Network switching method, apparatus and device

Abstract

The application provides a network switching method, device and equipment, first to-be-identified data and second to-be-identified data of a network device are collected; feature data corresponding to the first to-be-identified data and the second to-be-identified data is determined according to the first to-be-identified data and the second to-be-identified data; the feature data corresponding to the first to-be-identified data and the second to-be-identified data is input into a preset prediction model to obtain network quality information of the network device; and the network of the network device is switched according to the network quality information. By collecting network service quality data of the network device, the network quality advantages and disadvantages of the current main line and the standby line are predicted based on a deep learning model, the network is switched in advance according to the prediction result, and the network communication quality is improved.

Description

Technical Field

[0001] This application relates to the field of communication network technology, and in particular to a network switching method, apparatus and device. Background Technology

[0002] With the continuous growth of data communication services, the demand for global networking is also increasing. To meet these networking needs, Internet access services are provided through Software Defined Wide Area Network (SD-WAN) devices.

[0003] In existing technologies, a primary line and a backup line are set up in the SD-WAN terminal. If it is determined that the network of the SD-WAN primary line is disconnected, the network line of the SD-WAN terminal is switched, and the SD-WAN backup line provides Internet access service.

[0004] However, in the above method, the system only switches to the backup line when the main network is completely disconnected, which causes network outages during the switching process, thereby reducing the quality of network access services and affecting the user experience. Summary of the Invention

[0005] This application provides a network switching method, apparatus, and device to solve the problem of network outages during switching, which in turn reduces the quality of network access services.

[0006] In a first aspect, this application provides a network switching method, the method comprising:

[0007] Collect first and second data to be identified from the network device; wherein, the first data to be identified is the network service quality data of the main network of the network device within a preset time period; and the second data to be identified is the network service quality data of the backup network of the network device within a preset time period.

[0008] Based on the first data to be identified and the second data to be identified, determine the corresponding feature data between the first data to be identified and the second data to be identified;

[0009] The feature data corresponding to the first data to be identified and the second data to be identified are input into a preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device;

[0010] Based on the network quality information, switch the network of the network device.

[0011] In one example, based on the first data to be identified and the second data to be identified, determining the corresponding feature data between the first data to be identified and the second data to be identified includes:

[0012] The first data to be identified is vectorized to obtain the processed first data to be identified; and the second data to be identified is vectorized to obtain the processed second data to be identified.

[0013] The processed first data to be identified and the processed second data to be identified are concatenated to obtain the feature data corresponding to the first data to be identified and the second data to be identified.

[0014] In one example, before collecting the first and second data to be identified from the network device, the method further includes: determining data collection information; wherein, the data collection information is the collection frequency of the first and second data to be identified.

[0015] Collect the first and second data to be identified from the network device, including:

[0016] The first data to be identified and the second data to be identified are collected according to the sampling frequency indicated by the data acquisition information.

[0017] In one example, switching the network of the network device based on the network quality information includes:

[0018] If it is determined that the network quality information indicates that the network quality of the mainline network of the network device is better than the network quality of the backup network of the network device, then the network of the network device is switched to the mainline network;

[0019] If it is determined that the network quality information indicates that the network quality of the primary network of the network device is inferior to the network quality of the backup network of the network device, then the network of the network device is switched to the backup network.

[0020] In one example, the backup network for the network device is a 4G / 5G network.

[0021] In one example, the network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0022] Secondly, this application provides a method for determining a prediction model applied to network handover, the method comprising:

[0023] Acquire the first training data and the second training data of the network device; wherein, the first training data is the network service quality data of the main network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within a preset time period.

[0024] Based on the first training data and the second training data, determine the corresponding feature data between the first training data and the second training data;

[0025] The initial model is trained based on the feature data corresponding to the first training data and the second training data to obtain the preset prediction model; wherein, the preset prediction model is used to process the feature data in the method described in the first aspect to obtain the network quality information of the network device and then perform network switching.

[0026] In one example, based on the first training data and the second training data, determining the corresponding feature data between the first training data and the second training data includes:

[0027] The first training data is vectorized to obtain the processed first training data; and the second training data is vectorized to obtain the processed second training data.

[0028] The processed first training data and the processed second training data are concatenated to obtain the corresponding feature data between the first training data and the second training data.

[0029] In one example, the first training data and the second training data have actual network quality information; the actual network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0030] Based on the feature data corresponding to the first training data and the second training data, the initial model is trained to obtain the preset prediction model, including:

[0031] The feature data corresponding to the first training data and the second training data are input into the initial model to obtain the predicted network quality information of the network device; wherein, the predicted network quality information characterizes whether the predicted network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0032] Based on the predicted network quality information and the actual network quality information, the parameters of the initial model are trained to obtain the preset prediction model.

[0033] In one example, the network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0034] Thirdly, this application provides a network switching device, comprising:

[0035] The data acquisition unit is used to acquire first and second data to be identified from the network device; wherein, the first data to be identified is the network service quality data of the main network of the network device within a preset time period; and the second data to be identified is the network service quality data of the backup network of the network device within a preset time period.

[0036] The determining unit is configured to determine the feature data corresponding to the first data to be identified and the second data to be identified based on the first data to be identified and the second data to be identified.

[0037] The prediction unit is used to input the feature data corresponding to the first data to be identified and the second data to be identified into a preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device;

[0038] The switching unit is used to switch the network of the network device according to the network quality information.

[0039] In one example, the determining unit includes:

[0040] The first processing module is used to perform vectorization processing on the first data to be identified to obtain processed first data to be identified; and to perform vectorization processing on the second data to be identified to obtain processed second data to be identified.

