Service switching method and device, network equipment, terminal and storage medium

Abstract

Provided are a service switching method and apparatus, a network device, a terminal and a storage medium, the service switching method being applied to a first network device, the service switching method comprising: generating prediction state information according to first information, the first information being information related to quality of experience (QoE) of a user; according to the prediction state information, switching at least one service operated by the first terminal to a target network device for operation; and receiving second information sent by the target network equipment, wherein the second information is information related to service switching of the first terminal. According to the scheme of the invention, the problems of service interruption and QoE (Quality of Experience) decline during network switching of the first terminal can be avoided.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular, to a service switching method, apparatus, network device, terminal, and storage medium. Background Art

[0002] Quality of Experience (QoE) is the subjective feeling and experience of users. Some of its indicators can be presented as measurable indicators, such as latency, buffering time, etc. Therefore, QoE information is divided into subjective and objective parts. The subjective part is evaluated using the Mean Opinion Score (MOS), and the objective part is evaluated using measurable indicators.

[0003] The QoE report after traditional QoE measurement is directly transmitted to the QoE server, and the gNB does not perform parsing. However, current technologies have set new indicators for the Radio Access Network (RAN), such as buffer level, Playout delay for Media Start-up, etc. These indicators are RAN-side resolvable indicators, which can assist the RAN side in understanding the QoE information of the terminal (UE) and optimizing it.

[0004] However, when the terminal is switched from one gNB (gNB1) to another gNB (gNB2), there will be an interruption for a certain period of time during the handover process. Therefore, it is necessary to solve the problems of service interruption and QoE experience degradation that may occur during the handover process of the terminal. Summary of the Invention

[0005] The objective of the technical solution of the present invention is to provide a service switching method, apparatus, network device, terminal, and storage medium, so as to solve the problems of service interruption and QoE experience degradation that may occur during the handover process of the terminal in the prior art.

[0006] To solve the above technical problems, the embodiments of the present invention provide the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides a service switching method, which is applied to a first network device, and the method includes:

[0008] Generating predicted state information according to first information, where the first information is information related to Quality of Experience (QoE);

[0009] According to the predicted state information, switching at least one service running on a first terminal to run on a target network device;

[0010] Receive the second information sent by the target network device, where the second information is information related to service handover with the first terminal.

[0011] Optionally, before generating the predicted status information according to the first information, the method further includes:

[0012] Obtain the first information from the target device;

[0013] Wherein, the target device includes at least one of the following: at least one terminal, the first network device, at least one second network device;

[0014] Wherein, the first information includes at least one of the following:

[0015] Resource status information of the first network device;

[0016] Resource status information of the at least one second network device;

[0017] The number of Radio Resource Control (RRC) connections between the at least one terminal and the first network device;

[0018] The number of terminals accessed by the first network device;

[0019] Slice information of at least one service of the at least one terminal;

[0020] Radio Access Network visible Radio View Quality of Experience (RV QoE) information of at least one service of the at least one terminal;

[0021] The traffic of the at least one terminal.

[0022] Optionally, the predicted status information includes at least one of the following:

[0023] The predicted time of overload of the first network device;

[0024] At least one first terminal predicted to need network handover;

[0025] At least one target network device predicted to be switched to when the at least one first terminal performs network handover;

[0026] The predicted time of network handover of the at least one first terminal;

[0027] Quality of Experience (QoE) information of at least one service run by the at least one first terminal;

[0028] RV QoE information of at least one service run by the at least one first terminal;

[0029] Predicted QoE information required when the first terminal is switched from the first network device to the target network device;

[0030] Slicing information of at least one service run by the at least one first terminal;

[0031] Buffering degree information predicted to be required when the at least one first terminal is switched from the first network device to the target network device.

[0032] Optionally, the second information includes at least one of the following:

[0033] Buffering degree of at least one service when the first terminal receives the RRC reconfiguration message;

[0034] Buffering degree of at least one service when the first terminal sends the RRC reconfiguration complete message;

[0035] The first duration between the first terminal receiving the RRC reconfiguration message and the first terminal sending the RRC reconfiguration complete message;

[0036] The time when the first terminal is switched from the first network device to the target network device;

[0037] QoE information required when the first terminal is switched from the first network device to the target network device;

[0038] Performance information when the first terminal is switched from the first network device to the second network device.

