A traffic pool construction method and system

Abstract

The present invention discloses a traffic pool construction method and system. Based on historical traffic sequences, the system predicts multiple backup traffic packages corresponding to users at multiple future time nodes, and predicts the future time nodes and activation probabilities of users activating backup traffic packages. The backup traffic includes traffic values and traffic package prices, which improves the accuracy of traffic usage predictions and the accuracy of traffic package predictions. This allows for the preparation of appropriate traffic packages for users to avoid insufficient network traffic during peak network usage. On this basis, the backup traffic packages are partitioned and stored based on future time nodes, activation probabilities, traffic package prices, and traffic values to obtain a traffic pool. The backup traffic packages can be partitioned and stored to quickly retrieve backup traffic package data suitable for the user, thereby improving the accuracy of traffic usage predictions, enhancing traffic utilization rates, and enhancing user experience.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a traffic pool construction method and system. Background Art

[0002] Network traffic refers to the amount of data transmitted over a network within a certain period of time. The size of network traffic reflects network usage and the level of data transmission activity. In an era of explosive network development, during peak periods of user network usage, traffic demand often exceeds supply, significantly impacting users' work and lives. Therefore, this application proposes the construction of a traffic pool to address the problem of network inability to meet demand when user traffic usage explodes. Summary of the Invention

[0003] The purpose of the present invention is to provide a flow pool construction method and system to solve the above-mentioned problems existing in the prior art.

[0004] In a first aspect, an embodiment of the present invention provides a method for constructing a traffic pool, comprising:

[0005] Obtaining a user's historical traffic sequence, where the historical traffic sequence includes multiple historical traffic usages, each of which corresponds to a plurality of historical time nodes;

[0006] Based on historical traffic sequences, the system predicts multiple backup traffic packages corresponding to the user at multiple future time points, and predicts the future time points and activation probability of the user's backup traffic packages; the backup traffic includes traffic value and traffic package price;

[0007] The backup traffic packages are partitioned and stored based on future time nodes, activation probability, traffic package price and traffic value to obtain a traffic pool.

[0008] Furthermore, based on the historical traffic sequence, multiple backup traffic packages corresponding to the user at multiple future time nodes are predicted, and the future time nodes and activation probabilities of the user activating the backup traffic packages are predicted, including:

[0009] Perform Fourier transform on the historical traffic sequence and map it to the frequency domain to obtain frequency domain traffic data;

[0010] Based on historical traffic sequences and frequency domain traffic data, multiple backup traffic packages corresponding to multiple future time nodes, future time nodes at which users activate the backup traffic packages, and activation probabilities are obtained.

[0011] Furthermore, based on the historical traffic sequence and frequency domain traffic data, a plurality of backup traffic packages corresponding to a plurality of future time nodes, the future time nodes at which the user activates the backup traffic packages, and the activation probability are obtained, including:

[0012] The historical traffic sequence and frequency domain traffic data are respectively used as the input of the joint prediction network to obtain the time domain traffic prediction sequence and frequency domain traffic prediction sequence respectively;

[0013] Perform inverse Fourier transform on the frequency domain traffic prediction sequence to obtain the inverse traffic prediction sequence;

[0014] Obtaining a prediction difference sequence between the inverse flow prediction sequence and the time domain flow prediction sequence, the prediction difference sequence including a plurality of predicted flow differences;

[0015] Obtaining a first covariance sequence of the inverse flow prediction sequence and the time domain flow prediction sequence; the first covariance sequence includes a plurality of predicted flow covariances; the predicted flow covariances correspond to the predicted flow differences in a one-to-one manner;

[0016] If the predicted traffic covariance is greater than the set value, the value in the time domain traffic prediction sequence corresponding to the predicted traffic covariance is used as the traffic value of the backup traffic package;

[0017] Based on the traffic value and traffic price parameters, obtain the traffic package price;

[0018] The predicted traffic difference is marked on the backup traffic package at the future time node corresponding to the predicted difference sequence;

[0019] Obtain the variance of the frequency domain traffic prediction sequence, and use the cross entropy of the variance of the frequency domain traffic prediction sequence and the predicted traffic covariance as the activation probability;

[0020] The joint prediction network includes a first long short-term memory network and a second long short-term memory network. The input of the first long short-term memory network is frequency domain traffic data, and the input of the second long short-term memory network is historical traffic sequence.

