Metod and device for predicting link capacity in metropolitan area network, equipment and storage medium

Abstract

The present disclosure provides a metropolitan area network link capacity prediction method and device, equipment and storage medium, relating to the technical field of artificial intelligence. In some embodiments of the present disclosure, the type of metropolitan area network application scenario is obtained; according to the type of metropolitan area network application scenario, the link capacity prediction model corresponding to the type of metropolitan area network application scenario is used to predict the link capacity of the metropolitan area network, and the link capacity of the metropolitan area network corresponding to the type of metropolitan area network application scenario is obtained; according to the link capacity of the metropolitan area network, the work order distribution operation is performed; the present disclosure adopts the corresponding link capacity prediction model to predict the link capacity of the metropolitan area network for different metropolitan area network application scenarios, and improves the accuracy of the prediction of the link capacity of the metropolitan area network.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of artificial intelligence, in particular to a metropolitan area network link capacity prediction method and device, equipment and storage medium. BACKGROUND

[0002] In the aspect of metropolitan area network management, trend prediction plays a crucial role in many application scenarios, such as predicting abnormal traffic links, knowing the relevant index values in advance, and then assigning work orders to achieve accurate regulation of resources. With the development of artificial intelligence technology, AI prediction algorithms are constantly improving, and applying prediction models to network intelligent operation and maintenance is a hot research topic.

[0003] At present, the prediction accuracy of metropolitan area network management link capacity is low. SUMMARY

[0004] The present disclosure provides a metropolitan area network link capacity prediction method, device, equipment and storage medium to at least solve the problem of low prediction accuracy of existing metropolitan area network management link capacity.

[0005] The technical solution of the present disclosure is as follows:

[0006] The present disclosure provides a metropolitan area network link capacity prediction method, which comprises:

[0007] Obtaining a metropolitan area network application scenario type;

[0008] According to the metropolitan area network application scenario type, a link capacity prediction model corresponding to the metropolitan area network application scenario type is used to predict the metropolitan area network link capacity, and the metropolitan area network link capacity corresponding to the metropolitan area network application scenario type is obtained.

[0009] According to the metropolitan area network link capacity, a work order assignment operation is performed.

[0010] Optionally, if the metropolitan area network application scenario type is a city-level overall trend application scenario, the step of predicting the metropolitan area network link capacity according to the metropolitan area network application scenario type and using the link capacity prediction model corresponding to the metropolitan area network application scenario type comprises:

[0011] According to the city-level overall trend application scenario, a long-period prediction model corresponding to the city-level overall trend application scenario is used to predict the metropolitan area network link capacity, and the metropolitan area network link capacity corresponding to the city-level overall trend application scenario is obtained.

[0012] Optionally, if the metropolitan area network application scenario type is a single-cell single-index application scenario; the link capacity prediction of the metropolitan area network according to the metropolitan area network application scenario type, using the link capacity prediction model corresponding to the metropolitan area network application scenario type, obtaining the metropolitan area network link capacity corresponding to the metropolitan area network application scenario type, comprises:

[0013] According to the single-cell single-index application scenario, the link capacity prediction of the metropolitan area network is performed using the deep learning time series prediction model corresponding to the single-cell single-index application scenario, and the metropolitan area network link capacity corresponding to the single-cell single-index application scenario is obtained.

[0014] Optionally, if the metropolitan area network application scenario type is a single-link application scenario; the link capacity prediction of the metropolitan area network according to the metropolitan area network application scenario type, using the link capacity prediction model corresponding to the metropolitan area network application scenario type, obtaining the metropolitan area network link capacity corresponding to the metropolitan area network application scenario type, comprises:

[0015] According to the single-link application scenario, the link capacity prediction of the metropolitan area network is performed using the time series neural network fusion model corresponding to the single-link application scenario, and the metropolitan area network link capacity corresponding to the single-link application scenario is obtained.

[0016] Optionally, the time series neural network fusion model comprises a combination prediction model, a short-period prediction model and a machine learning prediction model.

[0017] Optionally, the combination prediction model is an XGBoost prediction model, the short-period prediction model is an LSTM model, and the machine learning prediction model is an RNN model.

[0018] The embodiments of the present disclosure also provide a metropolitan area network link capacity prediction device, comprising:

[0019] The acquisition module acquires a metropolitan area network application scenario type;

[0020] The prediction module performs the link capacity prediction of the metropolitan area network according to the metropolitan area network application scenario type, using the link capacity prediction model corresponding to the metropolitan area network application scenario type, and obtains the metropolitan area network link capacity corresponding to the metropolitan area network application scenario type;

[0021] The dispatch module performs a work order dispatch operation according to the metropolitan area network link capacity.

