Processing method and device

Abstract

The invention provides a processing method and device, and is applied to the technical field of communication. The processing method is applied to the terminal equipment and comprises the steps that network parameters of data transmission of the terminal equipment at a first moment are acquired, and the network parameters comprise a first network signal parameter and a second network signal parameter; the first network signal parameter is processed based on the first intelligent model, a first prediction result is determined, the first prediction result represents the influence degree of the first network signal parameter at the first moment on data transmission at the second moment, the second network signal parameter is processed based on the second intelligent model, and a second prediction result is determined, the second prediction result represents the influence degree of the second network signal parameter at the first moment on data transmission at the second moment, and the second moment is later than the first moment; and determining a target network for data transmission at the second moment according to the first prediction result and the second prediction result.

Description

Technical Field

[0001] The present disclosure relates to the field of communication technologies, and in particular, to a processing method and apparatus. Background Art

[0002] Currently, during data transmission, the network cannot be adjusted according to the network conditions, resulting in low data transmission efficiency and poor user experience. Summary of the Invention

[0003] In view of this, the present disclosure provides a processing method and apparatus.

[0004] According to a first aspect of the present disclosure, there is provided a processing method applied to a terminal device, including: obtaining network parameters for data transmission by the terminal device at a first moment, where the network parameters include a first network signal parameter and a second network signal parameter; processing the first network signal parameter based on a first intelligent model to determine a first prediction result, where the first prediction result represents the influence degree of the first network signal parameter at the first moment on data transmission at a second moment, and processing the second network signal parameter based on a second intelligent model to determine a second prediction result, where the second prediction result represents the influence degree of the second network signal parameter at the first moment on data transmission at the second moment, and the second moment is later than the first moment; and determining a target network for data transmission at the second moment according to the first prediction result and the second prediction result.

[0005] According to an embodiment of the present disclosure, determining a target network for data transmission at the second moment according to the first prediction result and the second prediction result includes: when the first prediction result represents that the first network signal parameter at the first moment causes data transmission failure at the second moment, and the second prediction result represents that the second network signal parameter at the first moment causes data transmission success at the second moment, determining that the target network for data transmission at the second moment is the second network.

[0006] According to an embodiment of the present disclosure, when data transmission at the first moment uses a first network and the intensity of the first network signal is higher than a first threshold, where the first threshold represents a limit value of the intensity of the first network signal when the network for data transmission is switched from the first network to the second network, the method further includes: when the registration of the IP multimedia subsystem of the terminal device is successful, determining that the priority of the second network for data transmission at the second moment is higher than that of the first network, and switching the network for data transmission at the second moment from the first network to the second network.

[0007] According to an embodiment of the present disclosure, determining that the priority of a second network for data transmission at a second moment is higher than that of a first network includes: performing a first process on the intensity of the first network signal at a first moment to make the intensity of the first network signal lower than a first threshold; and / or performing a second process on the intensity of the second network signal at the first moment to make the intensity of the second network signal higher than a second threshold, where the second threshold represents the minimum value of the second network signal intensity in data transmission.

[0008] According to an embodiment of the present disclosure, determining that the priority of a second network for data transmission at a second moment is higher than that of a first network includes: obtaining attribute information of the data transmission, where the attribute information represents the application scenario of the data transmission; when the attribute information meets a preset condition, updating the first threshold to a first target threshold, where the first target threshold is greater than the first threshold; and / or when the attribute information meets a preset condition, updating the second threshold to a second target threshold, where the second target threshold is less than the second threshold.

[0009] According to an embodiment of the present disclosure, the first intelligent model is trained in the following manner:

[0010] Obtaining a plurality of first sample data, where the plurality of first sample data includes first network call data of a plurality of sample users, and each first network call data includes: the first network signal intensity at a third moment, the registered network of the first network, the radio frequency band of the first network, and the first historical state of the first network call at a fourth moment, where the fourth moment is later than the third moment; inputting the first network signal intensity, the registered network of the first network, and the radio frequency band of the first network at the third moment into the first intelligent model to determine the first predicted state of the first network call at the fourth moment; adjusting the parameters of the first intelligent model according to the difference between the first predicted state and the first historical state of multiple first network call data until the difference converges.

[0011] According to an embodiment of the present disclosure, the second intelligent model is trained in the following manner: obtaining a plurality of second sample data, where the plurality of second sample data includes second network call data of a plurality of sample users, and each second network call data includes: the second network signal intensity at a fifth moment, the registered network of the second network, the radio frequency band of the second network, and the second historical state of the second network call at a sixth moment, where the sixth moment is later than the fifth moment; inputting the second network signal intensity, the registered network of the second network, and the radio frequency band of the second network at the fifth moment into the second intelligent model to determine the second predicted state of the second network call at the sixth moment; adjusting the parameters of the second intelligent model according to the difference between the second predicted state and the second historical state of multiple second network call data until the difference converges.

[0012] According to an embodiment of the present disclosure, the first historical state includes successful first network calls and failed first network calls, and the proportions of successful first network calls and failed first network calls among multiple first sample data are within a third threshold range; the second historical state includes successful second network calls and failed second network calls, and the proportions of successful second network calls and failed second network calls among multiple second sample data are within a fourth threshold range.

[0013] According to an embodiment of the present disclosure, when there are multiple task data transmissions at a first moment, the method further includes: determining the priorities of multiple tasks according to the attribute information of the data transmission; when the priority of a first task is greater than the priority of a second task, determining a target network for the first task to perform data transmission at a second moment, and in response to the completion of the first task data transmission, determining a target network for the second task to perform data transmission at a third moment, where the third moment is later than the second moment.

[0014] A second aspect of the present disclosure provides a processing device, including: an acquisition module, configured to acquire network parameters of a terminal device for data transmission at a first moment, where the network parameters include first network signal parameters and second network signal parameters; a first determination module, configured to process the first network signal parameters based on a first intelligent model to determine a first prediction result, where the first prediction result represents the influence degree of the first network signal parameters at the first moment on data transmission at a second moment, and process the second network signal parameters based on a second intelligent model to determine a second prediction result, where the second prediction result represents the influence degree of the second network signal parameters at the first moment on data transmission at a second moment, and the second moment is later than the first moment; a second determination module, configured to determine a target network for data transmission at the second moment according to the first prediction result and the second prediction result.

[0015] A third aspect of the present disclosure provides an electronic device, including: one or more processors; a memory, configured to store one or more programs, where when the one or more programs are executed by the one or more processors, the one or more processors are caused to execute the above-mentioned processing method.

[0016] A fourth aspect of the present disclosure further provides a computer-readable storage medium, on which executable instructions are stored, and when the instructions are executed by a processor, the processor is caused to execute the above-mentioned processing method.

[0017] A fifth aspect of the present disclosure further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the above-mentioned processing method is implemented.