[0021] As a further step, the backup traffic packages are partitioned and stored based on the future time node, activation probability, traffic package price and traffic value to obtain a traffic pool, including:

[0022] Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value;

[0023] If the access heat is greater than the first set value, the backup traffic packet is stored in the first block;

[0024] If the access heat is greater than the second set value and less than or equal to the first set value, the backup traffic packet is stored in the second block;

[0025] If the access heat is less than or equal to the second set value, the backup traffic packet is stored in the third block.

[0026] As a further step, access popularity is obtained based on future time nodes, activation probability, traffic package price and traffic value, including:

[0027] Obtaining a traffic jump value, which is the difference between the value in the time domain traffic prediction sequence corresponding to the future time node and the historical traffic usage corresponding to the current time node;

[0028] Obtain the traffic change rate, which is equal to the traffic jump value divided by the time distance between the future time node and the current time node;

[0029] Get the price change rate; the price change rate is equal to the rate of change between the traffic package price and the price of the traffic package used by the user at the current time point;

[0030] Determine the visit popularity equal to activation probability * price change rate / traffic change rate.

[0031] In a second aspect, an embodiment of the present invention further provides a traffic pool construction system, including:

[0032] An acquisition module is used to obtain a user's historical traffic sequence, where the historical traffic sequence includes multiple historical traffic usages, each of which corresponds to a plurality of historical time nodes.

[0033] A prediction module is used to predict multiple backup traffic packages corresponding to the user at multiple future time nodes based on historical traffic sequences, and to predict the future time nodes and activation probabilities of the user activating the backup traffic packages; the backup traffic includes traffic value and traffic package price;

[0034] The traffic pool construction module is used to partition and store the spare traffic packages based on future time nodes, activation probability, traffic package price and traffic value to obtain a traffic pool.

[0035] Compared with the prior art, the embodiments of the present invention achieve the following beneficial effects:

[0036] The embodiment of the present invention provides a traffic pool construction method and system, which predicts multiple backup traffic packages corresponding to users at multiple future time nodes based on historical traffic sequences, and predicts the future time nodes and activation probabilities of users activating backup traffic packages, wherein the backup traffic includes traffic values and traffic package prices, thereby improving the accuracy of traffic usage predictions and the accuracy of traffic package predictions, thereby preparing suitable traffic packages for users to avoid insufficient network traffic during peak network usage. On this basis, the backup traffic packages are partitioned and stored based on future time nodes, activation probabilities, traffic package prices, and traffic values to obtain a traffic pool. The backup traffic packages can be partitioned and stored so that the backup traffic package data suitable for the user can be quickly retrieved and provided to the user, thereby improving the accuracy of traffic usage predictions, enhancing traffic utilization, and enhancing user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a flow chart of a method for constructing a traffic pool provided by an embodiment of the present invention.

[0038] Figure 2 It is a structural diagram of an electronic device provided by an embodiment of the present invention.

[0039] Markings in the figure: 500 - bus; 501 - receiver; 502 - processor; 503 - transmitter; 504 - memory; 505 - bus interface. DETAILED DESCRIPTION

[0040] The present invention will be described in detail below with reference to the accompanying drawings.

[0041] Example

[0042] The embodiment of the present invention provides a method for constructing a flow pool. Figure 1 As shown, the traffic pool construction method includes the following steps:

[0043] S101: Obtain the user's historical traffic sequence.

[0044] The historical traffic sequence includes multiple historical traffic usages, each of which corresponds to multiple historical time nodes. The most recent historical time node is the current time node.

[0045] S102: Based on the historical traffic sequence, predict multiple backup traffic packages corresponding to the user at multiple future time nodes, and predict the future time nodes and activation probabilities of the user activating the backup traffic packages.

[0046] Among them, the standby traffic includes the traffic value and traffic package price.

[0047] S103: Partition and store the spare traffic packages based on the future time node, activation probability, traffic package price and traffic value to obtain a traffic pool.

[0048] By adopting the above solution, based on historical traffic sequences, multiple backup traffic packages corresponding to users at multiple future time nodes are predicted, as well as the future time nodes and activation probabilities of users activating backup traffic packages. The backup traffic includes traffic value and traffic package price, which improves the accuracy of traffic usage prediction and traffic package prediction. This allows for the preparation of appropriate traffic packages for users to avoid insufficient network traffic during peak network usage. On this basis, the backup traffic packages are partitioned and stored based on future time nodes, activation probabilities, traffic package prices, and traffic values to obtain a traffic pool. The backup traffic packages can be partitioned and stored to quickly retrieve backup traffic package data suitable for the user, thereby improving the accuracy of traffic usage prediction, enhancing traffic utilization, and enhancing user experience.