[0041] The second processing module is used to concatenate the processed first data to be identified and the processed second data to be identified to obtain the feature data corresponding to the first data to be identified and the second data to be identified.

[0042] In one example, before the acquisition unit acquires the first and second data to be identified from the network device, the system further includes a determining unit for determining data acquisition information; wherein the data acquisition information is the acquisition frequency of the first and second data to be identified.

[0043] The acquisition unit includes:

[0044] The acquisition module is used to acquire the first data to be identified and the second data to be identified according to the sampling frequency indicated by the data acquisition information.

[0045] In one example, the switching unit includes:

[0046] The first determining module is configured to switch the network of the network device to the main network if the network quality information indicates that the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0047] The second determining module is used to switch the network of the network device to the backup network if it is determined that the network quality information indicates that the network quality of the primary network of the network device is inferior to the network quality of the backup network of the network device.

[0048] In one example, the backup network for the network device is a 4G / 5G network.

[0049] In one example, the network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0050] Fourthly, this application provides a predictive model determination apparatus for network handover, comprising:

[0051] The acquisition unit is used to acquire first training data and second training data of the network device; wherein, the first training data is the network service quality data of the main network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within a preset time period.

[0052] The determining unit is configured to determine the feature data corresponding to the first training data and the second training data based on the first training data and the second training data.

[0053] The training unit is used to train the initial model based on the feature data corresponding to the first training data and the second training data to obtain the preset prediction model; wherein, the preset prediction model is used to process the feature data in the device as described in the third aspect, and then perform network switching after obtaining the network quality information of the network device.

[0054] In one example, the determining unit includes:

[0055] The first processing module is used to perform vectorization processing on the first training data to obtain the processed first training data; and to perform vectorization processing on the second training data to obtain the processed second training data.

[0056] The second processing module is used to concatenate the processed first training data and the processed second training data to obtain the feature data corresponding to the first training data and the second training data.

[0057] In one example, the first training data and the second training data have actual network quality information; the actual network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0058] The training unit includes:

[0059] The prediction module is used to input the feature data corresponding to the first training data and the second training data into the initial model to obtain the predicted network quality information of the network device; wherein, the predicted network quality information characterizes whether the predicted network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0060] The training module is used to train the parameters of the initial model based on the predicted network quality information and the actual network quality information to obtain the preset prediction model.

[0061] In one example, the network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0062] Fifthly, this application provides an electronic device, including: a processor, and a memory communicatively connected to the processor;

[0063] The memory stores computer-executed instructions;

[0064] The processor executes computer execution instructions stored in the memory to implement the methods described in the first and second aspects.

[0065] In a sixth aspect, this application provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the methods described in the first and second aspects.

[0066] In a seventh aspect, this application provides a computer program product comprising: a computer program stored in a readable storage medium, wherein at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the methods described in the first and second aspects.

[0067] This application provides a network switching method, apparatus, and device, which collects first and second data to be identified from a network device. The first data to be identified is the network service quality data of the primary network of the network device within a preset time period; the second data to be identified is the network service quality data of the backup network of the network device within the preset time period. Based on the first and second data to be identified, corresponding feature data is determined. The corresponding feature data is input into a preset prediction model to obtain network quality information of the network device. This network quality information indicates whether the network quality of the primary network of the network device is superior to that of the backup network. Based on the network quality information, the network of the network device is switched. By collecting network quality information of a user's network device over a period of time and using a deep learning model trained with artificial intelligence technology, the network quality of the primary and backup networks is predicted. If the prediction result indicates that the backup network quality is superior, network switching is performed, thereby achieving the purpose of switching to the backup network in advance and improving network communication quality. Attached Figure Description

[0068] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0069] Figure 1 A flowchart illustrating a network switching method provided in an embodiment of this application;

[0070] Figure 2 A flowchart illustrating another network switching method provided in an embodiment of this application;

[0071] Figure 3 A schematic diagram of a network architecture for deploying SD-WAN primary / backup terminal devices is provided in an embodiment of this application;

[0072] Figure 4 A flowchart illustrating a method for determining a prediction model applied to network handover, provided in an embodiment of this application;

[0073] Figure 5A flowchart illustrating another method for determining a prediction model applied to network handover, provided in an embodiment of this application;

[0074] Figure 6 This is a schematic diagram of the structure of a network switching device provided in an embodiment of this application;

[0075] Figure 7 This is a schematic diagram of another network switching device provided in an embodiment of this application;

[0076] Figure 8 A schematic diagram of a predictive model determination device for network handover provided in an embodiment of this application;

[0077] Figure 9 A schematic diagram of another predictive model determination device for network handover provided in an embodiment of this application;

[0078] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application;

[0079] Figure 11 This is a block diagram illustrating an electronic device according to an exemplary embodiment.

[0080] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0081] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0082] Currently, with the accelerated pace of enterprise globalization, the demand for global networking is increasing. Software Defined Wide Area Network (SD-WAN), as a new access technology, is becoming one of the choices for enterprise networking needs due to its lower cost and higher security. SD-WAN technology achieves customer networking needs by building an IPSEC-VPN (Internet Protocol Security) tunnel over the Internet.

[0083] In existing technologies, control scripts are written into the SD-WAN controller, and the system will switch to the backup line only after a short circuit is detected on the main line.

[0084] In one example, an SD-WAN device terminal is used to provide Internet access service. The SD-WAN terminal is configured with a primary line and a backup line. If it is determined that the SD-WAN primary line is disconnected, the network line of the SD-WAN terminal is switched, and the SD-WAN backup line provides Internet access service.

