Access network traffic prediction method and device, electronic equipment and storage medium

Abstract

The invention provides an access network traffic prediction method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring internet surfing data of a target user; classifying the internet surfing data to obtain useful flow data and useless flow data; performing time sequence analysis and behavior pattern analysis on the useful traffic to obtain time sequence characteristics and behavior characteristics; and fusing the time sequence characteristics and the behavior characteristics to obtain fusion characteristics, inputting the fusion characteristics into a traffic prediction model obtained by training to obtain predicted access network traffic of the target user in a preset time, and realizing deep mining and understanding of the user traffic characteristics through Internet access data analysis and machine learning technologies of the target user. Therefore, the traffic prediction model can accurately predict the access network traffic of the target user in the future preset time period, and further help the system to determine the network traffic change trend and the possible traffic peak in the future time period, thereby providing a scientific basis for traffic resource allocation.

Description

Technical Field

[0001] The present application relates to the technical field of access network traffic prediction, and in particular to an access network traffic prediction method, device, electronic device and storage medium. Background Art

[0002] As a bridge between users and the wide area network, the efficiency of traffic management and resource allocation of the access network directly affects the overall performance of the network and user experience. With the rapid development of Internet technology, especially the popularity of bandwidth-intensive services such as video streaming, online games and cloud computing, the data traffic faced by the access network has increased dramatically. This surge in traffic has brought huge challenges, especially during peak traffic periods. In current related technologies, historical averages or simple linear prediction models are usually used for traffic prediction. These models cannot accurately capture the time-varying and bursty nature of traffic, resulting in insufficient resources being provided when traffic surges, and may cause resource waste when traffic is scarce. Summary of the invention

[0003] In view of this, the purpose of the present application is to propose an access network traffic prediction method, device, electronic device and storage medium to solve or partially solve the above-mentioned technical problems.

[0004] Based on the above purpose, the present application provides an access network traffic prediction method, including:

[0005] Obtain the target user’s Internet access data;

[0006] Classifying the Internet access data to obtain useful traffic data and useless traffic data;

[0007] Performing time series analysis and behavior pattern analysis on the useful traffic to obtain time series characteristics and behavior characteristics;

[0008] The time series feature and the behavior feature are fused to obtain a fused feature, and the fused feature is input into a traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time.

[0009] Based on the same concept, the present application also proposes an access network traffic prediction device, including:

[0010] Acquisition module, to obtain the target user's Internet access data;

[0011] A classification module, which classifies the Internet access data to obtain useful flow data and useless flow data;

[0012] A feature analysis module performs time series analysis and behavior pattern analysis on the useful traffic to obtain time series features and behavior features;

[0013] The prediction module fuses the time series feature and the behavior feature to obtain a fused feature, and inputs the fused feature into the traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time.

[0014] Based on the same concept, the present application also proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the access network traffic prediction method as described above when executing the program.

[0015] Based on the same concept, the present application also proposes a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the access network traffic prediction method as described above.

[0016] Based on the same concept, the present application also proposes a computer program product, which includes a computer program. The computer program is executed by one or more processors so that the processors execute the access network traffic prediction method as described above.

[0017] From the above, it can be seen that the access network traffic prediction method, device, electronic device and storage medium provided by the present application obtain the Internet access data of the target user; classify the Internet access data to obtain useful traffic data and useless traffic data; perform time series analysis and behavior pattern analysis on the useful traffic to obtain time series characteristics and behavior characteristics; fuse the time series characteristics and the behavior characteristics to obtain fusion characteristics, and input the fusion characteristics into the trained traffic prediction model to obtain the predicted access network traffic of the target user within the preset time. Through the analysis of the Internet access data of the target user and machine learning technology, the user traffic characteristics are deeply mined and understood, so that the traffic prediction model can accurately predict the access network traffic of the target user in the future preset time period, and further help the system determine the network traffic change trend and possible traffic peak in the future period, thereby providing a scientific basis for traffic resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 A schematic diagram of an application scenario of an embodiment of the present application;

[0020] Figure 2 A schematic diagram of a flow chart of a method for predicting access network traffic according to an embodiment of the present application;

[0021] Figure 3 A schematic diagram of the structure of an access network traffic prediction device according to an embodiment of the present application;

[0022] Figure 4 A schematic diagram of a specific electronic device hardware structure of an embodiment of the present application. DETAILED DESCRIPTION

[0023] It is understandable that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and relevant provisions.