[0018] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. Brief Description of the Drawings

[0019] Through the following description of the embodiments of the present disclosure with reference to the accompanying drawings, the above and other objects, features, and advantages of the present disclosure will become clearer. In the drawings:

[0020] Figure 1 Schematically shows a schematic diagram of a voice call in the related art;

[0021] Figure 2 Schematically shows a flowchart of a processing method according to an embodiment of the present disclosure;

[0022] Figure 3 Schematically shows a flowchart of a first intelligent model training method according to an embodiment of the present disclosure;

[0023] Figure 4 Schematically shows a flowchart of a second intelligent model training method according to an embodiment of the present disclosure;

[0024] Figure 5 Schematically shows a schematic diagram of the principle of a processing method according to an embodiment of the present disclosure;

[0025] Figures 6A - 6C Schematically shows a statistical chart of different features for WiFi calls;

[0026] Figure 7A Schematically shows a statistical chart of different features for mobile calls according to the present disclosure;

[0027] Figure 7B Schematically shows the performance effect diagram of an intelligent model according to the present disclosure;

[0028] Figures 8A - 8C Schematically shows the effect diagram of using different classification algorithms according to the present disclosure;

[0029] Figures 9A - 9B Schematically shows the effect diagram of an intelligent model using different algorithms according to the present disclosure;

[0030] Figure 9C Schematically shows the prediction result diagram of an intelligent model using a decision tree according to an embodiment of the present disclosure;

[0031] Figure 10 Schematically shows a structural block diagram of a processing device according to an embodiment of the present disclosure;

[0032] Figure 11 Schematically shows a block diagram of an electronic device suitable for implementing a processing method according to an embodiment of the present disclosure. Detailed Description of the Embodiments

[0033] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present disclosure. In the following detailed description, for the sake of explanation, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure. However, evidently, one or more embodiments can also be implemented without these specific details. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessarily obscuring the concepts of the present disclosure.

[0034] The terms used herein are merely for describing specific embodiments and are not intended to limit the present disclosure. The terms "including", "comprising", etc. used herein indicate the presence of features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0035] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0036] In cases where expressions similar to "at least one of A, B, and C, etc." are used, generally, it should be interpreted according to the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include, but not be limited to, a system having only A, only B, only C, having A and B, having A and C, having B and C, and / or having A, B, and C, etc.).

[0037] Embodiments of the present disclosure provide a processing method and apparatus. Before introducing the technical solutions provided by the embodiments of the present disclosure, the related technologies involved in the present disclosure will be described first.

[0038] Currently, during data transmission, the network cannot be adjusted according to the network conditions, resulting in low data transmission efficiency and poor user experience.

[0039] In one example, voice over new radio (VoNR) technology, voice over long-term evolution (VoLTE) technology, and voice over Wi-Fi (VoWiFi) technology are three different voice call technologies. Among them, VoNR technology is a voice call technology based on a 5G network. VoLTE technology is a voice call technology based on a 4G network. VoWiFi is a technology for making voice calls based on a Wi-Fi network.

[0040] AsFigure 1 As shown, in VoLTE, the mobile phone connects to the P-CSCF, which is the operator's SIP VoIP server, through the SIP protocol. In VoWiFi, the mobile phone first establishes a secure connection with the ePDG, and the traffic is forwarded to the operator's network through a VPN (IPSec) tunnel to ensure the security of VoWiFi access. Establishing a VPN (IPSec) connection means connecting to the operator's dedicated VoWiFi VPN server. Then, it connects to the P-CSCF SIP server through the VPN tunnel in the same way as VoLTE.

[0041] Among them, IMS (IP Multimedia Subsystem) provides a unified IP multimedia service architecture and supports services such as VoLTE / VoWiFi. Its core functions include session management, user authentication, and service control. P-CSCF (Proxy-CallSession Control Function) refers to the "Proxy Call Session Control Function". It is a key component in the IMS (IP Multimedia Subsystem) architecture, mainly responsible for handling signaling and session management between user devices and the network. The main functions of P-CSCF include signaling forwarding, session establishment and termination, security guarantee, and policy control. This function is crucial for realizing efficient and reliable IP communication services (such as VoLTE and other multimedia services). ePDG (Evolved Packet Data Gateway) plays an important role in VoWiFi (Wi-Fi calling), allowing users to connect to the mobile network through Wi-Fi hotspots, enjoy high-definition audio and video calls, and seamlessly switch between VoLTE. The SIP protocol is the signaling protocol used to establish, modify, and terminate multimedia sessions (such as voice / video calls).

[0042] Because VoWifi (or Wifi Calling) involves more network elements than Volte / VoNR. For example, there is an additional EPDG for data forwarding, and dropped calls are more likely to occur under weak network or complex network conditions, resulting in a poor user experience.

[0043] To prevent call drops during voice calls. In one example, the operator requires the mobile device to pre - configure Volte or Vo Wifi priority according to its own network conditions. For example, Operator A gives priority to Volte, and Operator B gives priority to VoWifi. Although this method simplifies network management, it ignores the complexity and dynamic requirements of the actual network scenario. In another example, it is also possible to set the threshold for switching VoWiFi to VoLTE according to the operator's requirements. For example, Operator A stipulates that when the Wi - Fi signal strength (RSSI) drops to - 80 dBm (decibel - milliwatt), the system will perform the operation of switching from VoWiFi to VoLTE. This means that if the user is making a call using VoWiFi and the Wi - Fi signal becomes too weak (below - 80 dBm), the system will automatically switch to the more stable VoLTE network to maintain call quality. In this method, the threshold for network switching is fixed and cannot make dynamic decisions based on network conditions.

[0044] Embodiments of the present disclosure provide a processing method applied to a terminal device, including: obtaining network parameters for data transmission by the terminal device at a first moment, where the network parameters include a first network signal parameter and a second network signal parameter; processing the first network signal parameter based on a first intelligent model to determine a first prediction result, where the first prediction result characterizes the influence degree of the first network signal parameter at the first moment on data transmission at a second moment, and processing the second network signal parameter based on a second intelligent model to determine a second prediction result, where the second prediction result characterizes the influence degree of the second network signal parameter at the first moment on data transmission at the second moment, and the second moment is later than the first moment; determining a target network for data transmission at the second moment according to the first prediction result and the second prediction result.

[0045] The following will be described in detail Figure 2 ~ FIG. 6 for the processing method of the embodiments of the present disclosure.

[0046] Figure 2 Schematically shows an application scenario diagram of the processing method according to the embodiments of the present disclosure.

[0047] As Figure 2 shown, the processing method of this embodiment is applied to a terminal device, and the processing method of this embodiment includes operations S210~operation S230.

[0048] The terminal device in this disclosure can be referred to as a user equipment (UE), a mobile station (MS), a mobile terminal, a smart terminal, etc. The terminal device can communicate with one or more core networks via a radio access network (RAN). For example, the terminal device can be a mobile phone (or a "cellular" phone), a computer with a mobile terminal, etc. The terminal device can also be a portable, pocket-sized, handheld, computer-integrated, or vehicle-mounted mobile device, as well as a terminal device in a future New Radio (NR) network, which exchanges voice or data with the radio access network. Explanation of the terminal device: In this disclosure, the terminal device can also include a Relay, and anything that can communicate with a base station can be regarded as a terminal device. In this disclosure, a general UE will be introduced.