[0085] However, in the above method, the system only switches to the backup line when the main network is completely disconnected, which can cause network outages during the switching process and reduce the quality of network access services.

[0086] This application provides a network switching method, apparatus, and device, which aims to solve the above-mentioned technical problems of the prior art.

[0087] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0088] Figure 1 This is a flowchart illustrating a network handover method provided in an embodiment of this application, as shown below. Figure 1 As shown, the method includes:

[0089] S101. Collect the first and second data to be identified of the network device; wherein, the first data to be identified is the network service quality data of the main network of the network device within a preset time period; and the second data to be identified is the network service quality data of the backup network of the network device within a preset time period.

[0090] For example, the executing entity of this embodiment can be an electronic device, a server, a terminal device, or other apparatus or device capable of executing this embodiment. This embodiment uses an electronic device as an example for description.

[0091] Based on electronic devices, such as SD-WAN terminals, network devices can provide primary network services and backup network services. For such network devices, a controller can be set up to receive control commands from the controller to collect network service quality data to be identified provided by the network devices within a preset recent time period. This includes network service quality data provided by the primary and backup lines of the network devices, i.e., the first data to be identified, and network service quality data provided by the backup lines of the network devices, i.e., the second data to be identified.

[0092] S102. Based on the first data to be identified and the second data to be identified, determine the feature data corresponding to the first data to be identified and the second data to be identified.

[0093] For example, based on the network service quality data provided by the main line of the network device and the network service quality data provided by the backup line of the network device, feature processing is performed on both of these data to obtain the feature network service quality data corresponding to these two data, which is used for further identification and processing.

[0094] S103. Input the feature data corresponding to the first data to be identified and the second data to be identified into the preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0095] For example, a pre-set network quality prediction model is invoked, and the processed network service quality data to be identified is output to the prediction model. The prediction model analyzes and compares the network service quality data to be identified and outputs whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0096] In one example, a pre-set network quality prediction model, such as a time-series-based neural network model, is invoked. The processed network service quality data to be identified is then output to the neural network model. The neural network model analyzes and compares the network service quality data to be identified and obtains the prediction result. For example, if the neural network model outputs "1", it means that the network quality of the main network of the network device is better than that of the backup network of the network device; or, if the neural network model outputs "0", it means that the network quality of the main network of the network device is worse than that of the backup network of the network device.

[0097] S104. Switch the network of the network device according to the network quality information.

[0098] For example, based on the network quality information of the network device output by the network quality prediction model, the network device controller receives the information and controls the network device to switch the network supply line of the network device.

[0099] In one example, the network is currently powered by the primary line of a network device. Based on the network quality information output by the network quality prediction model, if the neural network model outputs "1", it means that the network quality of the primary network of the network device is better than that of the backup network. Upon receiving this "1" signal, the network device controller will issue a control command to prevent the network device from switching to the primary network. Conversely, if the neural network model outputs "0", it means that the network quality of the primary network is worse than that of the backup network. Upon receiving this "0" signal, the network device controller will issue a control command to prevent the network device from switching its primary network to its backup network in advance, thereby ensuring the quality of network service.

[0100] In this embodiment, first and second data to be identified from network devices are collected. The first data to be identified is the network service quality data of the primary network of the network device within a preset time period; the second data to be identified is the network service quality data of the backup network of the network device within the preset time period. Based on the first and second data to be identified, corresponding feature data is determined. The corresponding feature data is input into a preset prediction model to obtain network quality information of the network device. This network quality information indicates whether the network quality of the primary network of the network device is better than that of the backup network. Based on the network quality information, the network of the network device is switched. By collecting network quality information of the user's SD-WAN device over a period of time, and using a deep learning model trained with artificial intelligence technology, the network quality of the primary and backup networks is predicted. If the prediction result indicates that the backup network quality is better, the SD-WAN device is switched to the backup network in advance, thereby improving network communication quality.

[0101] Figure 2 A flowchart illustrating another network handover method provided in this application embodiment is shown below. Figure 2 As shown, the method includes:

[0102] S201. Determine the data acquisition information; wherein, the data acquisition information is the acquisition frequency of the first data to be identified and the second data to be identified.

[0103] In one example, network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0104] In one example, the backup network for the network device is a 4G / 5G network.

[0105] For example, the executing entity of this embodiment can be an electronic device, a server, a terminal device, or other apparatus or device capable of executing this embodiment. This embodiment uses an electronic device as an example for description.

[0106] Based on electronic devices and network equipment, such as SD-WAN terminals, primary network services and backup network services can be provided. The backup line accesses via 4G (4th Generation Mobile Communication Technology) / 5G (5th Generation Mobile Communication Technology). For this network equipment, a controller can be configured to receive control commands from the controller and set the time and frequency for collecting network service quality data for the primary and backup lines. The network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0107] In one example, Figure 3 This application provides a schematic diagram of a network architecture for deploying SD-WAN primary / backup terminal devices, as shown in the embodiments of this application. Figure 3 As shown, on the customer side, SD-WAN (Software-Defined Wide Area Network) primary and backup line terminals are deployed separately. The SD-WAN primary line connects to the POP (Point of Presence) via Internet access, while the backup line connects to the POP via 4G / 5G access. Both the deployed SD-WAN primary and backup lines are managed by the SD-WAN controller. Users can log in to their SD-WAN terminal device accounts and pre-set the time and frequency for collecting network service quality data from the primary and backup lines. The sampling frequency is once per hour. When the collection time arrives, the SD-WAN controller can be notified to collect network latency data, network jitter data, network packet loss rate, and other network service quality data.

[0108] S202. Collect the first and second data to be identified according to the sampling frequency indicated by the data acquisition information.