[0039] Optionally, generating the predicted state information according to the first information includes:

[0040] Inputting the first information into at least one artificial intelligence (AI) model to obtain the predicted state information;

[0041] Among them, the predicted state information output by different AI models is different.

[0042] Optionally, before switching at least one service run by the first terminal to run on the target network device according to the predicted state information, the method further includes:

[0043] Sending the predicted state information to the target network device;

[0044] Receiving first indication information sent by the target network device according to the predicted state information;

[0045] Among them, the first indication information indicates that the target radio resource state meets the QoE information predicted to be required when the first terminal is switched from the first network device to the target network device, and the target radio resource state is the remaining radio resource state corresponding to the target network device at the time predicted for the first terminal to perform network handover.

[0046] Optionally, switching at least one service running on the first terminal to run on the target network device includes at least one of the following:

[0047] According to the predicted cache level information and / or the first indication information required when at least one first terminal switches from the first network device to the target network device, switching at least one service running on the first terminal to run on the target network device;

[0048] When the predicted cache level information required when at least one first terminal switches from the first network device to the target network device meets the target cache level, according to the first indication information, switching at least one service running on the first terminal to run on the target network device;

[0049] When the predicted cache level information required when at least one first terminal switches from the first network device to the target network device does not meet the target cache level, allocating a cache level for the first terminal according to the target cache level;

[0050] Wherein, the target cache level is the cache level required to switch at least one service running on the first terminal to run on the target network device without jamming;

[0051] The first indication information indicates that the target radio resource state meets the QoE information required when the predicted first terminal switches from the first network device to the target network device, and the target radio resource state is the remaining radio resource state of the target network device corresponding to the time when the predicted first terminal performs network switching.

[0052] Optionally, the method further includes:

[0053] Optimizing at least one AI model according to the second information;

[0054] Wherein, optimizing at least one AI model according to the second information includes at least one of the following:

[0055] When the cache level of at least one service is greater than a preset value when the first terminal sends an RRC reconfiguration complete message, optimizing the at least one AI model according to the cache level of at least one service when the first terminal receives the RRC reconfiguration message and the cache level of at least one service when the first terminal sends the RRC reconfiguration complete message;

[0056] When the caching level of at least one service is less than or equal to a preset value when the first terminal sends an RRC reconfiguration complete message, optimize the at least one AI model according to the caching level of at least one service when the first terminal receives the RRC reconfiguration message, the caching level of at least one service when the first terminal sends the RRC reconfiguration complete message, the QoE information required for the first terminal to switch from the first network device to the target network device, and the time when the first terminal switches from the first network device to the target network device.

[0057] In a second aspect, an embodiment of the present invention further provides a service switching method, which is applied to a target network device, and the method includes:

[0058] Receive the predicted status information sent by the first network device.

[0059] Optionally, the predicted status information includes at least one of the following:

[0060] The predicted time of overload of the first network device;

[0061] At least one first terminal predicted to need to perform network switching;

[0062] At least one target network device predicted to be switched to when the at least one first terminal performs network switching;

[0063] The predicted time when the at least one first terminal performs network switching;

[0064] The user experience quality QoE information of at least one service run by the at least one first terminal;

[0065] The RV QoE information of at least one service run by the at least one first terminal;

[0066] The predicted QoE information required for the first terminal to switch from the first network device to the target network device;

[0067] The slice information of at least one service run by the at least one first terminal;

[0068] The predicted caching level information required for the at least one first terminal to switch from the first network device to the target network device.

[0069] Optionally, before receiving the predicted status information sent by the first network device, the method further includes:

[0070] Send the resource status information of the target network device to the first network device.

[0071] Optionally, the method further includes:

[0072] Send first indication information to the first network device according to the predicted status information;

[0073] Among them, the first indication information indicates that the target radio resource status meets the QoE information required for the predicted first terminal to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status corresponding to the target network device at the time predicted for the first terminal to perform network switching.