[0049] In operation S210, obtain the network parameters of the terminal device for data transmission at a first moment. The network parameters include a first network signal parameter and a second network signal parameter.

[0050] Exemplarily, data transmission refers to the process of transferring data from one device or system to another device or system via a network. This data can be text, audio, video, files, or other forms of information. The purpose of data transmission is to enable different devices to exchange and share information. For example, data transmission can be a voice call.

[0051] The network can be the type of multiple networks for data transmission. The network can be a virtual private network (VPN), a private network, a local area network, a mobile network, etc.

[0052] The network parameters can be multiple parameters related to network quality that can perform data transmission. For example, the network parameters can include signal strength, bandwidth, latency, load, etc. For example, in a voice call, the first network signal parameter can be the signal strength (RSRP), signal-to-noise ratio (SINR), network load, etc. of VoLTE (LTE network). The second network signal parameter can be the signal strength (RSSI), latency (Ping), packet loss rate, IPSec VPN connection status, etc. of VoWiFi (WiFi network). The embodiments of this disclosure do not make specific limitations on the type of the target network and the network signal parameters.

[0053] The first moment can be the current time point of data transmission, can also be a historical time point of data transmission, or can also be a historical period. When the first moment is a historical period, the network parameters can also include the change data of the network signal parameters of multiple networks for data transmission.

[0054] In operation S220, the first network signal parameter is processed based on the first intelligent model to determine a first prediction result, where the first prediction result represents the influence degree of the first network signal parameter at the first moment on data transmission at the second moment. Also, the second network signal parameter is processed based on the second intelligent model to determine a second prediction result, where the second prediction result represents the influence degree of the second network signal parameter at the first moment on data transmission at the second moment, and the second moment is later than the first moment.

[0055] Exemplarily, the first intelligent model uses the first network signal parameter to predict the influence of the first network signal at the first moment on data transmission at the second moment (the first prediction result).

[0056] The second intelligent model uses the second network signal parameter to predict the influence of the second network signal at the first moment on data transmission at the second moment (the second prediction result).

[0057] The prediction result can be the effect of data transmission using different network data at the second moment. For example, the prediction result can be the prediction of the probability of data transmission failure based on the quality of network signals of different networks at the second moment. For instance, the probability of call failure or success. Analyzing LTE network parameters predicts that the probability of call drop within the next 10 seconds is 5%. The prediction result can also be the prediction of data transmission signal parameters based on the quality of network signals of different networks at the second moment. For example, analyzing WiFi network parameters predicts that the delay may increase to 200 ms within the next 10 seconds.

[0058] For different network types, different intelligent models can be used for prediction, which can make full use of the specificities of different networks and improve the prediction accuracy. For different network types, the same intelligent model can also be used for prediction, which can improve the utilization rate of storage resources.

[0059] Refers to the future time window (such as the next 5 seconds) when the prediction takes effect, and the terminal needs to complete the network switching preparation before this moment.

[0060] In operation S230, based on the first prediction result and the second prediction result, the target network for data transmission at the second moment is determined.

[0061] Exemplarily, the target network can be the network with the best signal quality among multiple networks at the second moment. For example, by comparing the first prediction result and the second prediction result, according to their influence degrees (such as call drop risk, delay, throughput), the comprehensively optimal network is selected. If VoLTE predicts a high call drop risk (such as rapid attenuation of LTE signal), while VoWiFi predicts stability, then switch to VoWiFi. If VoWiFi predicts a sharp increase in delay (such as WiFi congestion), while the VoLTE load is low, then keep VoLTE.

[0062] It can be understood that the intelligent model predicts in advance the influence degree of multiple networks on data transmission at a future moment, and performs network switching according to the prediction results. On the one hand, it can judge in advance which network signal parameters are more beneficial to data transmission at a future moment, and select the best target network for data transmission. Thereby improving the stability and transmission efficiency of the network connection, and avoiding data loss or delay caused by network fluctuations. On the other hand, it can improve the working efficiency of the terminal device in different network environments, adjust the network selection strategy in real time, automatically select the appropriate network, and enhance the user experience.

[0063] In other embodiments, the first prediction result may also characterize the negative influence degree of the target network signal parameters at the first moment on data transmission at the second moment. The second prediction result may also characterize the positive influence degree of the target network signal parameters at the first moment on data transmission at the second moment.

[0064] The first intelligent model can predict the failure rate of data transmission of different networks at the second moment according to different network signal parameters at the first moment. The second intelligent model can predict the success rate of data transmission of different networks at the second moment according to different network signal parameters at the first moment. For example, the first intelligent model can determine the failure rate of data transmission of the first network at the second moment according to the first network signal parameter. The second intelligent model can determine the success rate of data transmission of the first network at the second moment according to the first network signal parameter.

[0065] As described above, in operation S230, according to the first prediction result and the second prediction result, determine the target network for data transmission at the second moment. In one implementable way, as Figure 4 shown, this operation may further include the operation of: when the first prediction result characterizes that the first network signal parameter at the first moment causes data transmission failure at the second moment, and the second prediction result characterizes that the second network signal parameter at the first moment causes data transmission success at the second moment, determine that the target network for data transmission at the second moment is the second network.

[0066] The result obtained by analyzing and predicting the first network signal parameters (such as network quality, signal strength, etc.) at the first moment. This result indicates that if the current first network continues to be used for data transmission at the second moment, the data transmission will fail. The result obtained by analyzing and predicting the second network signal parameters (such as network quality, signal strength, etc.) at the first moment. This result indicates that if the second network continues to be used for data transmission at the second moment, the data transmission will succeed.

[0067] In one example, continuing with the above data transmission for a voice call as an example. If the first prediction result indicates that the call drop probability of VoLTE (the first network) at the second moment is 90%, that is, it is easy to fail when using VoLTE for a call at the second moment. The second prediction result indicates that the call drop probability of VoWiFi (the second network) at the second moment is 30%, that is, it is not easy to fail when using VoWiFi for a call at the second moment. The probability of a successful call using the second network at the second moment is greater than that of using the first network for a call. Therefore, the target network for data transmission at the second moment can adopt VoWiFi. If the terminal device uses VoLTE for a call at the first moment, in the case where it is determined that the target network at the second moment adopts VoWiFi, the network for the call can be switched from VoLTE to VoWiFi.

[0068] In another example, if the first prediction result indicates that the call drop probability of VoLTE (the first network) at the second moment is 20%, that is, it is not easy to fail when using VoLTE for a call at the second moment. The second prediction result indicates that the call drop probability of VoWiFi (the second network) at the second moment is 80%, that is, it is easy to fail when using VoWiFi for a call at the second moment. The probability of a successful call using the first network at the second moment is greater than that of using the second network for a call. Therefore, the target network for data transmission at the second moment can adopt VoLTE. If the terminal device uses VoLTE for a call at the first moment, in the case where it is determined that the target network at the second moment adopts VoLTE, the network for the call can continue to be VoLTE without network switching.