[0109] For example, according to the preset sampling time and acquisition frequency, when the acquisition time arrives, the controller issues an instruction to acquire the network service quality data to be identified provided by the network device in a preset relatively recent time period, including the main line of the network device, i.e., the first data to be identified, and the network service quality data of the network provided by the backup line of the network device, i.e., the second data to be identified.

[0110] In one example, the user first logs into the SD-WAN terminal, sets the time for collecting network service quality data for the primary and backup lines to 15 days, and the sampling frequency to once per hour. According to the preset sampling time and collection frequency, when the collection time arrives, the controller issues an instruction to collect the network service quality data to be identified provided by the network device within 15 days, including the network service quality data provided by the primary and backup lines of the network device, as well as the network service quality data provided by the backup line of the network device.

[0111] S203. The first data to be identified is vectorized to obtain the processed first data to be identified; and the second data to be identified is vectorized to obtain the processed second data to be identified.

[0112] For example, based on the network service quality data provided by the main line of the network device and the network service quality data provided by the backup line of the network device, the network service quality data provided by the main line of the network device is vectorized, converting the numerical data into vector form, which is the processed main line network service quality data; at the same time, the network service quality data provided by the backup line of the network device is feature-processed and vectorized, converting the numerical data into vector form, which is the processed backup line network service quality data.

[0113] S204. The processed first data to be identified and the processed second data to be identified are concatenated to obtain the feature data corresponding to the first data to be identified and the second data to be identified.

[0114] For example, based on the vectorized primary network service quality data and the vectorized backup network service quality data, the two data are concatenated to obtain the feature data corresponding to the primary network service quality data to be identified and the backup network service quality data to be identified, which are then used for further identification and processing.

[0115] S205. Input the feature data corresponding to the first data to be identified and the second data to be identified into the preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0116] For example, this step can be referred to step S103, and will not be repeated here.

[0117] After step S205, step S206 or step S207 can be executed.

[0118] S206. If it is determined that the network quality information indicates that the network quality of the main network of the network device is better than that of the backup network of the network device, then the network of the network device shall be switched to the main network.

[0119] For example, after step S205, based on the network quality information of the network device output by the network quality prediction model, if it is determined that the network quality of the main network of the network device is better than the network quality of the backup network of the network device, then after receiving the information, the network device controller will issue a control command to the network device. If the current network device's network is the main network, then the network device's network will remain the main network; if the current network device's network is the backup network, then the network device's network will be switched to the main network.

[0120] In one example, the network is currently supplied to the main line of the network device. According to the network quality information of the network device output by the network quality prediction model, if the neural network model outputs "1", it means that the network quality of the main line network of the network device is better than the network quality of the backup line network of the network device. After receiving the "1" signal, the network device controller will issue a control command to the network device to prevent the network device from switching to the current main line network.

[0121] S207. If it is determined that the network quality information indicates that the network quality of the main network of the network device is inferior to that of the backup network of the network device, then the network of the network device shall be switched to the backup network.

[0122] For example, after step S205, based on the network quality information of the network device output by the network quality prediction model, if it is determined that the network quality of the main network of the network device is worse than the network quality of the backup network of the network device, then after receiving the information, the network device controller will issue a control command to the network device. If the current network device's network is the main network, then the network device's network will be switched to the backup network; if the current network device's network is the backup network, then the network device's network will remain the backup network.

[0123] In one example, the network is currently being supplied via the primary line of a network device. Based on the network quality information output by the network quality prediction model, if the neural network model outputs "0", it means that the network quality of the primary network of the network device is inferior to that of the backup network. Upon receiving this "0" signal, the network device controller will issue a control command to the network device to switch the primary network to the backup network in advance, thereby ensuring the quality of network service.

[0124] In this embodiment, based on the above embodiments, data acquisition information is determined; wherein, the data acquisition information is the acquisition frequency of the first and second data to be identified; the first and second data to be identified are acquired according to the sampling frequency indicated by the data acquisition information; the first data to be identified is vectorized to obtain the processed first data to be identified; and the second data to be identified is vectorized to obtain the processed second data to be identified; the processed first and second data to be identified are concatenated to obtain the feature data corresponding to the first and second data to be identified; the feature data corresponding to the first and second data to be identified is input into a preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device; if it is determined that the network quality information characterizes the network quality of the main network of the network device is better than the network quality of the backup network of the network device, then the network of the network device is switched to the main network; if it is determined that the network quality information characterizes the network quality of the main network of the network device is worse than the network quality of the backup network of the network device, then the network of the network device is switched to the backup network. By using a pre-set sampling frequency and connecting a backup line to the 4G / 5G network, the system collects primary and backup network service quality data for the network device to be identified within a recent time period. The collected data is then vectorized to obtain feature data. Based on a pre-set prediction model, this feature data is analyzed and compared to predict the network quality difference between the primary and backup networks of the current network device. Based on the prediction results, network switching is performed in advance to ensure network quality. When switching to the backup network, users can still use the 4G / 5G network, thus improving the user experience.

[0125] Figure 4 A flowchart illustrating a prediction model determination method for network handover provided in this application embodiment is shown below. Figure 4 As shown, the method includes:

[0126] S301. Obtain the first training data and the second training data of the network device; wherein, the first training data is the network service quality data of the main network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within a preset time period.

[0127] For example, the executing entity of this embodiment can be an electronic device, a server, a terminal device, or other apparatus or device capable of executing this embodiment. This embodiment uses an electronic device as an example for description.