[0022] Optionally, if the metropolitan area network application scenario type is a city-level overall trend application scenario; when the prediction module performs the link capacity prediction of the metropolitan area network according to the metropolitan area network application scenario type, using the link capacity prediction model corresponding to the metropolitan area network application scenario type, and obtains the metropolitan area network link capacity corresponding to the metropolitan area network application scenario type, it is used for:

[0023] According to the city-level overall trend application scenario, a long-period prediction model corresponding to the city-level overall trend application scenario is used to perform metro link capacity prediction, to obtain metro link capacity corresponding to the city-level overall trend application scenario.

[0024] Optionally, if the metro application scenario type is a single-cell single-index application scenario, the prediction module, when performing metro link capacity prediction according to the metro application scenario type by using a link capacity prediction model corresponding to the metro application scenario type, is configured to:

[0025] According to the single-cell single-index application scenario, a deep learning time series prediction model corresponding to the single-cell single-index application scenario is used to perform metro link capacity prediction, to obtain metro link capacity corresponding to the single-cell single-index application scenario.

[0026] Optionally, if the metro application scenario type is a single-link application scenario, the prediction module, when performing metro link capacity prediction according to the metro application scenario type by using a link capacity prediction model corresponding to the metro application scenario type, is configured to:

[0027] According to the single-link application scenario, a time series neural network fusion model corresponding to the single-link application scenario is used to perform metro link capacity prediction, to obtain metro link capacity corresponding to the single-link application scenario.

[0028] Optionally, the time series neural network fusion model includes a combination prediction model, a short-period prediction model, and a machine learning prediction model.

[0029] Optionally, the combination prediction model is an XGBoost prediction model, the short-period prediction model is an LSTM model, and the machine learning prediction model is an RNN model.

[0030] The embodiments of the present disclosure further provide an electronic device, including:

[0031] a processor;

[0032] a memory for storing instructions executable by the processor;

[0033] The processor is configured to execute the instructions to implement each step in the above method.

[0034] The embodiments of the present disclosure further provide a computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, implements each step in the above method.

[0035] The embodiments of the present disclosure also provide a computer program product comprising computer programs / instructions which, when executed by a processor, implement the steps of the above method.

[0036] The technical solutions provided by the embodiments of the present disclosure at least have the following beneficial effects:

[0037] In some embodiments of the present disclosure, a metropolitan area network application scenario type is acquired; according to the metropolitan area network application scenario type, a metropolitan area network link capacity prediction is performed by using a link capacity prediction model corresponding to the metropolitan area network application scenario type to obtain a metropolitan area network link capacity corresponding to the metropolitan area network application scenario type; and according to the metropolitan area network link capacity, a work order dispatching operation is performed; the present disclosure adopts a corresponding link capacity prediction model to perform a metropolitan area network link capacity prediction for different metropolitan area network application scenarios, thereby improving the prediction accuracy of the metropolitan area network management link capacity.

[0038] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0039] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure, and do not constitute an undue limitation on the present disclosure.

[0040] Figure 1 A flowchart of a metropolitan area network link capacity prediction method provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0041] Figure 2 A structure diagram of a combined prediction model provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0042] Figure 3 A structure diagram of an LSTM model provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0043] Figure 4 A time line expansion diagram of an RNN model provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0044] Figure 5 A schematic diagram of a LightGBM model provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0045] Figure 6 A structure diagram of a metropolitan area network link capacity prediction device provided for an exemplary embodiment of the present disclosure is shown in the figure;

[0046] Figure 7 A structure diagram of an electronic device provided for an exemplary embodiment of the present disclosure is shown in the figure. DETAILED DESCRIPTION

[0047] In order for those skilled in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings.

[0048] It should be noted that the terms "first", "second" and the like in the specification and claims of the present disclosure and the above drawings are used to distinguish similar objects, and do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Rather, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0049] It should be noted that the user information involved in the present disclosure includes but is not limited to user equipment information and user personal information; the collection, storage, use, processing, transmission, provision and disclosure of user information in the present disclosure comply with the provisions of relevant laws and regulations, and do not violate public order and good customs.

[0050] In the management of metropolitan area networks, trend prediction plays a crucial role in many application scenarios, such as predicting abnormal traffic links, knowing the relevant index values in advance, and then assigning work orders to achieve accurate regulation of resources; with the development of artificial intelligence technology, AI prediction algorithms are constantly improving, and applying prediction models to network intelligent operation and maintenance is a hot research topic.