[0074] Optionally, sending first indication information to the first network device according to the predicted status information includes:

[0075] Input the predicted status information and third information into a target artificial intelligence (AI) model to obtain the target radio resource status corresponding to the predicted overload time of the first network device output by the target AI model;

[0076] When the target radio resource status meets the QoE information required for the predicted first terminal to switch from the first network device to the second network device, send the first indication information to the first network device;

[0077] Among them, the third information includes at least one of the following:

[0078] Resource status information of the target network device;

[0079] The radio resource control (RRC) connection number between at least one terminal running services in the target network device and the target network device;

[0080] The number of terminals accessed by the target network device;

[0081] Slice information of at least one service of at least one terminal running services in the target network device;

[0082] Radio access network visible quality of experience (RV QoE) information of at least one service of at least one terminal running services in the target network device;

[0083] Traffic of at least one terminal running services in the target network device.

[0084] Optionally, the method further includes:

[0085] Receive fourth information sent by the first terminal;

[0086] Send second information to the first network device;

[0087] Among them, the fourth information includes at least one of the following:

[0088] The caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0089] The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0090] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message;

[0091] Wherein, the second information includes at least one of the following:

[0092] The fourth information;

[0093] The QoE information required for the first terminal to switch from the first network device to the target network device;

[0094] The time when the first terminal switches from the first network device to the target network device;

[0095] The performance information when the first terminal switches from the first network device to the target network device.

[0096] In a third aspect, an embodiment of the present invention further provides a service switching method, which is applied to a first terminal, and the method includes:

[0097] Sending at least one of the following information to the first network device:

[0098] The slice information of at least one service of the first terminal;

[0099] The radio access network visible user experience quality RV QoE information of at least one service of the first terminal;

[0100] The traffic of the first terminal.

[0101] Optionally, the method further includes:

[0102] Sending the fourth information to the target network device;

[0103] Wherein, the fourth information includes at least one of the following:

[0104] The caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0105] The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0106] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message.

[0107] Fourth aspect, an embodiment of the present invention further provides a service switching device, which is applied to a first network device. The device includes:

[0108] A first processing module, configured to generate predicted status information according to first information, where the first information is information related to Quality of Experience (QoE);

[0109] A second processing module, configured to switch at least one service running on a first terminal to run on a target network device according to the predicted status information;

[0110] A first receiving module, configured to receive second information sent by the target network device, where the second information is information related to service switching of the first terminal.

[0111] Fifth aspect, an embodiment of the present invention further provides a service switching device, which is applied to a target network device. The device includes:

[0112] A second receiving module, configured to receive the predicted status information sent by the first network device.

[0113] Sixth aspect, an embodiment of the present invention further provides a service switching device, which is applied to a first terminal. The device includes:

[0114] A first sending module, configured to send at least one of the following information to the first network device:

[0115] Slice information of at least one service of the first terminal;

[0116] Radio Access Network Visible Quality of Experience (RV QoE) information of at least one service of the first terminal;

[0117] Traffic of the first terminal.

[0118] Seventh aspect, an embodiment of the present invention further provides a network device, including: a processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, it implements the service switching method described in any item of the first aspect, or implements the service switching method described in any item of the second aspect when executed.

[0119] Eighth aspect, an embodiment of the present invention further provides a terminal, including: a processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, it implements the service switching method described in any item of the third aspect.

[0120] In a ninth aspect, an embodiment of the present invention further provides a readable storage medium, on which a program is stored, and when the program is executed by a processor, it implements the steps in the service switching method described in any one of the first aspect, or implements the steps in the service switching method described in any one of the second aspect, or implements the steps in the service switching method described in any one of the third aspect.