[0128] Based on electronic devices, such as SD-WAN terminals, network devices can provide primary network services and backup network services. For such network devices, a controller can be set up to receive control commands from the controller to collect network service quality data to be trained provided by the network devices over a preset long period of time. This includes network service quality data provided by the primary and backup lines of the network devices, i.e., the first training data to be trained, and network service quality data provided by the backup lines of the network devices, i.e., the second training data to be trained.

[0129] S302. Based on the first training data and the second training data, determine the corresponding feature data between the first training data and the second training data.

[0130] For example, based on the network service quality data of the network provided by the main line of the network device to be trained and the network service quality data of the network provided by the backup line of the network device, feature processing is performed on both data to obtain the feature network service quality data corresponding to the two data, which is used for model training.

[0131] S303. Based on the feature data corresponding to the first training data and the second training data, train the initial model to obtain a preset prediction model; wherein, the preset prediction model is used to process the feature data in the network switching method, obtain the network quality information of the network device, and then perform network switching.

[0132] For example, a pre-set initial network quality prediction model is invoked, and the processed feature data to be trained is output to the initial prediction model to train the initial identification model. The parameters of the initial identification model are optimized to obtain a preset prediction model, which is used to process the feature data in the network switching method, obtain the network quality information of the network device, and then perform network switching.

[0133] In this embodiment, first and second training data for network devices are acquired. The first training data is the network service quality data of the mainline network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within the preset time period. Based on the first and second training data, corresponding feature data is determined. An initial model is trained using the corresponding feature data to obtain a preset prediction model. This preset prediction model is used to process the feature data in the network switching method, obtaining the network quality information of the network device before performing network switching. By collecting network quality information of the user's network device over a period of time, the model is trained. Based on the trained prediction model, the network quality of the current mainline and backup lines is predicted, and then network switching of the network device is performed based on the prediction results, thereby improving network communication quality.

[0134] Figure 5 A flowchart illustrating another prediction model determination method for network handover provided in this application embodiment is shown below. Figure 5 As shown, the method includes:

[0135] S401. Obtain the first training data and the second training data of the network device; wherein, the first training data is the network service quality data of the main network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within a preset time period.

[0136] For example, this step can be referred to step S301, and will not be repeated here.

[0137] S402. The first training data is vectorized to obtain the processed first training data; and the second training data is vectorized to obtain the processed second training data.

[0138] For example, based on the network service quality data provided by the main line of the network device and the network service quality data provided by the backup line of the network device, the main line network service quality data to be trained of the network device is vectorized, converting the numerical data into vector form, which is the processed main line network service quality data to be trained; at the same time, the backup line network service quality data to be trained of the network device is feature-processed and vectorized, converting the numerical data into vector form, which is the processed backup line network service quality data to be trained.

[0139] S403. The processed first training data and the processed second training data are concatenated to obtain the feature data corresponding to the first training data and the second training data.

[0140] For example, based on the vectorized mainline network service quality data and the vectorized backup network service quality data, the two data are concatenated to obtain the feature data corresponding to the mainline network service quality data to be trained and the other backup network service quality data to be trained, which are used for further model training.

[0141] S404. Input the feature data corresponding to the first training data and the second training data into the initial model to obtain the predicted network quality information of the network device; wherein, the predicted network quality information represents whether the predicted network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0142] In one example, the first training data and the second training data contain actual network quality information; the actual network quality information indicates whether the network quality of the main network of the actual network device is better than the network quality of the backup network of the network device.

[0143] For example, based on the collected primary network service quality data and backup network service quality data, the actual network quality information of the network device can be analyzed, that is, whether the network quality of the primary network of the actual network device is better than the network quality of the backup network of the network device.

[0144] Furthermore, the pre-set initial network quality prediction model is invoked, and the processed feature data to be trained is output to the initial prediction model to obtain the predicted network quality information of the network device, that is, whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0145] S405. Based on the predicted network quality information and the actual network quality information, train the parameters of the initial model to obtain the preset prediction model.

[0146] For example, in order to obtain an accurate prediction model, an initial identification model is trained, and the parameters of the initial identification model are optimized based on the actual network quality information to obtain a preset prediction model, which is used to process the feature data in the network switching method, and then perform network switching after obtaining the network quality information of the network device.

[0147] In this embodiment, based on the above embodiments, the first training data is vectorized to obtain processed first training data; and the second training data is vectorized to obtain processed second training data; the processed first training data and the processed second training data are concatenated to obtain the corresponding feature data between the two; the corresponding feature data between the first training data and the second training data are input into the initial model to obtain the predicted network quality information of the network device; wherein, the predicted network quality information represents whether the predicted network quality of the main network of the network device is better than the network quality of the backup network of the network device; based on the predicted network quality information and the actual network quality information, the parameters of the initial model are trained to obtain a preset prediction model. By performing vectorization and vector concatenation on the collected primary and backup network service quality data to be trained, the feature quality of the training data is further improved, thereby improving the accuracy of the network quality prediction model. Based on the training data, the initial prediction model is trained multiple times to obtain a preset prediction model, which is used to predict the network quality of the current primary and backup lines. Then, based on the prediction results, network switching of network devices is performed to improve network communication quality.

[0148] Figure 6 This is a schematic diagram of the structure of a network switching device provided in an embodiment of this application, as shown below. Figure 6 As shown, the device 500 includes:

[0149] The acquisition unit 501 is used to acquire first data to be identified and second data to be identified from the network device; wherein, the first data to be identified is the network service quality data of the main network of the network device within a preset time period; and the second data to be identified is the network service quality data of the backup network of the network device within a preset time period.

[0150] The determining unit 502 is used to determine the feature data corresponding to the first data to be identified and the second data to be identified based on the first data to be identified and the second data to be identified.