[0051] Currently, the prediction accuracy of metropolitan network management link capacity is low.

[0052] In order to solve the above technical problems, in some embodiments of the present disclosure, the type of metropolitan area network application scenario is obtained; according to the type of metropolitan area network application scenario, the link capacity prediction model corresponding to the type of metropolitan area network application scenario is used to predict the link capacity of the metropolitan area network, and the link capacity of the metropolitan area network corresponding to the type of metropolitan area network application scenario is obtained; according to the link capacity of the metropolitan area network, a work order assignment operation is performed; the present disclosure uses the corresponding link capacity prediction model to predict the link capacity of the metropolitan area network for different metropolitan area network application scenarios, thereby improving the prediction accuracy of the metropolitan network management link capacity.

[0053] The technical solutions provided by the embodiments of the present disclosure will be described in detail below in conjunction with the drawings.

[0054] Figure 1 A flowchart of a metropolitan area network link capacity prediction method provided by an exemplary embodiment of the present disclosure is shown in FIG. Figure 1 As shown in the figure, the method comprises:

[0055] S101: Obtain a type of a metropolitan area network application scenario.

[0056] S102: According to the type of the metropolitan area network application scenario, perform metropolitan area network link capacity prediction by using a link capacity prediction model corresponding to the type of the metropolitan area network application scenario, to obtain metropolitan area network link capacity corresponding to the type of the metropolitan area network application scenario.

[0057] S103: Perform a work order distribution operation according to the metropolitan area network link capacity.

[0058] In the embodiment, the execution subject of the method can be a terminal device or a server.

[0059] The terminal device includes but is not limited to a mobile station (MS), a mobile terminal, a mobile telephone, a handset, a portable equipment, and the like. The terminal device can communicate with one or more core networks through a radio access network (RAN). For example, the terminal device can be a mobile phone (also referred to as a "cellular" phone), a computer with wireless communication function, and the like. The terminal device can also be a computer with wireless transceiver function, a virtual reality (VR) terminal device, an AR terminal device, a wireless terminal in industrial control, a wireless terminal in self driving, a wireless terminal in remote medical treatment, a wireless terminal in smart grid, a wireless terminal in transportation safety, a wireless terminal in smart city, a wireless terminal in smart home, and the like. An operating system installed on the terminal device includes but is not limited to an IOS, an Android, a windows, a linux, a Mac OS, and the like. In different networks, the terminal can be called by different names, such as a user equipment, a mobile station, a user unit, a station, a cellular phone, a personal digital assistant, a wireless modem, a wireless communication device, a handheld device, a laptop, a cordless phone, a wireless local loop station, a television, and the like. For the convenience of description, the terminal device is referred to as a terminal device in the embodiment.

[0060] In the embodiment, the implementation form of the server is not limited. For example, the server can be a conventional server, a cloud server, a cloud host, a virtual center, or the like. The server mainly includes a processor, a hard disk, a memory, a system bus, and the like, and a general computer architecture type.

[0061] In the embodiment, a metropolitan area network application scenario type is acquired. According to the metropolitan area network application scenario type, a link capacity prediction model corresponding to the metropolitan area network application scenario type is used to perform metropolitan area network link capacity prediction, to obtain metropolitan area network link capacity corresponding to the metropolitan area network application scenario type. According to the metropolitan area network link capacity, a work order dispatching operation is performed. The present disclosure uses a corresponding link capacity prediction model to perform metropolitan area network link capacity prediction for different metropolitan area network application scenarios, to improve the accuracy of metropolitan area network management link capacity prediction.

[0062] In the embodiment, different link capacity prediction models are used to perform metropolitan area network link capacity prediction for different metropolitan area network application scenario types, which can improve the accuracy of metropolitan area network management link capacity prediction. The metropolitan area network application scenario type includes but is not limited to a city-level overall trend application scenario, a single-cell single-index application scenario, and a single-link application scenario. Correspondingly, the link capacity prediction model corresponding to the city-level overall trend application scenario is a long-period prediction model; the link capacity prediction model corresponding to the single-cell single-index application scenario is a deep learning time series prediction model; and the link capacity prediction model corresponding to the single-link application scenario is a time series neural network fusion model.

[0063] In some embodiments of the present disclosure, according to the city-level overall trend application scenario, a long-period prediction model corresponding to the city-level overall trend application scenario is used to perform metropolitan area network link capacity prediction, to obtain metropolitan area network link capacity corresponding to the city-level overall trend application scenario. The long-period prediction model is a Prophet model.