[0121] At least one of the above technical solutions of the present invention has the following beneficial effects:

[0122] In the solution of the present invention, the first network device generates prediction status information according to the first information (the first information is information related to the quality of experience QoE of the user), and according to the prediction status information, switches at least one service running on the first terminal to run on the target network device, that is, performs service switching according to the prediction status information, which can avoid the problems of service interruption and QoE experience degradation during network switching of the first terminal. After receiving the second information related to the service switching of the first terminal sent by the target network device, the QoE experience can be further improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0123] Figure 1 It is a functional framework diagram of AI / ML provided by the present invention;

[0124] Figure 2 It is a flowchart of the service switching method applied to the first network device provided by an embodiment of the present invention;

[0125] Figure 3 It is a flowchart of the service switching method applied to the target network device provided by an embodiment of the present invention;

[0126] Figure 4 It is a flowchart of the service switching method applied to the first terminal provided by an embodiment of the present invention;

[0127] Figure 5 It is a schematic structural diagram of the service switching device applied to the first network device provided by an embodiment of the present invention;

[0128] Figure 6 It is a schematic structural diagram of the service switching device applied to the target network device provided by an embodiment of the present invention;

[0129] Figure 7 It is a schematic structural diagram of the service switching device applied to the first terminal provided by an embodiment of the present invention;

[0130] Figure 8 It is a schematic structural diagram of the network device provided by an embodiment of the present invention;

[0131] Figure 9Schematic diagram of the structure of the terminal provided by the embodiments of the present invention. Detailed implementation manners

[0132] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0133] Before describing the specific implementation manners of the present invention, the following explanations are made first:

[0134] The 5G network generates a vast amount of data resources all the time. The existing 5G network technologies cannot efficiently use these vast amounts of data and produce valuable data analysis and inferences to provide strong support for the increasingly complex operation and maintenance of the 5G network. As an enabling technology, Artificial Intelligence (AI) can effectively improve the intelligent level of the 5G network by reasonably utilizing and exploring the 5G data resources. Moreover, the problems faced in the operation of the 5G network are complex and diverse. By performing efficient data processing and reasonable inferences based on the AI model, the performance of the 5G network can be effectively improved.

[0135] As users are the users of services, operators need to provide good service experiences for users. The traditional Quality of Service (QoS) evaluation indicators only reflect the service performance from the service technical level and cannot well reflect the subjective factors of users. Quality of Experience (QoE) is an evaluation indicator based on a user experience, which directly reflects the degree of recognition of the services or operations used by users in a certain objective environment.

[0136] In the current technology, AI / Machine Learning (ML) functional frameworks such as Figure 1As shown in the figure, it includes several parts such as Data Collection, Model Training, Model Inference, and Actor. Based on the collected training data (i.e., inputting Training Data into Model Training for training), data analysis and processing are carried out to obtain the trained model in Model Inference. Data Collection also includes Inference Data for Model Inference. The Output of Model Inference outputs the prediction results or corresponding strategies to Actor, and the data Feedback output by Actor is fed back to Data Collection. The data of Data Collection can also be used for Model Monitoring / Update, which helps to improve the network management and optimization efficiency and enhance the user experience. Therefore, how to obtain effective input data for the model and how to effectively utilize the output data of the model are issues that need to be considered when using AI to assist QoE optimization.

[0137] When the UE accesses the current gNB1 and initiates services such as Virtual Reality (VR) and Dynamic Adaptive Streaming over HTTP (DASH), if gNB1 is overloaded or the resources of the slice used by the UE are insufficient, the UE may be switched to another gNB2. There will be a certain period of interruption during the handover process. How to ensure that the service is not interrupted and the QoE experience does not decline during the handover is a problem that needs to be solved. In addition, since video services such as VR and DASH have high requirements for metrics such as latency, even if the load of gNB1 is not overloaded but reaches a certain level, it may not be able to guarantee the QoE of the UE service. Therefore, it is necessary to introduce an AI model to predict the future gNB resource situation (e.g., the time when gNB will be overloaded in the future) and the QoE metric requirements of the UE running services (e.g., the buffer level required before handover to ensure that the service is not interrupted during the handover process), and propose solutions in advance based on the prediction information and perform corresponding optimizations to ensure the service experience of the UE.

[0138] As Figure 2 As shown in the figure, an embodiment of the present invention provides a service handover method, which is applied to a first network device. The method includes:

[0139] Step 201: Generate predicted status information according to first information, where the first information is information related to Quality of Experience (QoE).

[0140] It should be noted that in this embodiment, the first network device is a source base station (Source gNB).