[0069] As described above, in the case where the data transmission at the first moment uses the first network and the strength of the first network signal is higher than the first threshold, the first threshold characterizes the limit value of the strength of the first network signal when the network for data transmission is switched from the first network to the second network. The processing method of this embodiment may further include an operation: when the registration of the IP multimedia subsystem of the terminal device is successful, determine that the priority of the second network for data transmission at the second moment is higher than that of the first network, and switch the network for data transmission at the second moment from the first network to the second network.

[0070] Exemplarily, the first threshold may be the lowest threshold value set for the strength of the first network signal (such as the RSRP of LTE) when the terminal device performs a switch from the first network to the second network. For example, the first threshold can be set to RSRP ≥ -100dBm, indicating that only when the LTE signal strength is higher than -100dBm is it allowed to trigger the switch to VoWiFi.

[0071] In one example, continuing with the above data transmission being a voice call as an example. If the terminal device uses VoWiFi for a call at the first moment. At the first moment, the network parameters of VoWiFi (the first network) and VoLTE (the second network) are simultaneously obtained. The network parameters of VoWiFi are input into the first intelligent model, and a first prediction result is obtained: the call drop probability of VoWiFi (the first network) at the second moment is 90%. The network parameters of VoLTE are input into the second intelligent model, and a second prediction result is obtained: the call drop probability of VoLTE (the second network) at the second moment is 30%. The probability of a successful call using the second network at the second moment is greater than the probability of a successful call using the first network. Therefore, the target network for data transmission at the second moment can use VoLTE. At the same time, the signal strength of VoWiFi is -70 dBm, which is greater than the first threshold of -80 dBm. It is detected that the IP Multimedia Subsystem (IMS) of the terminal device is successfully registered. If the registration is successful, it is confirmed that VoWiFi is better than VoLTE for the call network at the second moment, and the current call network can be switched from VoWiFi to VoLTE.

[0072] In some other embodiments, when the data transmission at the first moment uses the second network and the signal strength of the second network is higher than a second threshold, where the second threshold represents the limit value of the signal strength of the second network when the network for data transmission is switched from the second network to the first network, the method further includes an operation: when the IP Multimedia Subsystem of the terminal device is successfully registered, determining that the priority of the first network for data transmission at the second moment is higher than that of the second network, and switching the network for data transmission at the second moment from the second network to the first network.

[0073] In one example, continuing with the above data transmission being a voice call as an example. If the terminal device uses VoLTE for a call at the first moment. At the first moment, the network parameters of VoLTE (the first network) and VoWiFi (the second network) are simultaneously obtained. The network parameters of VoLTE are input into the first intelligent model, and a first prediction result is obtained: the call drop probability of VoLTE (the first network) at the second moment is 90%. The network parameters of VoWiFi are input into the second intelligent model, and a second prediction result is obtained: the call drop probability of VoWiFi (the second network) at the second moment is 30%. The probability of a successful call using the second network at the second moment is greater than the probability of a successful call using the first network. Therefore, the target network for data transmission at the second moment can use VoWiFi. At the same time, the signal strength of VoLTE is -50 dBm, which is greater than the first threshold of -60 dBm. It is detected that the IP Multimedia Subsystem (IMS) of the terminal device is successfully registered. If the registration is successful, it is confirmed that VoWiFi is better than VoLTE for the call network at the second moment, and the current call network can be switched from VoLTE to VoWiFi.

[0074] It should be noted that in the case where the first network signal strength does not reach the first threshold at the first moment, an optimization operation is performed (i.e., predicting the data transmission effect of the second network and the first network at the second moment). If the prediction result shows that the second network is better than the first network, the second network is triggered to switch to the first network in advance. If the first network signal strength is lower than the first threshold at the first moment, it directly switches from the second network to the first network without predicting based on the parameters of the two networks.

[0075] It can be understood that by evaluating the network signal strength and priority at different moments, network switching can be achieved when it is predicted that the current network is not conducive to data transmission in the future, optimizing the quality and efficiency of data transmission.

[0076] As described above, for the above operation: it is determined that the priority of the second network for data transmission at the second moment is higher than that of the first network. In one implementable manner, this operation may further include an operation: performing a first process on the strength of the first network signal at the first moment to make the strength of the first network signal lower than the first threshold.

[0077] The first process can be through one or more means to ensure that the signal strength of the first network is reduced to a level lower than the first threshold. For example, the strength of the first network signal can be "penalized" to make the strength of the first network signal lower than the first threshold. That is, through the first process, the signal strength of the first network is adjusted to be lower than the preset "first threshold". If the signal strength of the first network is lower than the first threshold, it indicates that the quality of the first network is not sufficient to support good data transmission. Network switching can be triggered.

[0078] In an example, at the first moment, the system detects the signal strength of the first network. If the prediction result shows that the second network is more suitable for use at the second moment. To enable the second network at the second moment to obtain a higher priority, the system first performs a first process on the signal strength of the first network to reduce the signal strength of the first network and make it weaker. When the first network signal strength drops below the first threshold, the system will consider that the quality of the first network is no longer suitable to continue to undertake the data transmission task. Then it automatically switches the first network to the second network.

[0079] As described above, for the above operation: it is determined that the priority of the second network for data transmission at the second moment is higher than that of the first network. In another implementable manner, this operation may further include an operation: performing a second process on the strength of the second network signal at the first moment to make the strength of the second network signal higher than the second threshold, where the second threshold represents the minimum value of the second network signal strength in data transmission.

[0080] The second processing can be to ensure, by one or more means, that the signal strength of the second network is increased to a level higher than the second threshold. For example, the strength of the second network signal can be "rewarded" so that the strength of the second network signal is higher than the second threshold. That is, through the second processing, the signal strength of the second network is adjusted to be lower than the preset "second threshold". If the signal strength of the second network is higher than the second threshold, it indicates that the quality of the second network is sufficient to support good data transmission. Then the priority of the second network at the second moment is greater than that of the first network.

[0081] In one example, at the first moment, since the first network is used for data transmission at the first moment, it indicates that at the first moment, the signal strength of the first network is better than that of the second network. The system will not trigger a network switch at the first moment. At the first moment, the system will detect the signal strength of the second network. If the prediction result indicates that the second network is more suitable for use at the second moment. In order to enable the second network at the second moment to obtain a higher priority, the system first performs second processing on the signal strength of the second network to increase the signal strength of the second network and enhance it. When the signal strength of the second network drops to be higher than the second threshold, the system will consider that the quality of the second network is more suitable for continuing to undertake the data transmission task. Then the first network is automatically switched to the second network.