[0049] In an embodiment of the present invention, the backup traffic package can be used in addition to the existing package.

[0050] The method of predicting multiple backup traffic packages corresponding to the user at multiple future time points and predicting the future time points and activation probabilities of the user activating the backup traffic packages includes:

[0051] Perform Fourier transform on the historical traffic sequence and map it to the frequency domain space to obtain frequency domain traffic data. Based on the historical traffic sequence and frequency domain traffic data, obtain multiple backup traffic packages that correspond to multiple future time nodes, the future time nodes at which users activate the backup traffic packages, and the activation probability. Specifically, based on the historical traffic sequence and frequency domain traffic data, obtain multiple backup traffic packages that correspond to multiple future time nodes, the future time nodes at which users activate the backup traffic packages, and the activation probability, including:

[0052] The historical traffic series and frequency-domain traffic data are used as inputs to the joint prediction network, resulting in a time-domain traffic prediction series and a frequency-domain traffic prediction series. The time-domain traffic prediction series consists of multiple values, each of which is called a time-domain traffic prediction value. The frequency-domain traffic prediction series is then inverse Fourier transformed to obtain an inverse traffic prediction series. This ensures that the resulting inverse traffic prediction series has the same dimensionality as the time-domain traffic prediction series.

[0053] A prediction difference sequence between the inverse flow prediction sequence and the time domain flow prediction sequence is obtained, the prediction difference sequence including a plurality of predicted flow differences. A first covariance sequence between the inverse flow prediction sequence and the time domain flow prediction sequence is obtained. The first covariance sequence includes a plurality of predicted flow covariances, and the predicted flow covariances correspond one-to-one to the predicted flow differences.

[0054] If the predicted traffic covariance is greater than the set value, the value in the time domain traffic prediction sequence corresponding to the predicted traffic covariance is used as the traffic value of the backup traffic packet. In the embodiment of the present invention, the set value is a real number between 0.2 and 0.6, such as 0.3 or 0.4.

[0055] Based on the traffic value and traffic price parameter, the traffic package price is obtained. The traffic price parameter can be the unit price of traffic, which is determined by the operator. The traffic price parameter can be 5 or 10. The predicted traffic difference is annotated on the backup traffic package at the future time node corresponding to the predicted difference sequence. The variance of the frequency domain traffic prediction sequence is obtained, and the cross entropy of the variance of the frequency domain traffic prediction sequence and the predicted traffic covariance is used as the activation probability.

[0056] This embodiment of the present invention incorporates the concept of a Long Short-Term Memory (LSTM) network. The joint prediction network includes two LSTM networks: a first LSTM network and a second LSTM network. The first LSTM network receives frequency-domain traffic data as input, while the second LSTM network receives historical traffic data as input.

[0057] Optionally, the output of the first long short-term memory network is used as a hidden gate of the second long short-term memory network.

[0058] As a further step, the backup traffic packages are partitioned and stored based on the future time node, activation probability, traffic package price and traffic value to obtain a traffic pool, including:

[0059] Obtain access popularity based on future time nodes, activation probability, traffic package price and traffic value.

[0060] If the access popularity is greater than the first set value, the backup traffic packet is stored in the first block. If the access popularity is greater than the second set value and less than or equal to the first set value, the backup traffic packet is stored in the second block. If the access popularity is less than or equal to the second set value, the backup traffic packet is stored in the third block. The first set value ranges from 5 to 50. The second set value ranges from 0.5 to 30. In an embodiment of the present invention, the first set value is greater than the second set value.

[0061] In other words, the traffic pool is divided into three blocks, namely the first block, the second block and the third block. Different backup traffic packages are stored in different blocks, so that the more popular backup traffic packages can be activated first and conveniently.

[0062] The access popularity is obtained based on the future time node, activation probability, traffic package price and traffic value, including:

[0063] Obtain the traffic jump value, which is the difference between the value in the time domain traffic prediction sequence corresponding to the future time node and the historical traffic usage corresponding to the current time node. Obtain the traffic change rate, which is equal to the traffic jump value divided by the time distance between the future time node and the current time node. Obtain the price change rate, which is equal to the rate of change between the traffic package price and the price of the traffic package used by the user at the current time node. Determine the access popularity equal to the activation probability * price change rate / traffic change rate. To this end, the accuracy of the backup traffic package prediction of the backup traffic package is improved, thereby improving the accuracy of predicting the possibility of activating the backup traffic package.