[0141] In this step, the Source gNB obtains information related to Quality of Experience (QoE), where the information related to Quality of Experience (QoE) includes information related to historical QoE and / or information related to real-time QoE.

[0142] Step 202: According to the predicted status information, switch at least one service running on the first terminal to run on a target network device.

[0143] It should be noted that in this embodiment, the target network device is a target base station (target gNB).

[0144] That is, in this step, based on the above predicted status information, the service on the first terminal is switched, which can avoid problems such as service interruption and QoE experience degradation during network switching of the first terminal.

[0145] Among them, the predicted status information includes at least one of the following:

[0146] The predicted time of overload of the first network device (that is, the time of predicting source gNB overload), where the predicted time can also be understood as a time period;

[0147] At least one first terminal for which network switching is predicted (that is, the predicted UE(s) to be switched);

[0148] At least one target network device to which the at least one first terminal is predicted to be switched during network switching, where the target network device can also be understood as the cell to which the first terminal is predicted to be switched during network switching (predicted target gNB(s) or cell);

[0149] The predicted time of network switching of the at least one first terminal;

[0150] Quality of Experience (QoE) information of at least one service running on the at least one first terminal;

[0151] Radio Access Network Visible Quality of Experience (RV QoE) information of at least one service running on the at least one first terminal ((RVQoE of UE(s) ongoing service));

[0152] The QoE information required for the predicted handover of the first terminal from the first network device to the target network device (the QoE required for the predicted handover of UE(s) ongoing service);

[0153] The slice information of at least one service running on the at least one first terminal (the slice information of UE(s) ongoing service);

[0154] The predicted caching degree information required for the predicted handover of the at least one first terminal from the first network device to the target network device.

[0155] Step 203: Receive second information sent by the target network device, where the second information is information related to the service handover of the first terminal.

[0156] In this step, after at least one service on the first terminal is handed over to run on the target network device, the first terminal records one or more items in the second information, and after the first terminal accesses the target gNB, reports the second information to the target gNB. The target network device obtains the second information related to the service handover of the first terminal, which can further improve the QoE experience.

[0157] In an optional embodiment, before generating the predicted status information according to the first information, the method further includes:

[0158] Obtain the first information from a target device, where the target device includes at least one of the following: at least one terminal, the first network device, at least one second network device;

[0159] Wherein, the above first terminal is one or more terminals among the at least one terminal, and the target network device is one or more network devices among the at least one second network device;

[0160] Wherein, the first information includes at least one of the following:

[0161] The resource status information of the first network device;

[0162] The resource status information of the at least one second network device;

[0163] The number of Radio Resource Control (RRC) connections between the at least one terminal and the first network device;

[0164] The number of terminals accessed by the first network device (the number of Active UE);

[0165] Slice information of at least one service of the at least one terminal (slice information of UE ongoing service);

[0166] Radio access network visible quality of experience RV QoE information of at least one service of the at least one terminal (RV QoE of UE ongoing service);

[0167] Traffic of the at least one terminal (UE traffic).

[0168] Optionally, the second information includes at least one of the following:

[0169] The buffering level of at least one service when the first terminal receives the RRC reconfiguration message (i.e., the buffer level of ongoing service when receiving the RRC reconfiguration message);

[0170] The buffering level of at least one service when the first terminal sends the RRC reconfiguration complete message (i.e., the buffer level of ongoing service when sending the RRC reconfiguration complete message);

[0171] The first duration between the first terminal receiving the RRC reconfiguration message and the first terminal sending the RRC reconfiguration complete message (i.e., the time difference between the RRC reconfiguration message and the RRC reconfiguration complete message, which can be understood as the handover time);

[0172] The time when the first terminal switches from the first network device to the target network device;

[0173] The QoE information required when the first terminal switches from the first network device to the target network device (e.g., RV QoE, legacy QoE report, etc. If the source node (Source gNB) configures the feedback time after handover in the request message, then feedback according to the configuration);

[0174] The performance information when the first terminal switches from the first network device to the second network device, that is, the performance information of the UE after handover (if the source node (Source gNB) configures the feedback time after handover in the request message, then feedback according to the configuration), where the performance information of the UE after handover includes at least one of the following: throughput, average packet delay, average packet loss rate.