[0064] In other embodiments of the present disclosure, according to the single-cell single-index application scenario, a deep learning time series prediction model corresponding to the single-cell single-index application scenario is used to perform metropolitan area network link capacity prediction, to obtain metropolitan area network link capacity corresponding to the single-cell single-index application scenario. The deep learning time series prediction model is a LightGBM model.

[0065] In other embodiments of the present disclosure, according to the single-link application scenario, a time series neural network fusion model corresponding to the single-link application scenario is used to perform metropolitan area network link capacity prediction, to obtain metropolitan area network link capacity corresponding to the single-link application scenario. The time series neural network fusion model includes a combination prediction model, a short-period prediction model, and a machine learning prediction model. The combination prediction model is an XGBoost prediction model, the short-period prediction model is an LSTM model, and the machine learning prediction model is an RNN model.

[0066] The city area network link capacity prediction method of the present disclosure is described below in combination with specific embodiments.

[0067] I. Data collection

[0068] The historical actual data of the link capacity (peak traffic, peak bandwidth utilization) and the number of users (peak online user number) of 1000 network elements (BNG, OLT, PON port) in the city within half a year are called, and the data is cleaned. Perform outlier processing, remove links with outliers, remove invalid data to ensure data availability; perform missing value processing, remove links with time series data missing rate exceeding 70%; for links with missing rate less than 70%, use adjacent time values to fill in missing values. After cleaning the data, standardize the data type, and record it as a historical database in "day granularity", "hour granularity", "15 minute granularity", "province and city", "district", "network element", "link".

[0069] II. Prediction model construction

[0070] Through correlation filtering, variance filtering, feature selection, etc. Feature selection methods, respectively, for Prophet model, LightGBM model, XGBOOST prediction model, RNN model, LSTM model, these five groups of model construction data set, for the integrated learning of each model.

[0071] 1. Long-period prediction model.

[0072] Prophet model principle: y(t) = g(t) + s(t) + h(t) + ∈ t

[0073] The model consists of three parts: growth (trend), seasonality (seasonal trend), and holidays (the impact of holidays).

[0074] g(t) represents the trend item, which is used to fit the changing trend of the time series on the non-periodic surface;

[0075] s(t) represents the periodic term (seasonal term), such as every week, every year, etc.

[0076] h(t) represents the holiday term, which indicates whether there is a holiday or a sudden event on the day;

[0077] ε(t) is the error term, which represents the error not considered by the model.

[0078] The model is aimed at "wide range and large span", "holiday or activity day business mutation", based on long-term historical data, establishes a Prophet additive prediction mode, predicts the business changes of prefecture-level cities and holiday business mutation detection, and is used for regional long-period prediction.

[0079] 2. Combined prediction model.

[0080] The XGBoost prediction model is mainly used to solve the problems of GBDT that cannot converge and cannot be parallel, is based on time characteristics, can effectively learn time trend and fluctuation, and is suitable for medium and short period index prediction.

[0081] The SVR model is a support vector regression (SVR) model applied to regression problems by support vector machines. For a nonlinear model, a kernel function is used to map to a feature space, and then regression fitting is performed.

[0082] Figure 2 A structural diagram of a combined prediction model provided by an exemplary embodiment of the present disclosure is shown in FIG. 1. As shown in FIG. 1, the model is sensitive to time series recent data based on XGBoost and can receive multiple index inputs, constructs an XGBoost regression tree model, and uses SVM to fit and predict the error sequence, solving the problem that XGBoost will introduce prediction errors in multi-step time prediction. Figure 2

[0083] 3. Short-period prediction model

[0084] Figure 3 A structural diagram of an LSTM model provided by an exemplary embodiment of the present disclosure is shown in FIG. 2. As shown in FIG. 2, the LSTM model is an optimized improvement on RNN, adding three logic gates to control the saving or discarding of historical information, ensuring that the current iteration time cell information processing and the information of the past time interval have strong time correlation, so that LSTM can process long-period sequences. Figure 3

[0085] 4. Machine learning prediction model

[0086] Principle: Recurrent Neural Networks (RNN), RNN generally takes sequence data as input, and can effectively capture the relationship between sequences through the structure design of the network. It is generally output in sequence form. Recurrent neural network is a recognized sequence model suitable for sequence data.

[0087] Parameters:

[0088] input_size: the size of the feature dimension in the input tensor x.

[0089] ​​hidden_size: size of the feature dimension in the hidden layer tensor h.

[0090] num_layers: number of hidden layers.

[0091] nonlinearity: selection of the activation function, default is tanh.