[0082] It can be understood that when obtaining the intelligent network prediction result regarding the data transmission effects of multiple networks at the second moment, by adjusting the signal strength of the first network and the signal strength of the second network, the signal strength of the target network at the second moment is made higher than the switching threshold, and the signal strength of the non-target network at the second moment is made lower than the switching threshold, so as to trigger a network switch in advance. To adjust the network selection according to the real-time network conditions, ensure that the most suitable network is selected for data transmission at different moments, and thus improve the adaptability and efficiency of device data transmission.

[0083] In some other embodiments, if it is determined that the priority of the second network for data transmission at the second moment is higher than that of the first network, the first processing can be simultaneously performed on the signal strength of the first network at the first moment to make the signal strength of the first network lower than the first threshold, and the second processing can be performed on the signal strength of the second network at the first moment to make the signal strength of the second network higher than the second threshold. That is, the two networks are processed respectively, so that the signal strength of the target network at the second moment is higher compared to the switching threshold, and the signal strength of the non-target network is lower compared to the switching threshold. Thus, a network switch is triggered in advance.

[0084] As described above, for the above operation: it is determined that the priority of the second network for data transmission at the second moment is higher than that of the first network. In another implementable manner, this operation may further include the operation of obtaining attribute information of the data transmission, where the attribute information characterizes the application scenario of the data transmission; when the attribute information meets a preset condition, updating the first threshold to a first target threshold, and the first target threshold is greater than the first threshold.

[0085] Exemplarily, the attribute information may be the application scenario of the data transmission. For example, voice call, picture download, etc.

[0086] The preset condition may characterize that the data transmission is real-time transmission. For example, a voice call is real-time data transmission. Picture download is non-real-time data transmission.

[0087] In one example, for real-time transmitted data, at the first moment, the system detects the signal strength of the first network (-60 dBm). If the prediction result indicates that the second moment is more suitable for using the second network. To enable the second network at the second moment to obtain a higher priority, the system first adjusts the first threshold (-70 dBm) for network switching of the first network to increase the threshold for network switching of the first network (the first target threshold, such as -50 dBm)). So that the signal strength of the first network at the first moment is lower than the first target threshold, then the system will consider that the quality of the first network is no longer suitable for continuing to undertake the data transmission task. Then the first network is automatically switched to the second network.

[0088] For non-real-time transmitted data, it shows that the requirement for the immediacy of data transmission is not very high. At this time, there is no need to perform network prediction to save the computing resources of the device.

[0089] As described above, for the above operation: it is determined that the priority of the second network for data transmission at the second moment is higher than that of the first network. In yet another implementable manner, this operation may further include the operation of updating the second threshold to a second target threshold when the attribute information meets a preset condition, and the second target threshold is less than the second threshold.

[0090] In one example, at the first moment, the system detects the signal strength of the second network (-80 dBm). If the prediction result indicates that the second moment is more suitable for using the second network. To enable the second network at the second moment to obtain a higher priority, the system first adjusts the second threshold (-70 dBm) for network switching of the second network to decrease the threshold for network switching of the second network (the second target threshold, such as -90 dBm)). So that the signal strength of the second network at the first moment is greater than the second target threshold, then the system will consider that the quality of the second network is suitable for continuing to undertake the data transmission task. Then the first network is automatically switched to the second network.

[0091] Figure 3 Schematically shows a flowchart of a first intelligent model training method according to an embodiment of the present disclosure.

[0092] As described above, for the first intelligent model. In one implementable manner, as Figure 3 shown, the first intelligent model can be trained through operations S310 to S330.

[0093] In operation S310, a plurality of first sample data are obtained. The plurality of first sample data includes first network call data of a plurality of sample users. Each first network call data includes: the first network signal strength at a third moment, the registered network of the first network, the wireless frequency band of the first network, and the first historical state of the first network call at a fourth moment, where the fourth moment is later than the third moment.

[0094] In operation S320, the first network signal strength at the third moment, the registered network of the first network, and the wireless frequency band of the first network are input into the first intelligent model to determine the first predicted state of the first network call at the fourth moment.

[0095] In operation S330, according to the difference between the first predicted state and the first historical state of multiple first network call data, the parameters of the first intelligent model are adjusted until the difference converges.

[0096] In an example, call data of multiple users when using the first network is obtained. Each call data may include the following information: the type of the first network (e.g., VoWiFi), the signal strength of the first network measured at the third moment, the wireless frequency band of the first network, and the first historical state (success or failure) of the first network call at a fourth moment later than the third moment.

[0097] The type of the first network, the signal strength of the first network measured at the third moment, the wireless frequency band of the first network, etc. are input into the first intelligent model, and the first intelligent model predicts the call state of the first network at the fourth moment (the first predicted state, e.g., success or failure). By comparing the difference between the first predicted state predicted by the first intelligent model and the actual first historical state, the parameters of the first intelligent model are continuously adjusted until the difference is less than a preset threshold (e.g., 0.1) to obtain the trained first intelligent model.

[0098] Figure 4 Schematically shows a flowchart of a second intelligent model training method according to an embodiment of the present disclosure.

[0099] As described above, for the second intelligent model. In one implementable manner, as Figure 4 shown, the second intelligent model can be trained through operations S410 to S430.

[0100] In operation S410, multiple second sample data are obtained. The multiple second sample data include second network call data of multiple sample users. Each second network call data includes: the second network signal strength at the fifth moment, the registered network of the second network, the wireless frequency band of the second network, and the second historical state of the second network call at the sixth moment, where the sixth moment is later than the fifth moment.

[0101] In operation S420, the second network signal strength at the fifth moment, the registered network of the second network, and the wireless frequency band of the second network are input into the second intelligent model to determine the second predicted state of the second network call at the sixth moment.

[0102] In operation S430, according to the difference between the second predicted state and the second historical state of multiple second network call data, the parameters of the second intelligent model are adjusted until the difference converges.

[0103] In one example, call data of multiple users when using the second network is obtained. Each call data may include the following information: the type of the second network (e.g., VoLTE), the signal strength of the second network measured at the fifth moment, the wireless frequency band of the first network, and the second historical state (success or failure) of the second network call at the sixth moment which is later than the fifth moment.

[0104] The type of the second network, the signal strength of the second network measured at the fifth moment, the wireless frequency band of the second network, etc. are input into the second intelligent model. The second intelligent model predicts the call state of the second network at the sixth moment (the second predicted state, e.g., success or failure). By comparing the difference between the second predicted state predicted by the second intelligent model and the actual second historical state, the parameters of the second intelligent model are continuously adjusted until the difference is less than a preset threshold (e.g., 0.1) to obtain the trained second intelligent model.

[0105] In some embodiments, the first historical state includes successful first network calls and failed first network calls, and the proportions of successful first network calls and failed first network calls in the multiple first sample data are within the third threshold range.

[0106] The second historical state includes successful second network calls and failed second network calls, and the proportions of successful second network calls and failed second network calls in the multiple second sample data are within the fourth threshold range.