[0024] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with general skills in the field to which the present application belongs. The "first", "second" and similar words used in the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connecting" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, which can be direct connections, indirect connections, wired connections, and wireless connections. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly. In addition, in the description of the present application, unless otherwise specified, the term "multiple" refers to two or more. The term "and / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects are in an "or" relationship.

[0025] The principles and spirit of the present application will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided only to enable those skilled in the art to better understand and implement the present application, and are not intended to limit the scope of the present application in any way. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0026] It is understandable that before using the technical solutions of each embodiment of the present application, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0027] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to the electronic device, application, server, storage medium or other software or hardware that performs the operation of the technical solution of the present application according to the prompt message.

[0028] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0029] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation method of this application. Other methods that meet relevant laws and regulations may also be applied to the implementation method of this application.

[0030] It should be understood herein that any number of elements in the drawings is for illustration rather than limitation, and any naming is only for distinction rather than having any limiting meaning.

[0031] As described in the background technology section, in current related technologies, historical averages or simple linear prediction models are usually used for traffic prediction. These models cannot accurately capture the time-varying and bursty nature of traffic, resulting in the inability to provide sufficient resources when traffic surges, and may cause waste of resources when traffic is scarce.

[0032] In summary, in order to solve the above problems, the embodiment of the present application proposes an access network traffic prediction method, including:

[0033] Obtain the Internet access data of the target user; classify the Internet access data to obtain useful traffic data and useless traffic data; perform time series analysis and behavior pattern analysis on the useful traffic to obtain time series features and behavior features; fuse the time series features and the behavior features to obtain fused features, and input the fused features into the trained traffic prediction model to obtain the predicted access network traffic of the target user within a preset time. Through the Internet access data analysis and machine learning technology of the target user, in-depth mining and understanding of the user traffic characteristics are achieved, so that the traffic prediction model can accurately predict the access network traffic of the target user within a preset time period in the future, and further help the system determine the network traffic change trend and possible traffic peak in a future period of time, thereby providing a scientific basis for traffic resource allocation.

[0034] In some specific application scenarios, the access network traffic prediction method of the present application can be applied to various platforms or systems involving access network traffic prediction. As an example, the application scenario includes at least one server and at least one terminal. Figure 1 , which is a schematic diagram of an application scenario of an access network traffic prediction method provided by an exemplary embodiment of the present disclosure. The application scenario includes a terminal device 101, a server 102, and a data storage system 103. Among them, the terminal device 101, the server 102 and the data storage system 103 can be connected through a wired or wireless communication network. The terminal device includes but is not limited to a desktop computer, a mobile phone, a mobile computer, a tablet computer, a media player, a smart wearable device, a personal digital assistant (PDA) or other electronic devices capable of realizing the above functions. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server and the terminal can communicate through the network to realize data transmission. Among them, the network can be a wired network or a wireless network, and this application does not specifically limit this.

[0035] The server can be a server that provides various services. Specifically, the server can be used to provide background services for applications running on the terminal. Optionally, in some implementations, the access network traffic prediction method provided in the embodiment of the present application can be executed by the server. Optionally, in some implementations, the access network traffic prediction method provided in the embodiment of the present application can be directly executed by the terminal device. It should be noted that when executed directly by the terminal device, a high-performance computing board can be installed on the optical terminal device (OLT) side, and the computing board is equipped with a high-speed processor and storage system capable of processing large-scale data and performing complex calculations. Such a configuration can not only collect and analyze data from the user side in real time, but also quickly perform data processing and analysis tasks without increasing the burden on the network subject. In addition, considering the unique structure and technical characteristics of the passive optical network, this method of centralized data processing on the OLT side is more conducive to the centralized management and optimal allocation of network resources, and can effectively reduce the overall complexity and operation and maintenance costs of the network.

[0036] Optionally, the server can be hardware or software. When the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or it can be implemented as a single server. When the server is software, it can be implemented as multiple software or software modules (for example, software or software modules for providing distributed services), or it can be implemented as a single software or software module. This embodiment of the application does not specifically limit this.

[0037] Optionally, the wireless network or wired network uses standard communication technology and / or protocol. The network is usually the Internet, but it can also be any network, including but not limited to a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a dedicated network or any combination of a virtual private network). In some embodiments, the data exchanged through the network is represented by technologies and / or formats including hypertext mark-up language (HTML), extensible markup language (XML), etc. In addition, conventional encryption technologies such as secure socket layer (SSL), transport layer security (TLS), virtual private network (VPN), and Internet protocol security (IPsec) can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above data communication technologies.