[0092] Figure 4 An RNN model timeline expansion diagram is provided for an exemplary embodiment of the present disclosure. As shown in the diagram, the model establishes a recurrent neural network prediction model based on short-term historical data for "multiple inputs", "short periodicity", and predicts the link capacity trend in the short term at the prefecture level, which is used for regional short-period prediction. Figure 4

[0093] 5, deep learning time series prediction model.

[0094] Figure 5 A schematic diagram of a LightGBM model is provided for an exemplary embodiment of the present disclosure. As shown in the diagram, LightGBM is mainly proposed to solve the problems encountered by GBDT in massive data, based on time characteristics, which can effectively learn the time series trend and fluctuation, and Hyperopt hyperparameter self-optimization, suitable for medium and short-term index prediction. Figure 5

[0095] Parameters:

[0096] num_iterations: number of iterations, self-optimization range 50 to 300 times;

[0097] num_leaf: number of leaf nodes, self-optimization range 5 to 100;

[0098] learning_rate: learning rate, self-optimization range 0 to 1;

[0099] max_depth: maximum depth of the tree, selected as 3, 4, 5.

[0100] Three, model fusion prediction.

[0101] Combine model characteristics and data granularity to match application scenarios: match Prophet model for prefecture-level overall trend application scenarios; match LightGBM model for single cell and single index application scenarios; match time series neural network fusion model based on time series neural network fusion formed by combination of XGBoost model, RNN model and LSTM model for single link application scenarios.

[0102] 1, based on link feature data, train, validate and use the above-mentioned constructed XGBoost model, RNN model and LSTM model for prediction. ​​

[0103] 2, call history database link feature data, extraction link timing data features in peak bandwidth utilization and peak online user number two characteristic data. Respectively, under the same feature conditions "feature data", "feature data combination" and "feature data trend", the three models are compared with the prediction results, respectively, to predict the most accurate model as the prediction model of the feature condition to predict.

[0104] (1) Feature data

[0105] Peak bandwidth utilization corresponding to the prediction model. As shown in Table 1 below.

[0106] Peak bandwidth utilization (%) Optimal model ＜70 XGBoost 70-85 LSTM model 85-90 RNN model 90-93 XGBoost ＞93 LSTM model

[0107] Table 1 peak online user number corresponding to the prediction model. As shown in Table 2 below.

[0108] Peak online user number (ten thousand) Optimal model ＜120 RNN model 120-160 LSTM model 160-200 RNN model 200-250 XGBoost 250-300 XGBoost

[0109] Table 2 (2) feature data combination. As shown in Table 3 below.

[0110]

[0111]

[0112] Table 3

[0113] (3) Feature data trend

[0114] The feature data trend in this disclosure is mainly for predicting the daily variation rate of peak bandwidth utilization. As shown in Table 4 below.

[0115]

[0116] Table 4

[0117] Four, model iteration

[0118] According to the characteristics of data seasonality, periodicity, suddenness, etc., the model is corrected and iterated, and the feature condition data is changed to continue training.

[0119] The present disclosure firstly trains five groups of models of Prophet, LightGBM, SVM+XGBOOST, RNN and LSTM through integrated learning; then matches the application scenarios according to the model characteristics and data granularity: matches the Prophet model for the overall trend application scenario of the city level; matches the LightGBM model for the single-cell single-index application scenario; then combines the XGBoost model, the RNN model and the LSTM model to form a link capacity prediction model based on the fusion of time series neural networks in the single-link application scenario. The time series data features of each link are fully learned, and the time series prediction model of "precise prediction according to needs" is constructed according to the model characteristics and data granularity, and a flexible, efficient, iterative and adaptable integrated learning and time series neural network fusion metropolitan area network link capacity prediction solution is established.

[0120] The present disclosure trains, verifies and predicts the XGBoost model, the RNN model and the LSTM model based on link feature data; extracts the peak bandwidth utilization and the peak online user number in the link time series data features in the link feature data, compares the prediction results of the three models under the same feature conditions such as feature data, feature data combination or feature data trend, and uses the model with the most accurate prediction result as the prediction model for prediction. The present disclosure explores the AI time series prediction algorithm, comprehensively compares and analyzes, establishes multiple prediction model schemes, adapts to various scenes of "wide range and large span", "single link and short period", "multiple indexes and medium-short period", matches the model characteristics and data granularity, and constructs the time series prediction model of "precise prediction according to needs".