[0151] The prediction unit 503 is used to input the feature data corresponding to the first data to be identified and the second data to be identified into a preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0152] The switching unit 504 is used to switch the network of the network device based on network quality information.

[0153] The apparatus in this embodiment can execute the technical solutions in the above method. Its specific implementation process and technical principles are the same, and will not be repeated here.

[0154] Figure 7 This is a schematic diagram of another network switching device provided in an embodiment of this application, as shown below. Figure 7 As shown, the device 600 includes:

[0155] The acquisition unit 601 is used to acquire first data to be identified and second data to be identified from the network device; wherein, the first data to be identified is the network service quality data of the main network of the network device within a preset time period; and the second data to be identified is the network service quality data of the backup network of the network device within a preset time period.

[0156] The determining unit 602 is used to determine the feature data corresponding to the first data to be identified and the second data to be identified based on the first data to be identified and the second data to be identified.

[0157] The prediction unit 603 is used to input the feature data corresponding to the first data to be identified and the second data to be identified into a preset prediction model to obtain the network quality information of the network device; wherein, the network quality information characterizes whether the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0158] The switching unit 604 is used to switch the network of the network device based on network quality information.

[0159] In one example, unit 602 is defined as including:

[0160] The first processing module 6021 is used to perform vectorization processing on the first data to be identified to obtain the processed first data to be identified; and to perform vectorization processing on the second data to be identified to obtain the processed second data to be identified.

[0161] The second processing module 6022 is used to concatenate the processed first data to be identified and the processed second data to be identified to obtain the feature data corresponding to the first data to be identified and the second data to be identified.

[0162] In one example, before the acquisition unit 601 acquires the first and second data to be identified from the network device, it further includes a determination unit 605, which is used to determine data acquisition information; wherein, the data acquisition information is the acquisition frequency of the first and second data to be identified.

[0163] The acquisition unit 601 includes:

[0164] The acquisition module 6011 is used to acquire the first data to be identified and the second data to be identified according to the sampling frequency indicated by the data acquisition information.

[0165] In one example, switching unit 604 includes:

[0166] The first determining module 6041 is used to switch the network of the network device to the main network if the network quality information indicates that the network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0167] The second determining module 6042 is used to switch the network of the network device to the backup network if the network quality information indicates that the network quality of the main network of the network device is inferior to that of the backup network of the network device.

[0168] In one example, the backup network for the network device is a 4G / 5G network.

[0169] In one example, network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0170] The apparatus in this embodiment can execute the technical solutions in the above method. Its specific implementation process and technical principles are the same, and will not be repeated here.

[0171] Figure 8 This is a schematic diagram of a predictive model determination device for network handover provided in an embodiment of this application, as shown below. Figure 8 As shown, the device 700 includes:

[0172] The acquisition unit 701 is used to acquire the first training data and the second training data of the network device; wherein, the first training data is the network service quality data of the main network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within a preset time period.

[0173] The determining unit 702 is used to determine the feature data corresponding to the first training data and the second training data based on the first training data and the second training data.

[0174] The training unit 703 is used to train the initial model based on the feature data corresponding to the first training data and the second training data to obtain a preset prediction model; wherein, the preset prediction model is used to process the feature data in the network switching device to obtain the network quality information of the network device and then perform network switching.

[0175] The apparatus in this embodiment can execute the technical solutions in the above method. Its specific implementation process and technical principles are the same, and will not be repeated here.

[0176] Figure 9 A schematic diagram of another predictive model determination device for network handover provided in this application embodiment is shown below. Figure 9 As shown, the device 800 includes:

[0177] The acquisition unit 801 is used to acquire the first training data and the second training data of the network device; wherein, the first training data is the network service quality data of the main network of the network device within a preset time period, and the second training data is the network service quality data of the backup network of the network device within a preset time period.

[0178] The determining unit 802 is used to determine the feature data corresponding to the first training data and the second training data based on the first training data and the second training data.

[0179] The training unit 803 is used to train the initial model based on the feature data corresponding to the first training data and the second training data to obtain a preset prediction model; wherein, the preset prediction model is used to process the feature data in the network switching device to obtain the network quality information of the network device and then perform network switching.

[0180] In one example, unit 802 is defined as including:

[0181] The first processing module 8021 is used to perform vectorization processing on the first training data to obtain the processed first training data; and to perform vectorization processing on the second training data to obtain the processed second training data.

[0182] The second processing module 8022 is used to concatenate the processed first training data and the processed second training data to obtain the feature data corresponding to the first training data and the second training data.

[0183] In one example, the first training data and the second training data contain actual network quality information; the actual network quality information indicates whether the network quality of the main network of the actual network device is better than the network quality of the backup network of the network device.

[0184] Training unit 803 includes:

[0185] The prediction module 8031 ​​is used to input the feature data corresponding to the first training data and the second training data into the initial model to obtain the predicted network quality information of the network device; wherein, the predicted network quality information represents whether the predicted network quality of the main network of the network device is better than the network quality of the backup network of the network device.

[0186] The training module 8032 is used to train the parameters of the initial model based on the predicted network quality information and the actual network quality information to obtain the preset prediction model.

[0187] In one example, network service quality data includes one or more of the following: network latency data, network jitter data, and network packet loss rate.

[0188] The apparatus in this embodiment can execute the technical solutions in the above method. Its specific implementation process and technical principles are the same, and will not be repeated here.

[0189] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application, such as... Figure 10 As shown, the electronic device 900 includes: a memory 91 and a processor 92; the memory 91 is a memory used to store instructions executable by the processor 92.

[0190] The processor 92 is configured to perform the methods provided in the embodiments described above.