[0064] Based on the above-mentioned traffic pool construction method, an embodiment of the present invention provides a traffic pool construction system for executing the above-mentioned traffic pool construction method. In an embodiment of the present invention, the traffic pool construction system includes:

[0065] The acquisition module is used to obtain the user's historical traffic sequence, where the historical traffic sequence includes multiple historical traffic usages, and the multiple historical traffic usages correspond one-to-one to multiple historical time nodes.

[0066] The prediction module is used to predict multiple backup traffic packages corresponding to users at multiple future time nodes based on historical traffic sequences, and to predict the future time nodes and activation probability of users activating backup traffic packages; the backup traffic includes traffic value and traffic package price.

[0067] The traffic pool construction module is used to partition and store the spare traffic packages based on future time nodes, activation probability, traffic package price and traffic value to obtain a traffic pool.

[0068] The specific implementation methods of the functions of each module of the above system are as described in the above traffic pool construction method, and will not be repeated here.

[0069] The embodiment of the present invention also provides an electronic device for integrating the above-mentioned flow pool construction system, such as Figure 2 As shown, the electronic device includes a memory 504, a processor 502, and a computer program stored in the memory 504 and executable on the processor 502. When the processor 502 executes the program, the steps of any of the aforementioned methods for constructing a traffic pool are implemented.

[0070] Among them, Figure 2In the embodiment of the present invention, a bus architecture (represented by bus 500) is shown. Bus 500 may include any number of interconnected buses and bridges, and bus 500 links various circuits including one or more processors represented by processor 502 and memory represented by memory 504. Bus 500 may also link various other circuits such as peripherals, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 505 provides an interface between bus 500 and receiver 501 and transmitter 503. Receiver 501 and transmitter 503 may be the same component, namely a transceiver, which provides a unit for communicating with various other devices over a transmission medium. Processor 502 is responsible for managing bus 500 and general processing, while memory 504 may be used to store data used by processor 502 when performing operations.

[0071] An embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the traffic pool construction methods described above.

[0072] The algorithm and display provided herein are not inherently related to any particular computer, virtual system or other device. Various general-purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing this type of system. In addition, the present invention is not directed to any specific programming language. It should be understood that various programming languages can be utilized to realize the content of the present invention described herein, and the above description of specific languages is for the purpose of disclosing the best mode of the present invention.

[0073] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0074] Similarly, it should be understood that in order to streamline the present disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed method should not be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as reflected in the claims below, inventive aspects lie in less than all the features of the individual embodiments disclosed above. Accordingly, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the invention.

[0075] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition may be divided into multiple submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed herein may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0076] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination.

[0077] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components in the apparatus according to an embodiment of the present invention. The present invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for executing a part or all of the methods described herein. Such a program implementing the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

Claims

1. A method for constructing a flow pool, characterized in that: include: Obtaining a historical traffic sequence of the user, where the historical traffic sequence includes multiple historical traffic usages, each of which corresponds to a plurality of historical time nodes; Perform Fourier transform on the historical traffic sequence and map it to the frequency domain to obtain frequency domain traffic data; The historical traffic sequence and frequency domain traffic data are respectively used as the input of the joint prediction network to obtain the time domain traffic prediction sequence and frequency domain traffic prediction sequence respectively; Perform inverse Fourier transform on the frequency domain traffic prediction sequence to obtain the inverse traffic prediction sequence; Obtaining a prediction difference sequence between the inverse flow prediction sequence and the time domain flow prediction sequence, the prediction difference sequence including a plurality of predicted flow differences; Obtaining a first covariance sequence of the inverse flow prediction sequence and the time domain flow prediction sequence; the first covariance sequence includes a plurality of predicted flow covariances; the predicted flow covariances correspond to the predicted flow differences in a one-to-one manner; If the predicted traffic covariance is greater than the set value, the value in the time domain traffic prediction sequence corresponding to the predicted traffic covariance is used as the traffic value of the backup traffic package; the backup traffic package includes the traffic value and the traffic package price; Based on the traffic value and traffic price parameters, obtain the traffic package price; The predicted traffic difference is marked on the backup traffic package at the future time node corresponding to the predicted difference sequence; Obtain the variance of the frequency domain traffic prediction sequence, and use the cross entropy of the variance of the frequency domain traffic prediction sequence and the predicted traffic covariance as the activation probability; The joint prediction network includes a first long short-term memory network and a second long short-term memory network. The input of the first long short-term memory network is frequency domain traffic data, and the input of the second long short-term memory network is historical traffic sequence. The backup traffic packages are partitioned and stored based on future time nodes, activation probability, traffic package price and traffic value to obtain a traffic pool.