[0121] In the above method embodiment of the present disclosure, the type of the metropolitan area network application scenario is obtained; according to the type of the metropolitan area network application scenario, the link capacity prediction model corresponding to the type of the metropolitan area network application scenario is used to predict the link capacity of the metropolitan area network, and the link capacity of the metropolitan area network corresponding to the type of the metropolitan area network application scenario is obtained; according to the link capacity of the metropolitan area network, the work order dispatching operation is performed; the present disclosure uses the corresponding link capacity prediction model to predict the link capacity of the metropolitan area network for different metropolitan area network application scenarios, and improves the prediction accuracy of the link capacity of the metropolitan area network management.

[0122] Figure 6 A structure diagram of a metropolitan area network link capacity prediction device 60 provided for an exemplary embodiment of the present disclosure is shown in FIG. 6. As shown in FIG. 6, the metropolitan area network link capacity prediction device 60 includes an acquisition module 61, a prediction module 62 and a dispatching module 63. Figure 6

[0123] The acquisition module 61 acquires the type of the metropolitan area network application scenario.

[0124] ​The prediction module 62 predicts the link capacity of the metropolitan area network according to the metropolitan area network application scenario type by using the link capacity prediction model corresponding to the metropolitan area network application scenario type, to obtain the link capacity of the metropolitan area network corresponding to the metropolitan area network application scenario type.

[0125] The dispatch module 63 performs work order dispatching operation according to the link capacity of the metropolitan area network.

[0126] Optionally, if the metropolitan area network application scenario type is a city-level overall trend application scenario, when the prediction module 62 predicts the link capacity of the metropolitan area network according to the metropolitan area network application scenario type by using the link capacity prediction model corresponding to the metropolitan area network application scenario type, to obtain the link capacity of the metropolitan area network corresponding to the metropolitan area network application scenario type, it is used for:

[0127] According to the city-level overall trend application scenario, the long-period prediction model corresponding to the city-level overall trend application scenario is used to predict the link capacity of the metropolitan area network, to obtain the link capacity of the metropolitan area network corresponding to the city-level overall trend application scenario.

[0128] Optionally, if the metropolitan area network application scenario type is a single-cell single-index application scenario, when the prediction module 62 predicts the link capacity of the metropolitan area network according to the metropolitan area network application scenario type by using the link capacity prediction model corresponding to the metropolitan area network application scenario type, to obtain the link capacity of the metropolitan area network corresponding to the metropolitan area network application scenario type, it is used for:

[0129] According to the single-cell single-index application scenario, the deep learning time series prediction model corresponding to the single-cell single-index application scenario is used to predict the link capacity of the metropolitan area network, to obtain the link capacity of the metropolitan area network corresponding to the single-cell single-index application scenario.

[0130] Optionally, if the metropolitan area network application scenario type is a single-link application scenario, when the prediction module 62 predicts the link capacity of the metropolitan area network according to the metropolitan area network application scenario type by using the link capacity prediction model corresponding to the metropolitan area network application scenario type, to obtain the link capacity of the metropolitan area network corresponding to the metropolitan area network application scenario type, it is used for:

[0131] According to the single-link application scenario, the time series neural network fusion model corresponding to the single-link application scenario is used to predict the link capacity of the metropolitan area network, to obtain the link capacity of the metropolitan area network corresponding to the single-link application scenario.

[0132] Optionally, the time series neural network fusion model includes a combination prediction model, a short-period prediction model, and a machine learning prediction model.

[0133] Optionally, the combination prediction model is an XGBoost prediction model, the short-period prediction model is an LSTM model, and the machine learning prediction model is an RNN model.

[0134] With regard to the apparatus in the above-described embodiments, the specific manner in which the respective modules perform operations has been described in detail in the embodiments related to the method, and thus will not be described in detail here.

[0135] Figure 7 A structural schematic diagram of an electronic device is provided for the exemplary embodiments of the present disclosure. As shown in the figure, the electronic device includes a memory 71 and a processor 72. In addition, the electronic device also includes a power supply component 73 and a communication component 74. Figure 7

[0136] The memory 71 is used to store computer programs and can be configured to store other various data to support operations on the electronic device. Examples of these data include instructions for any application program or method operating on the electronic device.

[0137] The memory 71 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0138] The communication component 74 is used for data transmission with other devices.

[0139] The processor 72 can execute computer instructions stored in the memory 71 for: obtaining a metropolitan area network application scenario type; performing metropolitan area network link capacity prediction according to the metropolitan area network application scenario type by using a link capacity prediction model corresponding to the metropolitan area network application scenario type, to obtain a metropolitan area network link capacity corresponding to the metropolitan area network application scenario type; and performing a work order distribution operation according to the metropolitan area network link capacity.