[0107] It is found through experiments that when the proportion of successful network calls is 97% and the proportion of failed calls is 3%, overfitting occurs during the training process of the intelligent model.

[0108] Exemplarily, the proportion of successful first network calls and failed first network calls is within a third threshold range, which can be 4:6 - 6:4. Ensure that the proportion of successful first network calls and failed first network calls is balanced.

[0109] The proportion of successful second network calls and failed second network calls is within a fourth threshold range, which can be 4:6 - 6:4. Ensure that the proportion of successful second network calls and failed second network calls is balanced.

[0110] It can be understood that only by controlling the ratio of successful and failed network calls within a certain range can overfitting during model training be prevented, thereby improving the model prediction effect.

[0111] In some embodiments, when there are multiple task data transmissions at a first moment, the processing method of this embodiment further includes operations: determining the priorities of multiple tasks according to the attribute information of the data transmission; when the priority of a first task is greater than the priority of a second task, determining the target network for the first task to perform data transmission at a second moment, and in response to the completion of the first task data transmission, determining the target network for the second task to perform data transmission at a third moment, where the third moment is later than the second moment.

[0112] In one example, if there are multiple tasks performing data transmission at the first moment. The priorities of each task can be divided according to the application scenario of the data transmission. For example, the priority order: voice call > audio playback > picture download. The embodiments of the present disclosure do not make specific limitations on the setting of task priorities and can be set according to user needs.

[0113] For example, at the first moment, there are multiple tasks such as voice call, audio playback, and picture download. Then the priority of the voice call is greater than that of the audio playback, which is greater than that of the picture download. At this time, the task of the voice call can be processed preferentially, that is, first predict the network for the voice call to determine the target network for the voice call at the second moment. After the voice call is completed, then predict the network for the audio playback to determine the target network for the audio playback at the second moment. Predict the networks of multiple tasks at future moments in order of priority.

[0114] For the convenience of understanding the processing method of the embodiments of the present disclosure, the following will be combined with Figure 5 Describe the processing method of this embodiment in detail.

[0115] Figure 5 Schematically shows the schematic diagram of the processing method according to the embodiments of the present disclosure.

[0116] As Figure 5As shown below, taking a voice call as an example, at the first moment, the terminal makes a call using the first network. The target network parameters for data transmission of the terminal device are obtained. The network includes the first network and the second network. The parameters of the first network are input into the first intelligent model to obtain the influence result (success or failure) of the first network on the call at the second moment later than the first moment, that is, the first prediction result. The parameters of the second network are input into the second intelligent model to obtain the influence result (success or failure) of the second network on the call at the second moment later than the first moment, that is, the second prediction result. If the first prediction result indicates that the success rate of the second network for the call at the second moment is greater than that of the first network. Then the network is optimized by reducing the signal strength of the first network at the first moment to be lower than the first threshold (network switching threshold), and at the same time, increasing the signal strength of the second network at the first moment to be higher than the second threshold (network switching threshold). After the optimization is completed, if the adjusted signal strength of the first network is lower than the first threshold and the adjusted signal strength of the second network is higher than the second threshold, then the network switch is automatically triggered to ensure the success of data transmission at the second moment.

[0117] For different networks, it is necessary to collect the network parameters for the intelligent model prediction results. The following will combine Figures 6A - 7B to illustrate the comparison results of the parameters collected for different networks.

[0118] Figures 6A - 6C Schematically shows a statistical chart of different characteristics for WiFi calls.

[0119] During the call, the sample data of VoWiFi is analyzed, and the results are as follows:

[0120] In Figure 6A , it can be analyzed that when the signal strength of the wireless local area network is less than -55 dBm, the call is more likely to drop, and when the signal strength of the wireless local area network is greater than -55 dBm, it is more likely to drop. In Figure 6B , it can be analyzed that the delay of the signal of the wireless local area network is more concentrated when the call is successful, and the delay of the signal of the wireless local area network is more dispersed when the call drops. In Figure 6C , it can be analyzed that the packet loss rate of the wireless local area network is more concentrated when the call is successful, and the packet loss rate of the signal of the wireless local area network is more dispersed when the call drops.

[0121] Figure 7A Schematically shows a statistical chart of different characteristics for mobile calls according to the present disclosure; Figure 7B Schematically shows the performance effect diagram of the intelligent model according to the present disclosure.

[0122] In Figure 7A , it shows the influence degree of signal strength, quality, absorption rate, and power characteristics on the prediction results of mobile network calls.

[0123] In Figure 7B it can be analyzed that during the training stage of the intelligent model, the accuracy rate of call success prediction reaches 97.6%, and the accuracy rate of call drop prediction reaches 97.6%. This indicates that the accuracy rate of the prediction results of the intelligent model is relatively high.

[0124] It should be noted that the embodiments of the present disclosure include but are not limited to the parameter characteristics of the network shown in the above figures.

[0125] The model will have an impact on the prediction results when using different algorithms. The following will combine Figures 8A - 8C to illustrate the comparison of the model prediction results of different algorithms.

[0126] Figures 8A - 8C Schematically shows the effect diagrams of different classification algorithms according to the present disclosure.

[0127] For the historical call data of different types of networks, different classification algorithms are respectively used for data classification, and the call data is divided into three categories: class0 (CALL_ORIG_FAILURE) indicates: call drop (i.e., the call is initiated successfully but disconnected midway); class1 (CALL_DROP) indicates: call failure (i.e., the call is not initiated successfully); class2 (CALL_PERF_STATS_SUCCESS) indicates: call success.

[0128] The classification models respectively adopt: Logistic Regression (logistic regression), Linear Discriminant Analysis (LDA, linear discriminant analysis), Quadratic Discriminant Analysis (QDA, quadratic discriminant analysis).

[0129] From Figure 8A it can be analyzed that the results of classifying historical call data using the logistic regression algorithm perform very well for Class 1 and Class 2, and the AUC value and the indicators in the classification report are very high. The performance for Class 0 is relatively poor, with a lower recall rate and F1-score. This may be because the number of samples in Class 0 is relatively small, resulting in insufficient recognition ability of the model. Among them, the AUC value of class0 is 91%, the AUC value of class1 is 100%, and the AUC value of class2 is 94%.

[0130] From Figure 8BAnalysis shows that: For the results of classifying historical call data using the linear discrimination algorithm, it also performs well for Class 1 and Class 2, with high AUC values and metrics in the classification report. Its performance for Class 0 is slightly worse, especially with relatively low recall rate and F1-score, indicating that the model has certain difficulties in identifying Class 0. Among them, the AUC value of class0 is 93%, the AUC value of class1 is 99%, and the AUC value of class2 is 98%.

[0131] From Figure 8C Analysis shows that: For the results of classifying historical call data using the quadratic discrimination algorithm, it performs very well for Class 2, with both high AUC value and metrics in the classification report. Its performance for Class 1 is very poor, with an AUC value of only 0.66 and a recall rate of only 0.33, indicating that the model has difficulty in identifying Class 1. Its performance for Class 0 is also not ideal, with relatively low precision and recall rate. Among them, the AUC value of class0 is 95%, the AUC value of class1 is 66%, and the AUC value of class2 is 99%.

