operation and management services for the digital twin network from the dimensions of network status, hardware, software, data, service capabilities, users, security and maturity, and provides digital twin network operation and management capability services, capability orchestration and capability exposure functions, which improves the reliability of the service.

[0080] The present disclosure also provides a network operation management system. Figure 7 As shown, the network operation management system 700 may include:

[0081] An external system 701 and a target network 702; an operation management system 703, which is used to extract spatial domain features from the network data to be processed to obtain corresponding airspace features; extract time domain features from the network data to be processed to obtain time domain features; fuse the time domain features and the airspace features to obtain fused features, and perform multi-scale feature extraction on the fused features to determine the network traffic prediction value corresponding to the network data to be processed, so as to perform operation management on the target network based on the network traffic prediction value.

[0082] In an exemplary embodiment of the present disclosure, the extracting spatial domain features of the network data to be processed to obtain corresponding spatial domain features includes: obtaining a forward hidden vector of the network data to be processed, and determining a backward hidden vector of the network data to be processed; fusing the forward hidden vector and the backward hidden vector to obtain the spatial domain features.

[0083] In an exemplary embodiment of the present disclosure, the fusing of the forward hidden vector and the backward hidden vector to obtain the spatial domain feature includes: concatenating and combining the forward hidden vector and the backward hidden vector to determine the spatial domain vector.

[0084] In an exemplary embodiment of the present disclosure, extracting time domain features from the network data to be processed to obtain the time domain features includes: converting the network data to be processed into the frequency domain, and performing a convolution operation on the network data to be processed in the frequency domain to obtain the time domain features.

[0085] In an exemplary embodiment of the present disclosure, the convolution operation is performed on the network data to be processed in the frequency domain to obtain the time domain features, including: determining the vector of the current layer of the network data to be processed in the graph convolution network, and performing a convolution operation on the vector of each node in the current layer to obtain a convolution result; updating the vector of each node in the next layer according to the convolution result, and continuing to perform a convolution operation on the vector of each node in the next layer to obtain a convolution result, until the expected number of layers is reached, so as to obtain the time domain features.

[0086] In an exemplary embodiment of the present disclosure, the multi-scale feature extraction is performed on the fused features to determine the network traffic prediction value corresponding to the network data to be processed, including: performing multi-scale feature extraction on the fused features respectively to obtain multi-scale features at different levels; performing a convolution operation on the multi-scale features to determine the network traffic prediction value.

[0087] In an exemplary embodiment of the present disclosure, the convolution operation is performed on the multi-scale features to determine the network traffic prediction value, including: performing a convolution operation on the multi-scale features based on a multi-scale jump connection module, and processing the multi-scale convolution result through an activation function to obtain the network traffic prediction value.

[0088] In an exemplary embodiment of the present disclosure, the operational management of the target network based on the network traffic prediction value includes: managing the network status of the target network based on the network traffic prediction value.

[0089] It should be noted that the specific details of each module in the above network operation management system have been described in detail in the corresponding network operation management method, so they will not be repeated here.

[0090] It should be noted that, although several modules or units of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more modules or units described above can be embodied in one module or unit. On the contrary, the features and functions of one module or unit described above can be further divided into multiple modules or units to be embodied.

[0091] In addition, although the steps of the method in the present disclosure are described in a specific order in the drawings, this does not require or imply that the steps must be performed in this specific order, or that all the steps shown must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps, etc.

[0092] In an exemplary embodiment of the present disclosure, an electronic device capable of implementing the above method is also provided.

[0093] Those skilled in the art will appreciate that various aspects of the present disclosure may be implemented as systems, methods or program products. Therefore, various aspects of the present disclosure may be specifically implemented in the following forms, namely: complete hardware implementation, complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software, which may be collectively referred to herein as "circuits", "modules" or "systems".

[0094] Refer to the following Figure 8 The electronic device 800 according to this embodiment of the present disclosure is described. Figure 8 The electronic device 800 shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0095] like Figure 8 As shown, the electronic device 800 is in the form of a general computing device. The components of the electronic device 800 may include, but are not limited to: the at least one processing unit 810, the at least one storage unit 820, a bus 830 connecting different system components (including the storage unit 820 and the processing unit 810), and a display unit 840.

[0096] The storage unit stores program codes, which can be executed by the processing unit 810, so that the processing unit 810 performs the steps according to various exemplary embodiments of the present disclosure described in the above “Exemplary Method” section of this specification. For example, the processing unit 810 can perform the following steps: Figure 1 Follow the steps shown in .

[0097] The storage unit 820 may include a readable medium in the form of a volatile storage unit, such as a random access memory unit (RAM) 8201 and / or a cache memory unit 8202 , and may further include a read-only memory unit (ROM) 8203 .

[0098] The storage unit 820 may also include a program / utility 8204 having a set (at least one) of program modules 8205, such program modules 8205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment.

[0099] Bus 830 may represent one or more of several types of bus structures, including a memory unit bus or memory unit controller, a peripheral bus, an accelerated graphics port, a processing unit, or a local bus using any of a variety of bus architectures.