[0038] The data storage system stores a large amount of data. Optionally, the data storage system can partition the data according to different uses and data types. The source of the data includes but is not limited to an existing database, data crawled from the Internet, or data uploaded by the user when using the client.

[0039] The following describes the access network traffic prediction method according to the exemplary implementation of the present application in conjunction with a specific application scenario. It should be noted that the above application scenario is only shown to facilitate understanding of the spirit and principle of the present application, and the implementation of the present application is not limited in this regard. On the contrary, the implementation of the present application can be applied to any applicable scenario.

[0040] refer to Figure 2 , is a flow chart of a method for predicting access network traffic according to an embodiment of the present application, the method comprising the following steps:

[0041] S101, obtaining the Internet access data of the target user.

[0042] In the specific implementation, in order to ensure the accuracy and reliability of the traffic prediction model, it is necessary to first collect high-quality user Internet access data, which includes but is not limited to the user's Internet access time, service type, data packet size, and sending and receiving timestamps. Optionally, the target user is the user who currently needs to perform access network traffic prediction. In some embodiments, the target user can be any user. Optionally, in some embodiments, these Internet access data are usually collected in real time by a high-performance computing board installed on the OLT side. This configuration of high-speed processors and storage systems enables the device to quickly perform data collection and preliminary processing tasks without increasing the network burden.

[0043] In some embodiments, in order to ensure the quality and consistency of the input data, after the Internet access data of the target user is obtained, the Internet access data can be preprocessed. Optionally, the preprocessing step includes removing noise and outliers in the data, filling missing values, and performing data standardization. De-noising usually involves filtering out data fluctuations that are not caused by user behavior, such as false data points caused by network equipment failures. Outlier processing focuses on identifying and correcting data points that do not conform to normal user behavior patterns, such as unexpected traffic peaks, which may be caused by DDoS attacks or system errors. Data standardization is another key step that aims to adjust the data to a uniform scale and eliminate the effects of different units or magnitudes. Common standardization methods include minimum-maximum standardization and Z-score standardization (standard deviation standardization).

[0044] The formula for min-max normalization is:

[0045]

[0046] Where x is the original data point, min(x) and max(x) are the minimum and maximum values ​​in the data set, respectively. ′ is the normalized data point. This method normalizes the scale of all features to the interval [0,1]. The Z-score normalization is calculated by the following formula:

[0047]

[0048] Among them, μ and σ are the mean and standard deviation of the data set, respectively, and z represents the standardized data point. This method is centered on 0, taking into account the mean and fluctuation range of the data, making the data distribution close to the standard normal distribution.

[0049] S102, classifying the Internet access data to obtain useful traffic data and useless traffic data.

[0050] In specific implementation, in order to more accurately understand the traffic demand and make more effective resource allocation decisions. After obtaining the above-mentioned Internet access data, the Internet access data is classified to obtain useful traffic data and useless traffic data. In the classification process, useful traffic can be defined as data streams directly related to user needs, such as video streaming, file downloads, online games, etc.; while automatic software updates, advertising content, etc. can be defined as useless traffic. Optionally, algorithms such as support vector machines (SVM) and decision trees can be used to classify traffic according to its characteristics such as data packet size, transmission frequency, service type, etc.

[0051] In some embodiments, the Internet access data is classified to obtain useful traffic data and useless traffic data, specifically including:

[0052] Extracting classification features of the Internet access data; the classification features include: time interval between sending and receiving data packets, service type code and session duration;

[0053] The classification features are input into the classification model obtained through training to obtain useful flow data and useless flow data; wherein the classification model is a support vector machine.

[0054] In specific implementation, the support vector machine can be used to seek to maximize the interval between different categories.

[0055] In some embodiments, the online data may be classified by the following formula:

[0056]

[0057] sty i (w·x i +b)≥1-ξ i ,ξ i ≥0;

[0058] Among them, w represents the normal vector of the classification hyperplane, C is the regularization parameter, ξ i is the slack variable, y i is the class label of the data point, x i is the eigenvector.

[0059] S103, performing time series analysis and behavior pattern analysis on the useful traffic to obtain time series characteristics and behavior characteristics.