[0191] The terminal device also includes a receiver 93 and a transmitter 94. The receiver 93 is used to receive instructions and data sent by other devices, and the transmitter 94 is used to send instructions and data to external devices.

[0192] Figure 11 This is a block diagram illustrating an electronic device according to an exemplary embodiment. The device may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness device, personal digital assistant, etc.

[0193] Electronic device 1000 may include one or more of the following components: processing component 1002, memory 1004, power supply component 1006, multimedia component 1008, audio component 1010, input / output (I / O) interface 1012, sensor component 1014, and communication component 1016.

[0194] Processing component 1002 typically controls the overall operation of electronic device 1000, such as operations associated with display, telephone calls, data communication, camera operation, and recording operations. Processing component 1002 may include one or more processors 1020 to execute instructions to perform all or part of the steps of the methods described above. Furthermore, processing component 1002 may include one or more modules to facilitate interaction between processing component 1002 and other components. For example, processing component 1002 may include a multimedia module to facilitate interaction between multimedia component 10010 and processing component 1002.

[0195] Memory 1004 is configured to store various types of data to support the operation of electronic device 1000. Examples of this data include instructions for any application or method operating on electronic device 1000, contact data, phonebook data, messages, pictures, videos, etc. Memory 1004 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0196] Power supply component 1006 provides power to various components of electronic device 1000. Power supply component 1006 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to electronic device 1000.

[0197] Multimedia component 1008 includes a screen that provides an output interface between electronic device 1000 and a user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of touch or swipe actions but also the duration and pressure associated with the touch or swipe operation. In some embodiments, multimedia component 10010 includes a front-facing camera and / or a rear-facing camera. When electronic device 1000 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0198] Audio component 1010 is configured to output and / or input audio signals. For example, audio component 1010 includes a microphone (MIC) configured to receive external audio signals when electronic device 1000 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in memory 1004 or transmitted via communication component 1016. In some embodiments, audio component 1010 also includes a speaker for outputting audio signals.

[0199] I / O interface 1012 provides an interface between processing component 1002 and peripheral interface modules, such as keyboards, click wheels, buttons, etc. These buttons may include, but are not limited to, home buttons, volume buttons, power buttons, and lock buttons.

[0200] Sensor assembly 1014 includes one or more sensors for providing state assessments of various aspects of electronic device 1000. For example, sensor assembly 1014 may detect the on / off state of electronic device 1000, the relative positioning of components such as the display and keypad of electronic device 1000, changes in position of electronic device 1000 or a component of electronic device 1000, the presence or absence of user contact with electronic device 1000, the orientation or acceleration / deceleration of electronic device 1000, and temperature changes of electronic device 1000. Sensor assembly 1014 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 1014 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 1014 may also include an accelerometer, gyroscope, magnetometer, pressure sensor, or temperature sensor.

[0201] Communication component 1016 is configured to facilitate wired or wireless communication between electronic device 1000 and other devices. Electronic device 1000 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 1016 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 1016 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0202] In an exemplary embodiment, the electronic device 1000 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the methods described above.

[0203] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 1004 including instructions, which can be executed by a processor 1020 of an electronic device 1000 to perform the above-described method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0204] This application also provides a non-transitory computer-readable storage medium, which, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to perform the above-described method.

[0205] This application also provides a computer program product, which includes: a computer program stored in a readable storage medium, at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the solution provided in any of the above embodiments.

[0206] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the following claims.

[0207] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A network handover method, characterized by, The method comprises: Collecting first and second to-be-identified data of a network device; wherein the first to-be-identified data is network service quality data of a main network of the network device within a preset time period; the second to-be-identified data is network service quality data of a backup network of the network device within a preset time period; the network device is an SD-WAN terminal; the main network is the Internet; and the backup network is a 4G / 5G network; Vectorizing the first to-be-identified data to obtain processed first to-be-identified data, and vectorizing the second to-be-identified data to obtain processed second to-be-identified data; Splicing the processed first to-be-identified data and the processed second to-be-identified data to obtain feature data corresponding to the first to-be-identified data and the second to-be-identified data, and inputting the feature data into a preset prediction model to obtain network quality information of the network device; wherein the preset prediction model is a time-series neural network model, and the network quality information indicates whether the network quality of the main network of the network device is better than that of the backup network of the network device; Switching the network of the network device according to the network quality information.

2. The method of claim 1, wherein, Before collecting the first and second to-be-identified data of the network device, the method further comprises determining data collection information; wherein the data collection information is a collection frequency for collecting the first and second to-be-identified data. Collecting the first and second to-be-identified data of the network device comprises: Collecting the first and second to-be-identified data according to the sampling frequency indicated by the data collection information.

3. The method of claim 1, wherein, Switching the network of the network device according to the network quality information comprises: If it is determined that the network quality information indicates that the network quality of the main network of the network device is better than that of the backup network of the network device, switching the network of the network device to the main network; If it is determined that the network quality information indicates that the network quality of the main network of the network device is worse than that of the backup network of the network device, switching the network of the network device to the backup network.

4. The method according to any one of claims 1 to 3, characterized in that, The network service quality data comprises one or more of the following: network delay data, network jitter data, and network packet loss rate. 5.A method for determining a prediction model applied to network switching, characterized in that, The method comprises: Obtaining first and second training data of a network device; wherein the first training data is network service quality data of a main network of the network device within a preset time period, and the second training data is network service quality data of a backup network of the network device within a preset time period; Determining feature data corresponding to the first and second training data according to the first and second training data; According to the feature data corresponding between the first training data and the second training data, the initial model is trained to obtain the preset prediction model; wherein the preset prediction model is used to process the feature data in the method as claimed in any one of claims 1-4 to obtain the network quality information of the network device for network switching.