[0132] Based on the above analysis, the classification model using logistic regression performs the best among the three models, and is very accurate in identifying Class 1 and Class 2, but there is still room for improvement in identifying Class 0. The classification model using linear discrimination also performs well, but its recognition effect for Class 0 is not as good as that of Logistic Regression. The classification model using quadratic discrimination performs very well for Class 2, but its recognition effects for Class 1 and Class 2 are relatively poor, especially for Class 1. Therefore, the classification model using logistic regression is the optimal choice, but it needs to be further optimized to improve the recognition ability for Class 2. It can be considered to increase the sample size, adjust the model parameters or try other more complex models.

[0133] Figures 9A - 9B Schematically shows the effect diagrams of intelligent models using different algorithms according to the present disclosure; Figure 9C Schematically shows the prediction result diagram of the intelligent model using a decision tree according to an embodiment of the present disclosure.

[0134] The intelligent model respectively uses Algorithm 1: Support Vector Machine (SVM), and Algorithm 2: Decision Tree for comparative experiments. The results are as follows:

[0135] From Figure 9AAnalysis shows that when using a support vector machine to predict the call network results, the AUC value for class0 is 83%, the AUC value for class1 is 82%, and the AUC value for class2 is 86%.

[0136] From Figure 9B Analysis shows that when using an intelligent model with a decision tree to predict the call network results, the AUC value for class0 is 98%, the AUC value for class1 is 100%, and the AUC value for class2 is 99%.

[0137] From Figure 9C Analysis shows that when using an intelligent model with a decision tree, the prediction accuracy reaches 99%. The effect of the confusion matrix is also good.

[0138] Based on the above analysis, it can be obtained that using the decision tree algorithm for the intelligent model results in more accurate prediction.

[0139] Based on the above processing method, the present disclosure also provides a processing device. The following will be combined with Figure 10 to describe the device in detail.

[0140] Figure 10 The structural block diagram of the processing device according to an embodiment of the present disclosure is schematically shown.

[0141] As Figure 10 shown, the processing device 500 of this embodiment includes an acquisition module 510, a first determination module 520, and a second determination module 530.

[0142] The acquisition module 510 is used to acquire the network parameters of the terminal device for data transmission at the first moment. The network parameters include the first network signal parameter and the second network signal parameter. In one embodiment, the acquisition module 510 can be used to perform the operation S210 described above, which will not be elaborated here.

[0143] The first determination module 520 is used to process the first network signal parameter based on the first intelligent model to determine the first prediction result. The first prediction result characterizes the influence degree of the first network signal parameter at the first moment on the data transmission at the second moment, and process the second network signal parameter based on the second intelligent model to determine the second prediction result. The second prediction result characterizes the influence degree of the second network signal parameter at the first moment on the data transmission at the second moment, and the second moment is later than the first moment. In one embodiment, the first determination module 520 can be used to perform the operation S220 described above, which will not be elaborated here.

[0144] The second determination module 530 is used to determine the target network for data transmission at the second moment according to the first prediction result and the second prediction result. In one embodiment, the second determination module 530 can be used to perform the operation S230 described above, which will not be elaborated here.

[0145] According to an embodiment of the present disclosure, any one or more of the acquisition module 510, the first determination module 520, and the second determination module 530 may be combined and implemented in one module, or any one of them may be split into multiple modules. Alternatively, at least part of the functions of one or more of these modules may be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the acquisition module 510, the first determination module 520, and the second determination module 530 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on a substrate, a system on a package, an application specific integrated circuit (ASIC), or any other reasonable manner that can integrate or package circuits, etc., in hardware or firmware, or implemented in any one of the three implementation manners of software, hardware, and firmware, or in any appropriate combination of several of them. Alternatively, at least one of the acquisition module 510, the first determination module 520, and the second determination module 530 may be at least partially implemented as a computer program module, and when the computer program module is run, it can execute the corresponding functions.

[0146] Figure 11 A block diagram of an electronic device suitable for implementing a processing method according to an embodiment of the present disclosure is schematically shown.

[0147] As Figure 11 shown, the electronic device 600 according to an embodiment of the present disclosure includes a processor 601, which can perform various appropriate actions and processes according to a program stored in a read only memory (ROM) 602 or a program loaded from a storage section 608 into a random access memory (RAM) 603. The processor 601 may include, for example, a general microprocessor (such as a CPU), an instruction set processor, and / or a related chipset, and / or a dedicated microprocessor (such as an application specific integrated circuit (ASIC)), etc. The processor 601 may also include on board memory for caching purposes. The processor 601 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0148] In the RAM 603, various programs and data required for the operation of the electronic device 600 are stored. The processor 601, the ROM 602, and the RAM 603 are connected to each other via the bus 604. The processor 601 performs various operations of the method flow according to the embodiments of the present disclosure by executing the programs in the ROM 602 and / or the RAM 603. It should be noted that the programs can also be stored in one or more memories other than the ROM 602 and the RAM 603. The processor 601 can also perform various operations of the method flow according to the embodiments of the present disclosure by executing the programs stored in one or more memories.

[0149] According to an embodiment of the present disclosure, the electronic device 600 may further include an input / output (I / O) interface 605, and the input / output (I / O) interface 605 is also connected to the bus 604. The electronic device 600 may further include one or more of the following components connected to the I / O interface 605: an input part 606 including a keyboard, a mouse, etc.; an output part 607 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage part 608 including a hard disk, etc.; and a communication part 609 including a network interface card such as a LAN card, a modem, etc. The communication part 609 performs communication processing via a network such as the Internet. The drive 610 is also connected to the I / O interface 605 as needed. A removable medium 611, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 610 as needed so that a computer program read from it can be installed into the storage part 608 as needed.

[0150] The present disclosure also provides a computer-readable storage medium, which may be included in the device / device / system described in the above embodiments; or may exist separately without being assembled into the device / device / system. The above computer-readable storage medium carries one or more programs, and when the one or more programs are executed, the method according to the embodiments of the present disclosure is implemented.

[0151] According to an embodiment of the present disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, which may include, for example, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program, and the program may be used by or in combination with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present disclosure, the computer-readable storage medium may include one or more memories other than the above-described ROM 602 and / or RAM 603 and / or ROM 602 and RAM 603.

[0152] An embodiment of the present disclosure further includes a computer program product, which includes a computer program that contains program code for executing the method shown in the flowchart. When the computer program product runs in a computer system, the program code is used to enable the computer system to implement the processing method provided by the embodiment of the present disclosure.

[0153] When the computer program is executed by the processor 601, it executes the above functions defined in the system / apparatus of the embodiment of the present disclosure. According to an embodiment of the present disclosure, the above-described systems, apparatuses, modules, units, etc. may be implemented by computer program modules.