[0060] In the specific implementation, in order to extract the most effective features for the model, in this embodiment, time series analysis is used to capture the time-dependent characteristics of traffic data, that is, time series features, and at the same time, behavioral pattern analysis is used to obtain the target user's Internet access frequency, service type preference and other features, that is, behavioral features.

[0061] In some embodiments, time series analysis is performed on the useful flow data to obtain time series features, specifically including:

[0062] Determine the average flow rate and flow fluctuation degree of the useful flow data within a preset time window;

[0063] The average flow rate and the flow rate fluctuation degree are determined as time series features.

[0064] In specific implementation, the sliding window technology can be used to calculate statistical indicators in each time window, such as average flow and flow fluctuation. Sliding window statistics provide a local perspective on flow changes, and the choice of window size depends on the needs of the prediction model. The formula for calculating the average flow in the time window is:

[0065]

[0066] The formula for calculating the degree of flow fluctuation within the time window is:

[0067]

[0068] Among them, x i is the flow rate at the i-th time point, W is the window size, and t is the current time point. In this way, the average flow rate and fluctuation level in each time period can be captured, providing richer input features for the flow prediction model.

[0069] In some embodiments, behavior pattern analysis is performed by constructing a user behavior matrix that includes characteristics such as user surfing frequency, service type preferences, etc. This matrix enables the application of various machine learning algorithms to identify specific user behavior patterns and correlate these patterns with traffic fluctuations. The user behavior matrix is ​​in the form of:

[0070]

[0071] Where m represents the number of users, n represents the number of behavioral features, and element a ij Represents the value of the i-th user on the h-th behavioral feature.

[0072] In some embodiments, the behavior pattern analysis is performed on the useful traffic to obtain the behavior characteristics, which specifically includes:

[0073] Determining a plurality of initial representations corresponding to the useful flow;

[0074] Determining the contribution corresponding to each of the initial representations;

[0075] The initial representation whose contribution is greater than a preset threshold is confirmed as the behavior feature.

[0076] In the specific implementation, taking into account the different contributions or importance of different behavioral characteristics to the predicted access network traffic, therefore, when extracting behavioral characteristics, first determine the multiple initial representations corresponding to the useful traffic, and then determine the contribution corresponding to each of the initial representations. Only the initial representation whose contribution is greater than the preset threshold will be confirmed as the behavioral feature, that is, those initial representations whose contribution is less than or equal to the preset threshold will be deleted.

[0077] In some embodiments, the contribution corresponding to the initial representation is calculated by the following formula:

[0078] IG(Y|X)=H(Y)-H(Y|X);

[0079] Among them, H(Y) is the entropy of the target variable, that is, the predicted access network traffic of the target user within the preset time, and H(Y|X) is the conditional entropy of the target variable after the given feature, that is, the initial representation. In addition, model-based feature selection methods can also be used to determine the importance (contribution) of features, such as random forests and gradient boosting machines. These methods directly evaluate the effectiveness of each behavioral feature during the model training process.

[0080] S104, fusing the time series feature and the behavior feature to obtain a fused feature, and inputting the fused feature into a traffic prediction model obtained through training to obtain a predicted access network traffic of the target user within a preset time.

[0081] In specific implementation, after obtaining the time series features and behavior features, the two are first fused to obtain fused features, and then the fused features are input into the trained traffic prediction model to obtain the predicted access network traffic of the target user within a preset time. It should be noted that feature fusion is to combine features extracted from different methods to form a comprehensive feature set to enhance the prediction ability of the model. For example, features obtained from time series analysis and features obtained from user behavior analysis can be combined to provide a more comprehensive data view for the model. Optionally, when fusing features, the time series features and behavior features can be first mapped to the same dimension and then spliced ​​together.

[0082] In some embodiments, the traffic prediction model includes a convolutional neural network layer and a long short-term memory network layer; inputting the fusion feature into the trained traffic prediction model to obtain the predicted access network traffic of the target user within a preset time specifically includes:

[0083] Inputting the fused features into the convolutional neural network layer to obtain spatial features;

[0084] The spatial features are input into the long short-term memory network layer to obtain the predicted access network traffic of the target user within a preset time.

[0085] In specific implementation, in order to accurately predict the access network traffic, in an embodiment of the present application, the traffic prediction model includes a convolutional neural network layer and a long short-term memory network layer, that is, the convolutional neural network (CNN) and the long short-term memory network (LSTM) are combined together to form a new hybrid neural network structure as the structure of the traffic prediction model.