[0140] Optionally, if the metropolitan area network application scenario type is a city-level overall trend application scenario, the processor 72, when performing metropolitan area network link capacity prediction according to the metropolitan area network application scenario type by using a link capacity prediction model corresponding to the metropolitan area network application scenario type, to obtain a metropolitan area network link capacity corresponding to the metropolitan area network application scenario type, is configured to:

[0141] Performing metropolitan area network link capacity prediction according to the city-level overall trend application scenario by using a long-period prediction model corresponding to the city-level overall trend application scenario, to obtain a metropolitan area network link capacity corresponding to the city-level overall trend application scenario.

[0142] ​Optionally, if the metropolitan area network application scenario type is a single-cell single-index application scenario, the processor 72, when performing metropolitan area network link capacity prediction according to the metropolitan area network application scenario type by using a link capacity prediction model corresponding to the metropolitan area network application scenario type, obtains metropolitan area network link capacity corresponding to the metropolitan area network application scenario type, is configured to:

[0143] According to the single-cell single-index application scenario, the metropolitan area network link capacity prediction is performed by using a deep learning time series prediction model corresponding to the single-cell single-index application scenario, and metropolitan area network link capacity corresponding to the single-cell single-index application scenario is obtained.

[0144] Optionally, if the metropolitan area network application scenario type is a single-link application scenario, the processor 72, when performing metropolitan area network link capacity prediction according to the metropolitan area network application scenario type by using a link capacity prediction model corresponding to the metropolitan area network application scenario type, obtains metropolitan area network link capacity corresponding to the metropolitan area network application scenario type, is configured to:

[0145] According to the single-link application scenario, the metropolitan area network link capacity prediction is performed by using a time series neural network fusion model corresponding to the single-link application scenario, and metropolitan area network link capacity corresponding to the single-link application scenario is obtained.

[0146] Optionally, the time series neural network fusion model includes a combination prediction model, a short-period prediction model, and a machine learning prediction model.

[0147] Optionally, the combination prediction model is an XGBoost prediction model, the short-period prediction model is an LSTM model, and the machine learning prediction model is an RNN model.

[0148] Correspondingly, the embodiment of the present disclosure also provides a computer readable storage medium storing a computer program. When the computer readable storage medium stores the computer program, and the computer program is executed by one or more processors, the one or more processors are caused to perform Figure 1 the steps in the method embodiment.

[0149] Correspondingly, the embodiment of the present disclosure also provides a computer program product, which includes a computer program / instruction, and the computer program / instruction is executed by a processor to perform Figure 1 the steps in the method embodiment.

[0150] The above Figure 7The communication component is configured to facilitate wired or wireless communication between the device containing the communication component and other devices. The device containing the communication component can access wireless networks based on communication standards, such as WiFi, 2G, 3G, 4G / LTE, 5G, or combinations thereof. In one exemplary embodiment, the communication component receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, the communication component also includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID), Infrared Data Association (IrDA) technology, Ultra-Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0151] The above Figure 7 The power supply component provides power to the various components of the device in which it resides. The power supply component may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which it resides.

[0152] The aforementioned electronic devices also include a display screen and audio components.

[0153] The display includes a screen, which may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touchscreen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors can sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation.

[0154] An audio component may be configured to output and / or input audio signals. For example, the audio component includes a microphone (MIC) configured to receive external audio signals when the device containing the audio component is in an operating mode, such as call mode, recording mode, or voice recognition mode. The received audio signals may be further stored in memory or transmitted via a communication component. In some embodiments, the audio component also includes a speaker for outputting audio signals.

[0155] In the embodiments of the apparatus, device, storage medium, and computer program products disclosed herein, the metropolitan area network (MAN) application scenario type is obtained; based on the MAN application scenario type, the MAN link capacity is predicted using the link capacity prediction model corresponding to the MAN application scenario type, thereby obtaining the MAN link capacity corresponding to the MAN application scenario type; based on the MAN link capacity, a work order dispatch operation is performed; this disclosure adopts a corresponding link capacity prediction model for different MAN application scenarios to predict the MAN link capacity, thereby improving the accuracy of MAN management link capacity prediction.

[0156] Those skilled in the art will appreciate that embodiments of the disclosure can be supplied as a method, a system, or a computer program product. Accordingly, the disclosure can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the disclosure can take the form of a computer program product on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, and the like) embodying computer-readable program code.