[0154] In one embodiment, the computer program may rely on tangible storage media such as optical storage devices and magnetic storage devices. In another embodiment, the computer program may also be transmitted and distributed in the form of a signal on a network medium, and be downloaded and installed through the communication part 609, and / or be installed from the removable medium 611. The program code included in the computer program may be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0155] In such an embodiment, the computer program may be downloaded and installed from the network through the communication part 609, and / or be installed from the removable medium 611. When the computer program is executed by the processor 601, it executes the above functions defined in the system of the embodiment of the present disclosure. According to an embodiment of the present disclosure, the above-described systems, devices, apparatuses, modules, units, etc. may be implemented by computer program modules.

[0156] According to embodiments of the present disclosure, program code for executing the computer programs provided by the embodiments of the present disclosure can be written in any combination of one or more programming languages. Specifically, these computing programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages include, but are not limited to, such as Java, C++, Python, the "C" language, or similar programming languages. The program code can be executed entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user's computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, by using an Internet service provider to connect through the Internet).

[0157] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and combinations of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0158] Those skilled in the art can understand that the features recited in the various embodiments and / or claims of the present disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly recited in the present disclosure. In particular, without departing from the spirit and teachings of the present disclosure, the features recited in the various embodiments and / or claims of the present disclosure can be combined and combined in various ways. All such combinations and / or combinations fall within the scope of the present disclosure.

[0159] The embodiments of the present disclosure have been described above. However, these embodiments are merely for illustrative purposes and are not intended to limit the scope of the present disclosure. Although the embodiments have been described separately above, this does not mean that the measures in each embodiment cannot be used advantageously in combination. The scope of the present disclosure is defined by the appended claims and their equivalents. Without departing from the scope of the present disclosure, those skilled in the art can make various substitutions and modifications, and these substitutions and modifications should fall within the scope of the present disclosure.

Claims

1. A processing method, applied to a terminal device, comprising: Acquire network parameters for data transmission performed by the terminal device at a first moment, the network parameters including first network signal parameters and second network signal parameters; Processing the first network signal parameter based on a first intelligent model to determine a first prediction result, the first prediction result characterizing the influence of the first network signal parameter at a first moment on data transmission at a second moment, and processing the second network signal parameter based on a second intelligent model to determine a second prediction result, the second prediction result characterizing the influence of the second network signal parameter at the first moment on data transmission at a second moment, the second moment being later than the first moment; A target network for data transmission at the second moment is determined according to the first prediction result and the second prediction result.

2. The method according to claim 1, determining a target network for data transmission at the second moment according to the first prediction result and the second prediction result, comprising: When the first prediction result represents that the first network signal parameter at the first moment causes data transmission failure at the second moment, and the second prediction result represents that the second network signal parameter at the first moment causes data transmission success at the second moment, it is determined that the target network used for data transmission at the second moment is the second network.

3. The method according to claim 2, wherein, when the data transmission at the first moment uses the first network and the strength of the first network signal is higher than a first threshold, the first threshold represents a limit value of the strength of the first network signal when the network for data transmission is switched from the first network to the second network, the method further comprises: When the IP multimedia subsystem of the terminal device is successfully registered, it is determined that the priority of the second network for data transmission at the second moment is higher than that of the first network, and the network for data transmission at the second moment is switched from the first network to the second network.

4. The method according to claim 3, determining that the priority of the second network performing data transmission at the second moment is higher than that of the first network, comprises: performing a first processing on the strength of the first network signal at the first moment so that the strength of the first network signal is lower than the first threshold; and / or A second process is performed on the strength of the second network signal at the first moment so that the strength of the second network signal is higher than the second threshold, where the second threshold represents a minimum value of the strength of the second network signal in data transmission.

5. The method according to claim 3, determining that the priority of the second network performing data transmission at the second moment is higher than that of the first network, comprises: Acquire attribute information of the data transmission, where the attribute information represents an application scenario of the data transmission; When the attribute information satisfies a preset condition, updating the first threshold to a first target threshold, where the first target threshold is greater than the first threshold; and / or When the attribute information satisfies a preset condition, the second threshold is updated to a second target threshold, and the second target threshold is smaller than the second threshold.

6. According to the method of claim 1, the first intelligent model is trained by: Acquire a plurality of first sample data, wherein the plurality of first sample data include first network call data of a plurality of sample users, and each of the first network call data includes: The first network signal strength, the registered network of the first network, the wireless frequency band of the first network at a third moment, and the first historical state of the first network call at a fourth moment, the fourth moment being later than the third moment; Inputting the first network signal strength at the third moment, the registered network of the first network, and the wireless frequency band of the first network into the first intelligent model to determine a first predicted state of the first network call at the fourth moment; According to the difference between the first predicted state and the first historical state of the plurality of first network call data, the parameters of the first intelligent model are adjusted until the difference converges.

7. According to the method of claim 1, the second intelligent model is trained by: Acquire a plurality of second sample data, wherein the plurality of second sample data includes second network call data of a plurality of sample users, and each of the second network call data includes: the second network signal strength, the registered network of the second network, the wireless frequency band of the second network at a fifth moment, and the second historical status of the second network call at a sixth moment, the sixth moment being later than the fifth moment; Inputting the second network signal strength at the fifth moment, the registered network of the second network, and the wireless frequency band of the second network into the second intelligent model to determine the second predicted state of the second network call at the sixth moment; According to the difference between the second predicted state and the second historical state of the plurality of second network call data, the parameters of the second intelligent model are adjusted until the difference converges.

8. The method according to claim 6 or 7, wherein the first historical status includes a first network call success and a first network call failure, and the proportion of the first network call success and the first network call failure in the plurality of first sample data is within a third threshold range; The second historical status includes a second network call success and a second network call failure, and the proportion of the second network call success and the second network call failure in the plurality of second sample data is within a fourth threshold range.

9. The method according to claim 3, wherein when there are multiple task data transmissions at the first moment, the method further comprises: Determining priorities of the plurality of tasks according to attribute information of the data transmission; In a case where the priority of the first task is greater than the priority of the second task, determining a target network for data transmission of the first task at the second moment, In response to completion of the data transmission of the first task, a target network for data transmission of the second task is determined at the third time, and the third time is later than the second time.

10. A processing device comprising: An acquisition module, configured to acquire network parameters for data transmission by the terminal device at a first moment, the network parameters including a first network signal parameter and a second network signal parameter; A first determination module is used to process the first network signal parameter based on a first intelligent model to determine a first prediction result, wherein the first prediction result represents the influence of the first network signal parameter at a first moment on data transmission at a second moment, and to process the second network signal parameter based on a second intelligent model to determine a second prediction result, wherein the second prediction result represents the influence of the second network signal parameter at the first moment on data transmission at a second moment, wherein the second moment is later than the first moment; A second determination module is used to determine a target network for data transmission at the second moment according to the first prediction result and the second prediction result.

Images