[0086] It should be noted that convolutional neural networks are mainly used to extract local features in time series data. CNN effectively identifies spatial patterns in data, such as traffic peaks or periodic changes, through multiple layers of filters, which is crucial for capturing short-term trends in traffic data. Each convolutional layer is applied through the following formula:

[0087] h out =σ(W*h in + b);

[0088] Among them, h in is the output of the input layer or the previous layer, W is the convolution kernel weight, b is the bias term, σ is the nonlinear activation function, * represents the convolution operation, h out is the convolutional layer output.

[0089] Long short-term memory networks specifically deal with long-term dependencies in data, enabling the model to remember the long-term history of input data, which is extremely important for understanding time-based data flow patterns. The core of the LSTM unit is the unit state that it maintains and updates. The specific update process is carried out through the following formula:

[0090] f t =σ(W f ·[h t-1 ,x t ]+b f );

[0091] i t =σ(W i ·[h t-1 ,x t ]+b i );

[0092]

[0093] o t =σ(W o ·[h t-1 ,x t ]+b o );

[0094] h t =o t ⊙tanh(C t );

[0095] Among them, x t is the input feature at time t, h t is the output state, C t is the cell state, f t ,i t ,o t are the activation states of the forget gate, input gate, and output gate, respectively, and W and b are the corresponding weights and biases.

[0096] In some embodiments, the process of training the traffic prediction model includes:

[0097] The parameters of the traffic prediction model are adjusted using a hyperparameter automatic adjustment technique based on Bayesian optimization.

[0098] In specific implementation, the embodiment of the present application adopts the hyperparameter automatic adjustment technology based on Bayesian optimization to improve the prediction accuracy and processing speed by systematically exploring and optimizing model parameters.

[0099] It should be noted that Bayesian optimization is an efficient hyperparameter optimization method that uses Bayesian statistical theory to guide the search process to find the best model parameter combination. This method is based on the Gaussian process and predicts the performance of unexplored parameters by observing historical evaluation results, thereby guiding the search direction. The advantage of this method is that it can find hyperparameters with better performance in a smaller number of evaluations, which is especially suitable for situations where model training is expensive. The specific Bayesian optimization formula can be expressed as:

[0100] α(x)=μ(x)-κσ(x);

[0101] Where μ(x) and σ(x) are the predicted mean and standard deviation of the model performance under hyperparameter x, respectively, and κ is a hyperparameter that balances exploration and exploitation. The acquisition function is used to determine the next most promising hyperparameter point to be evaluated at each step.

[0102] During the optimization process, the improvement of model performance is not only reflected in the increase in prediction accuracy, but also in the improvement of response speed and processing efficiency to complex traffic patterns. By optimizing the model structure and parameters, such as adjusting the depth of the network layer, the number of neurons, and the learning rate, the model can adapt to new traffic conditions more quickly, thereby reducing response delays and improving the overall operating efficiency of the system. Performance optimization also includes the scale processing of the model to ensure that the model can run stably in network environments of different sizes. Through the dynamic adjustment and optimization of the model, the network can provide higher service quality during peak traffic periods, while saving resources during low traffic periods, thereby optimizing resource allocation and reducing operating costs. This dynamic adjustment mechanism enables network service providers to adjust resource allocation based on real-time data, significantly improving network flexibility and user satisfaction.

[0103] In some embodiments, after inputting the fusion feature into the traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time, the method further includes:

[0104] The network resource allocation strategy is adjusted based on the predicted access network traffic; wherein the network resource allocation strategy includes bandwidth allocation, routing selection and cache strategy.

[0105] In specific implementation, after obtaining the predicted access network traffic of the target user within the preset time, the network resource allocation strategy is dynamically adjusted according to the traffic prediction results to optimize bandwidth allocation, routing selection and caching strategy to ensure efficient operation of the network and improve user experience. According to the predicted traffic changes, the system will prioritize allocating more bandwidth for peak traffic periods and reclaim bandwidth resources during low traffic periods. Specifically, the system will calculate the current traffic share and future traffic share of each node, and prioritize the node combination with the smallest sum of future traffic share as the transmission path. During the path selection process, if the resources of a candidate path are insufficient, the system will reject the traffic allocation request of the path and try other candidate paths until a path with sufficient resources is found. This dynamic adjustment mechanism not only optimizes resource utilization, but also reduces network congestion, improves the overall performance of the network and user satisfaction, thereby achieving the optimal allocation of network resources.