[0157] The disclosure is described in reference to the flow diagrams and / or block diagrams of the methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flow diagrams and / or block diagrams, and combinations of blocks in the flow diagrams and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processing device, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 means for performing the function specified by the flow diagram and / or block diagram block or blocks.

[0158] These computer program instructions can also be stored in a computer- readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instructions which implement the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 means for performing the function specified by the flow diagram and / or block diagram block or blocks.

[0159] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the flow diagrams and / or block diagrams block or blocks. Figure 1 one or more functions specified in the flow diagram and / or block diagram block or blocks. Figure 1 means for performing the function specified by the flow diagram and / or block diagram block or blocks.

[0160] In one typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0161] The memory can include non-persistent memory, Random Access Memory (RAM), and / or non-volatile memory, such as Read Only Memory (ROM) or flash memory, among others. The memory is an example of computer-readable media.

[0162] Computer-readable media includes permanent and non-permanent, movable and non-movable media that can be implemented by any method or technology to store information. The information can be computer-readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information accessible to a computing device. According to the definition herein, computer-readable media does not include transitory media such as modulated data signals and carriers.

[0163] It should be noted that, in the present document, relational terms such as "first" and "second", and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without limitation, an element preceded by "comprises... a" does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus.

[0164] The above merely provides specific embodiments of the present disclosure, enabling those skilled in the art to understand or implement the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure will not be limited to these embodiments herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting the capacity of metropolitan area network links, characterized in that, include: Obtain the application scenario type for metropolitan area networks; Based on the metropolitan area network (MAN) application scenario type, the MAN link capacity is predicted using the link capacity prediction model corresponding to the MAN application scenario type, thereby obtaining the MAN link capacity corresponding to the MAN application scenario type. The MAN application scenario types include: city-level overall trend application scenario, single cell single indicator application scenario, and single link application scenario. Based on the metropolitan area network link capacity, work orders are dispatched. The step of predicting the metropolitan area network (MAN) link capacity based on the MAN application scenario type using the link capacity prediction model corresponding to the MAN application scenario type, and obtaining the MAN link capacity corresponding to the MAN application scenario type, includes: Based on the overall trend application scenario at the prefecture-level city, the metropolitan area network link capacity is predicted by using the long-term prediction model corresponding to the overall trend application scenario at the prefecture-level city. Based on the single-cell single-index application scenario, the metropolitan area network link capacity is predicted by using the deep learning time series prediction model corresponding to the single-cell single-index application scenario, and the metropolitan area network link capacity corresponding to the single-cell single-index application scenario is obtained. Based on the single-link application scenario, the metropolitan area network link capacity is predicted using the temporal neural network fusion model corresponding to the single-link application scenario, thereby obtaining the metropolitan area network link capacity corresponding to the single-link application scenario.

2. The method according to claim 1, characterized in that, The temporal neural network fusion model includes: a combined prediction model, a short-cycle prediction model, and a machine learning prediction model.

3. The method according to claim 2, characterized in that, The combined prediction model is the XGBoost prediction model, the short-cycle prediction model is the LSTM model, and the machine learning prediction model is the RNN model.

4. A metropolitan area network link capacity prediction device, characterized in that, include: The module retrieves the application scenario type for the metropolitan area network. The prediction module predicts the metropolitan area network (MAN) link capacity based on the MAN application scenario type using the link capacity prediction model corresponding to the MAN application scenario type, thereby obtaining the MAN link capacity corresponding to the MAN application scenario type. The MAN application scenario types include: city-level overall trend application scenario, single cell single indicator application scenario, and single link application scenario. The dispatch module performs work order dispatch operations based on the capacity of the metropolitan area network link; The prediction module uses a long-term prediction model corresponding to the overall trend application scenario at the prefecture-level city level to predict the metropolitan area network link capacity based on the overall trend application scenario at the prefecture-level city level, and obtains the metropolitan area network link capacity corresponding to the overall trend application scenario at the prefecture-level city level. Based on the single-cell single-index application scenario, the metropolitan area network link capacity is predicted by using the deep learning time series prediction model corresponding to the single-cell single-index application scenario, and the metropolitan area network link capacity corresponding to the single-cell single-index application scenario is obtained. Based on the single-link application scenario, the metropolitan area network link capacity is predicted using the temporal neural network fusion model corresponding to the single-link application scenario, thereby obtaining the metropolitan area network link capacity corresponding to the single-link application scenario.

5. An electronic device, characterized in that, include: processor; Memory used to store the processor's executable instructions; The processor is configured to execute the instructions to implement the steps of the method as described in any one of claims 1-3.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-3.

7. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1-3.

Images