6. The method of claim 5, wherein, According to the first training data and the second training data, the feature data corresponding between the first training data and the second training data is determined, including: The first training data is vectorized to obtain the processed first training data; and the second training data is vectorized to obtain the processed second training data; The processed first training data and the processed second training data are spliced to obtain the feature data corresponding between the first training data and the second training data.

7. The method of claim 5, wherein, The first training data and the second training data have actual network quality information; the actual network quality information represents whether the network quality of the main network of the network device is better than the network quality of the standby network of the network device; According to the feature data corresponding between the first training data and the second training data, the initial model is trained to obtain the preset prediction model, including: The feature data corresponding between the first training data and the second training data is input into the initial model to obtain the predicted network quality information of the network of the network device; wherein the predicted network quality information represents whether the predicted network quality of the main network of the network device is better than the network quality of the standby network of the network device; According to the predicted network quality information and the actual network quality information, the parameters of the initial model are trained to obtain the preset prediction model.

8. The method according to any one of claims 5-7, characterized in that, The network service quality data includes one or more of the following: network delay data, network jitter data, network packet loss rate.

9. A network switching apparatus, characterized by comprising: The device includes: The acquisition unit is configured to acquire first identification data and second identification data of a network device; wherein the first identification data is network service quality data of a main network of the network device within a preset time period; the second identification data is network service quality data of a standby network of the network device within a preset time period; the network device is an SD-WAN terminal; the main network is the Internet; and the standby network is a 4G / 5G network; The determination unit is configured to determine feature data corresponding between the first identification data and the second identification data according to the first identification data and the second identification data. The prediction unit is configured to input the feature data corresponding to the first to-be-identified data and the second to-be-identified data into a preset prediction model to obtain network quality information of the network device, wherein the preset prediction model is a time-series neural network model, and the network quality information indicates whether the network quality of the main line network of the network device is better than the network quality of the standby line network of the network device. The switching unit is configured to switch the network of the network device according to the network quality information. The determination unit comprises: The first processing module is configured to perform vectorization processing on the first to-be-identified data to obtain processed first to-be-identified data, and perform vectorization processing on the second to-be-identified data to obtain processed second to-be-identified data. The second processing module is configured to perform splicing processing on the processed first to-be-identified data and the processed second to-be-identified data to obtain the feature data corresponding to the first to-be-identified data and the second to-be-identified data.

10. The apparatus of claim 9, wherein, Before the collection unit is configured to collect the first to-be-identified data and the second to-be-identified data of the network device, the determination unit is further configured to determine data collection information, wherein the data collection information is a collection frequency of the first to-be-identified data and the second to-be-identified data. The collection unit comprises: The collection module is configured to collect the first to-be-identified data and the second to-be-identified data according to the sampling frequency indicated by the data collection information.

11. The apparatus of claim 9, wherein, The switching unit comprises: The first determination module is configured to switch the network of the network device to the main line network if it is determined that the network quality information indicates that the network quality of the main line network of the network device is better than the network quality of the standby line network of the network device. The second determination module is configured to switch the network of the network device to the standby line network if it is determined that the network quality information indicates that the network quality of the main line network of the network device is worse than the network quality of the standby line network of the network device.

12. The apparatus of any one of claims 9-11, wherein, The network service quality data comprises one or more of the following: network delay data, network jitter data, and network packet loss rate.

13. A prediction model determination apparatus applied to network switching, characterized in that, The apparatus comprises: The acquisition unit is configured to acquire first to-be-trained data and second to-be-trained data of a network device, wherein the first to-be-trained data is network service quality data of a main line network of the network device within a preset time period, and the second to-be-trained data is network service quality data of a standby line network of the network device within the preset time period. The determination unit is configured to determine feature data corresponding to the first to-be-trained data and the second to-be-trained data according to the first to-be-trained data and the second to-be-trained data. The training unit is configured to train an initial model according to the feature data corresponding to the first to-be-trained data and the second to-be-trained data to obtain the preset prediction model, wherein the preset prediction model is used to process feature data in the apparatus of any one of claims 9-12 to obtain network quality information of a network device and then perform network switching.

14. The apparatus of claim 13, wherein, The determining unit comprises: The first processing module is configured to perform vectorization processing on the first training data to obtain processed first training data, and perform vectorization processing on the second training data to obtain processed second training data; The second processing module is configured to perform splicing processing on the processed first training data and the processed second training data to obtain feature data corresponding to the first training data and the second training data.

15. The apparatus of claim 13, wherein, The first training data and the second training data have actual network quality information; the actual network quality information represents whether the network quality of the main network of the network device is better than the network quality of the standby network of the network device; The training unit comprises: The prediction module is configured to input the feature data corresponding to the first training data and the second training data into the initial model to obtain predicted network quality information of the network of the network device; wherein the predicted network quality information represents whether the network quality of the main network of the network device is better than the network quality of the standby network of the network device; The training module is configured to train parameters of the initial model according to the predicted network quality information and the actual network quality information to obtain the preset prediction model.

16. The apparatus of any one of claims 13-15, wherein, The network service quality data includes one or more of the following: network delay data, network jitter data, network packet loss rate.

17. An electronic device, comprising: It comprises: A processor and a memory connected in communication with the processor; The memory stores computer execution instructions; The processor executes the computer execution instructions stored in the memory to implement the method of any one of claims 1-8.

18. A computer-readable storage medium, characterized in that, The computer readable storage medium stores computer execution instructions, and the computer execution instructions are executed by the processor to implement the method of any one of claims 1-8.

Images