[0106] In some embodiments, the embodiments of the present application use Docker containerization technology and Kubernetes automatic expansion technology to ensure that the model can run stably in various network environments and has sufficient scalability. Through Docker containerization, the model and all its dependencies are encapsulated in a lightweight, portable container to achieve cross-platform consistency and rapid deployment. Through Kubernetes automatic expansion technology, the number of copies of the container is dynamically adjusted according to the actual network traffic, ensuring that resource processing capacity is automatically increased during traffic peaks and resource usage is reduced during traffic troughs to optimize costs.

[0107] In some embodiments, the embodiments of the present application deploy an online learning system, adopt streaming data processing and incremental learning strategies, so that the model can adapt to new data and traffic changes in real time. The online learning system can process continuously flowing data without retraining the entire model, thereby quickly responding to changes in network conditions. Streaming data processing technology will update the data set in real time to maintain the timeliness and processing efficiency of the data. Through continuous iteration and learning, the model can continuously improve the prediction accuracy and response speed, ensure that the network resource allocation strategy can be dynamically adjusted, optimize network performance and improve user satisfaction. Optionally, in some embodiments, the traffic prediction model can be updated by the following formula:

[0108]

[0109] Among them, w old and w new are the model weights before and after the update, η is the learning rate, y is the actual flow value, is the traffic value predicted by the model, and x is the input feature. It should be noted that the use of streaming data processing technology can process large-scale, high-speed data streams. This includes using data window partitioning technology to update and maintain data sets in real time to ensure the real-time and processing efficiency of data. Streaming data processing not only increases the flexibility of the system to process data, but also greatly reduces memory consumption and improves data processing speed.

[0110] The access network traffic prediction method provided by the present application obtains the Internet access data of the target user; classifies the Internet access data to obtain useful traffic data and useless traffic data; performs time series analysis and behavior pattern analysis on the useful traffic to obtain time series features and behavior features; fuses the time series features and the behavior features to obtain fused features, and inputs the fused features into the traffic prediction model obtained by training to obtain the predicted access network traffic of the target user within a preset time. By analyzing the Internet access data of the target user and using machine learning technology, in-depth mining and understanding of the user traffic characteristics are achieved, so that the traffic prediction model can accurately predict the access network traffic of the target user within a preset time period in the future, further helping the system to determine the network traffic change trend and possible traffic peaks in a future period of time, thereby providing a scientific basis for traffic resource allocation.

[0111] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides an access network traffic prediction device.

[0112] refer to Figure 3 , the access network traffic prediction device comprises:

[0113] Acquisition module 201, acquiring Internet access data of a target user;

[0114] A classification module 202 classifies the Internet access data to obtain useful traffic data and useless traffic data;

[0115] The feature analysis module 203 performs time series analysis and behavior pattern analysis on the useful traffic to obtain time series features and behavior features;

[0116] The prediction module 204 fuses the time series feature and the behavior feature to obtain a fused feature, and inputs the fused feature into the trained traffic prediction model to obtain the predicted access network traffic of the target user within a preset time.

[0117] In some embodiments, the classification module is specifically used to:

[0118] Extracting classification features of the Internet access data; the classification features include: time interval between sending and receiving data packets, service type code and session duration;

[0119] The classification features are input into the classification model obtained through training to obtain useful flow data and useless flow data; wherein the classification model is a support vector machine.

[0120] In some embodiments, the feature analysis module is specifically used to:

[0121] Determine the average flow rate and flow fluctuation degree of the useful flow data within a preset time window;

[0122] The average flow rate and the flow rate fluctuation degree are determined as time series features.

[0123] In some embodiments, the feature analysis module is specifically used to:

[0124] Determining a plurality of initial representations corresponding to the useful flow;

[0125] Determining the contribution corresponding to each of the initial representations;

[0126] The initial representation whose contribution is greater than a preset threshold is confirmed as the behavior feature.

[0127] In some embodiments, the traffic prediction model includes a convolutional neural network layer and a long short-term memory network layer; the prediction module is specifically used to:

[0128] Inputting the fused features into the convolutional neural network layer to obtain spatial features;

[0129] The spatial features are input into the long short-term memory network layer to obtain the predicted access network traffic of the target user within a preset time.

[0130] In some embodiments, the device further comprises a model training module for:

[0131] The parameters of the traffic prediction model are adjusted using a hyperparameter automatic adjustment technique based on Bayesian optimization.