2. The flow pool construction method according to claim 1, characterized in that: The backup traffic packages are partitioned and stored based on future time nodes, activation probability, traffic package price, and traffic value to obtain a traffic pool, including: Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value; If the access heat is greater than the first set value, the backup traffic packet is stored in the first block; If the access heat is greater than the second set value and less than or equal to the first set value, the backup traffic packet is stored in the second block; If the access heat is less than or equal to the second set value, the backup traffic packet is stored in the third block.

3. The flow pool construction method according to claim 2, characterized in that, Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value, including: Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value, including: Obtaining a traffic jump value, which is the difference between the value in the time domain traffic prediction sequence corresponding to the future time node and the historical traffic usage corresponding to the current time node; Obtain the traffic change rate, which is equal to the traffic jump value divided by the time distance between the future time node and the current time node; Get the price change rate; the price change rate is equal to the rate of change between the traffic package price and the price of the traffic package used by the user at the current time point; Determine the visit popularity equal to activation probability * price change rate / traffic change rate.

4. A traffic pool construction system, characterized in that: include: An acquisition module is used to obtain a user's historical traffic sequence, where the historical traffic sequence includes multiple historical traffic usages, each of which corresponds to a plurality of historical time nodes. Prediction module for Perform Fourier transform on the historical traffic sequence and map it to the frequency domain to obtain frequency domain traffic data; The historical traffic sequence and frequency domain traffic data are respectively used as the input of the joint prediction network to obtain the time domain traffic prediction sequence and frequency domain traffic prediction sequence respectively; Perform inverse Fourier transform on the frequency domain traffic prediction sequence to obtain the inverse traffic prediction sequence; Obtaining a prediction difference sequence between the inverse flow prediction sequence and the time domain flow prediction sequence, the prediction difference sequence including a plurality of predicted flow differences; Obtaining a first covariance sequence of the inverse flow prediction sequence and the time domain flow prediction sequence; the first covariance sequence includes a plurality of predicted flow covariances; the predicted flow covariances correspond to the predicted flow differences in a one-to-one manner; If the predicted traffic covariance is greater than the set value, the value in the time domain traffic prediction sequence corresponding to the predicted traffic covariance is used as the traffic value of the backup traffic package; the backup traffic package includes the traffic value and the traffic package price; Based on the traffic value and traffic price parameters, obtain the traffic package price; The predicted traffic difference is marked on the backup traffic package at the future time node corresponding to the predicted difference sequence; Obtain the variance of the frequency domain traffic prediction sequence, and use the cross entropy of the variance of the frequency domain traffic prediction sequence and the predicted traffic covariance as the activation probability; The joint prediction network includes a first long short-term memory network and a second long short-term memory network. The input of the first long short-term memory network is frequency domain traffic data, and the input of the second long short-term memory network is historical traffic sequence. The traffic pool construction module is used to partition and store the spare traffic packages based on future time nodes, activation probability, traffic package price and traffic value to obtain a traffic pool.

5. The flow pool construction system according to claim 4, characterized in that: The backup traffic packages are partitioned and stored based on future time nodes, activation probability, traffic package price, and traffic value to obtain a traffic pool, including: Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value; If the access heat is greater than the first set value, the backup traffic packet is stored in the first block; If the access heat is greater than the second set value and less than or equal to the first set value, the backup traffic packet is stored in the second block; If the access heat is less than or equal to the second set value, the backup traffic packet is stored in the third block.

6. The flow pool construction system according to claim 5, characterized in that: Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value, including: Obtain access popularity based on future time nodes, activation probability, traffic package price, and traffic value, including: Obtaining a traffic jump value, which is the difference between the value in the time domain traffic prediction sequence corresponding to the future time node and the historical traffic usage corresponding to the current time node; Obtain the traffic change rate, which is equal to the traffic jump value divided by the time distance between the future time node and the current time node; Get the price change rate; the price change rate is equal to the rate of change between the traffic package price and the price of the traffic package used by the user at the current time point; Determine the visit popularity equal to activation probability * price change rate / traffic change rate.

Images