[0100] The electronic device 800 may also communicate with one or more external devices 900 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 800, and / or communicate with any device that enables the electronic device 800 to communicate with one or more other computing devices (e.g., routers, modems, etc.). Such communication may be performed via an input / output (I / O) interface 850. Furthermore, the electronic device 800 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 860. As shown, the network adapter 860 communicates with other modules of the electronic device 800 via a bus 830. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the electronic device 800, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0101] Through the description of the above implementation, it is easy for those skilled in the art to understand that the example implementation described here can be implemented by software, or by software combined with necessary hardware. Therefore, the technical solution according to the implementation of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (which can be a personal computer, a server, a terminal device, or an electronic device, etc.) to execute the method according to the implementation of the present disclosure.

[0102] In an exemplary embodiment of the present disclosure, a computer-readable storage medium is also provided, on which a program product capable of implementing the above method of the present specification is stored. In some possible implementations, various aspects of the present disclosure may also be implemented in the form of a program product, which includes a program code, and when the program product is run on a terminal device, the program code is used to enable the terminal device to execute the steps according to various exemplary implementations of the present disclosure described in the above "Exemplary Method" section of the present specification.

[0103] According to the program product for implementing the above method in the embodiment of the present disclosure, it can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be run on a terminal device, such as a personal computer. However, the program product of the present disclosure is not limited thereto. In this document, a readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, an apparatus or a device.

[0104] The program product may use any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0105] Computer readable signal media may include data signals propagated in baseband or as part of a carrier wave, in which readable program code is carried. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Readable signal media may also be any readable medium other than a readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0106] The program code embodied on the readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.

[0107] Program code for performing the operations of the present disclosure may be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving a remote computing device, the remote computing device may be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computing device (e.g., through the Internet using an Internet service provider).

[0108] In addition, the above-mentioned figures are only schematic illustrations of the processes included in the method according to the exemplary embodiments of the present disclosure, and are not intended to be limiting. It is easy to understand that the processes shown in the above-mentioned figures do not indicate or limit the time sequence of these processes. In addition, it is also easy to understand that these processes can be performed synchronously or asynchronously, for example, in multiple modules.

[0109] Other embodiments of the present disclosure will be readily apparent to those skilled in the art after considering the specification and practicing the inventions invented herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not invented by the present disclosure. The specification and examples are to be considered merely exemplary, and the true scope and spirit of the present disclosure are indicated by the claims.

Claims

1. A network operation management method, characterized in that: include: Acquire the network data to be processed, and extract spatial domain features from the network data to be processed to obtain corresponding spatial domain features; Extracting time domain features from the network data to be processed to obtain time domain features corresponding to the network data to be processed; The time domain features and the spatial domain features are fused to obtain fused features, and multi-scale feature extraction is performed on the fused features to determine the network traffic prediction value corresponding to the network data to be processed, so that the operation management system can perform operation management on the target network based on the network traffic prediction value.

2. The network operation management method according to claim 1, characterized in that: The extracting spatial domain features of the network data to be processed to obtain corresponding spatial domain features includes: Acquire a forward hidden vector of the network data to be processed, and determine a backward hidden vector of the network data to be processed; The forward hidden vector and the backward hidden vector are fused to obtain the spatial domain feature.

3. The network operation management method according to claim 2, characterized in that: The fusing the forward hidden vector and the backward hidden vector to obtain the spatial domain feature includes: The forward hidden vector and the backward hidden vector are concatenated and combined to determine the spatial domain vector.

4. The network operation management method according to claim 1, characterized in that: The time domain feature extraction of the network data to be processed to obtain the time domain feature includes: The network data to be processed is converted into a frequency domain, and a convolution operation is performed on the network data to be processed in the frequency domain to obtain the time domain features.

5. The network operation management method according to claim 4, characterized in that: The performing a convolution operation on the to-be-processed network data in the frequency domain to obtain the time domain feature includes: Determine the vector of the current layer of the network data to be processed in the graph convolutional network, and perform a convolution operation on the vector of each node in the current layer to obtain the convolution result; According to the convolution result, the vector of each node in the next layer is updated, and the vector of each node in the next layer is continued to be convolved to obtain the convolution result until the expected number of layers is reached to obtain the time domain features.

6. The network operation management method according to claim 1, characterized in that: The performing multi-scale feature extraction on the fused features to determine the network traffic prediction value corresponding to the network data to be processed includes: Perform multi-scale feature extraction on the fused features to obtain multi-scale features at different levels; A convolution operation is performed on the multi-scale features to determine the network traffic prediction value.

7. The network operation management method according to claim 6, characterized in that: The performing a convolution operation on the multi-scale features to determine the network traffic prediction value includes: The multi-scale features are convolved based on a multi-scale jump connection module, and the multi-scale convolution results are processed through an activation function to obtain the network traffic prediction value.

8. The network operation management method according to claim 1, characterized in that: The operating and managing the target network based on the network traffic prediction value includes: The network status of the target network is managed based on the network traffic prediction value.

9. A network operation management system, characterized in that: include: External systems and target networks; An operation management system is used to extract spatial domain features from the network data to be processed to obtain corresponding airspace features; extract time domain features from the network data to be processed to obtain time domain features; fuse the time domain features and the airspace features to obtain fused features, and perform multi-scale feature extraction on the fused features to determine the network traffic prediction value corresponding to the network data to be processed, so as to perform operation management on the target network based on the network traffic prediction value.

10. An electronic device, characterized in that: include: processor; as well as A memory, configured to store executable instructions of the processor; The processor is configured to execute the network operation management method according to any one of claims 1 to 8 by executing the executable instructions.

Images