[0132] In some embodiments, the apparatus further comprises a distribution module for:

[0133] The network resource allocation strategy is adjusted based on the predicted access network traffic; wherein the network resource allocation strategy includes bandwidth allocation, routing selection and cache strategy.

[0134] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0135] The access network traffic prediction device of the above embodiment is used to implement the corresponding access network traffic prediction method in any of the above embodiments, and has the beneficial effects of the corresponding access network traffic prediction method embodiment, which will not be repeated here.

[0136] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present disclosure also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the access network traffic prediction method described in any of the above embodiments is implemented.

[0137] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0138] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0139] Figure 4 A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040 and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030 and the communication interface 1040 are connected to each other in communication within the device through the bus 1050.

[0140] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0141] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0142] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0143] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and the power equipment. The communication module can realize communication through a wired method (such as USB, network cable, etc.) or a wireless method (such as mobile network, WIFI, Bluetooth, etc.).

[0144] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0145] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0146] The electronic device of the above embodiment is used to implement the corresponding access network traffic prediction method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0147] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present disclosure also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the access network traffic prediction method described in any of the above embodiments.

[0148] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, 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 technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0149] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the access network traffic prediction method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0150] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides a computer program product, which includes a computer program. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor execute the access network traffic prediction method described in the above-mentioned embodiments. Corresponding to the execution subject corresponding to each step in each embodiment of the access network traffic prediction method, the processor that executes the corresponding step can belong to the corresponding execution subject.

[0151] To simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the following fact, that is, the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuit) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0152] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0153] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the present application as described above, which are not provided in detail for the sake of simplicity.

[0154] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A method for predicting access network traffic, characterized in that: include: Obtain the target user’s Internet access data; Classifying the Internet access data to obtain useful traffic data and useless traffic data; Performing time series analysis and behavior pattern analysis on the useful traffic to obtain time series characteristics and behavior characteristics; The time series feature and the behavior feature are fused to obtain a fused feature, and the fused feature is input into a traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time.

2. The method according to claim 1, characterized in that The Internet access data is classified to obtain useful traffic data and useless traffic data, specifically including: Extracting classification features of the Internet access data; the classification features include: time interval between sending and receiving data packets, service type code and session duration; The classification features are input into the classification model obtained through training to obtain useful flow data and useless flow data; wherein the classification model is a support vector machine.

3. The method according to claim 1, characterized in that The useful flow data is subjected to time series analysis to obtain time series characteristics, including: Determine the average flow rate and flow fluctuation degree of the useful flow data within a preset time window; The average flow rate and the flow rate fluctuation degree are determined as time series features.

4. The method according to claim 1, characterized in that: The useful traffic is subjected to a behavior pattern analysis to obtain behavior characteristics, including: Determining a plurality of initial representations corresponding to the useful flow; Determining the contribution corresponding to each of the initial representations; The initial representation whose contribution is greater than a preset threshold is confirmed as the behavior feature.

5. The method according to claim 1, characterized in that: The traffic prediction model includes a convolutional neural network layer and a long short-term memory network layer; the fusion feature is input into the traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time, specifically including: Inputting the fused features into the convolutional neural network layer to obtain spatial features; The spatial features are input into the long short-term memory network layer to obtain the predicted access network traffic of the target user within a preset time.

6. The method according to claim 1, characterized in that The process of training the traffic prediction model includes: The parameters of the traffic prediction model are adjusted using a hyperparameter automatic adjustment technique based on Bayesian optimization.

7. The method according to claim 1, characterized in that After inputting the fusion feature into the traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time, the method further includes: The network resource allocation strategy is adjusted based on the predicted access network traffic; wherein the network resource allocation strategy includes bandwidth allocation, routing selection and cache strategy.

8. An access network traffic prediction device, characterized in that: include: Acquisition module, to obtain the Internet access data of the target user; A classification module, which classifies the Internet access data to obtain useful flow data and useless flow data; A feature analysis module performs time series analysis and behavior pattern analysis on the useful traffic to obtain time series features and behavior features; The prediction module fuses the time series feature and the behavior feature to obtain a fused feature, and inputs the fused feature into the traffic prediction model obtained through training to obtain the predicted access network traffic of the target user within a preset time.

9. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 7 is implemented.

10. A non-transitory computer-readable storage medium, characterized in that: The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 7.

Images