[0175] In an optional embodiment, generating the predicted state information according to the first information includes:

[0176] Inputting the first information into at least one artificial intelligence (AI) model to obtain the predicted state information;

[0177] Among them, the predicted state information output by different AI models is different.

[0178] It should be noted that the AI model can be trained by the Operation Administration and Maintenance (OAM) network element side, and after the training is completed, the AI model is sent to the first network device.

[0179] The training data used for training the AI model includes a large amount of historical data from multiple aspects such as source gNB, neighboring base stations (neighbourgNB), and UEs. Among them, the training data includes at least one of the following:

[0180] Historical resource state information at each historical time point obtained from the Source gNB;

[0181] The number of historical active terminals (Active UEs) obtained from the Source gNB;

[0182] The number of historical Radio Resource Control (RRC) connections obtained from the Source gNB;

[0183] Historical UE traffic information obtained from the Source gNB;

[0184] The RV QoE report corresponding to the historical terminal obtained from the Source gNB;

[0185] Historical resource state information at each historical time point obtained from the neighbour gNB;

[0186] The number of historical active terminals (Active UEs) obtained from the neighbour gNB;

[0187] The number of historical Radio Resource Control (RRC) connections obtained from the neighbour gNB;

[0188] Historical UE traffic information obtained from the neighbour gNB;

[0189] RV QoE reports corresponding to historical terminals obtained from a neighbor gNB;

[0190] Services used by the UE at different historical time points obtained from the UE;

[0191] Slice information corresponding to the services used by the UE at different historical time points obtained from the UE;

[0192] RV QoE reports corresponding to the services used by the UE at different historical time points obtained from the UE;

[0193] Legacy QoE reports corresponding to the services used by the UE at different historical time points obtained from the UE.

[0194] It should be noted that there can be one or more AI models. If there is one AI model, all the above prediction status information is output by one AI model. If there are multiple AI models, for example, including two AI models, namely the first AI model and the second AI model. The input of the first AI model is the RV QoE reports corresponding to the services used by the UE at different historical time points obtained from the UE and the legacy QoE reports corresponding to the services used by the UE at different historical time points obtained from the UE, and the output is the RV QoE information of at least one service run by the at least one first terminal. The input of the second AI model is the historical resource status information at each historical time point obtained from the Source gNB and the number of historical active UEs obtained from the Source gNB, and the output is the predicted time of overload of the first network device / at least one first terminal that needs to perform network handover. The specific number of AI models is not limited in this embodiment.

[0195] It should also be noted that in the embodiments of the present invention, the AI model can also be understood as an ML model.

[0196] That is, the Source gNB uses the AI model based on the above first information to predict that after a period of time, the source gNB will be overloaded, or cannot meet the QoE requirements of all UE ongoing services, or the slice resources of some UE ongoing services are insufficient, and at least one of these needs to be switched to other gNBs.

[0197] In an alternative embodiment, before switching at least one service run by the first terminal to a target network device according to the prediction status information, the method further includes:

[0198] Send the predicted status information to the target network device, that is, the Source gNB and the Target gNB exchange one or more of the above predicted status information;

[0199] Receive the first indication information sent by the target network device according to the predicted status information, where the first indication information indicates that the target radio resource status meets the QoE information required for the first terminal to switch from the first network device to the target network device as predicted, and the target radio resource status is the remaining radio resource status of the target network device at the time predicted for the first terminal to perform network handover.

[0200] Specifically, the Target gNB obtains the radio resource status of the target network device at the time when the first terminal performs network handover through an AI model, an ML model or other prediction methods, and confirms the remaining radio resource status (i.e., the target radio resource status), and determines whether the target radio resource status can meet the QoE information (or slice resources) required for the first terminal to switch from the first network device to the target network device as predicted. Then, the target network device sends the first indication information to the first network device, that is, the first indication information fed back by the Target gNB indicates that the remaining radio resource status of the target network device meets the QoE information required for the first terminal to perform network handover.

[0201] Further, if the remaining radio resource status of the target network device does not meet the QoE information required for the first terminal to perform network handover, the Target gNB feeds back the second indication information to the first network device. The second indication information indicates that the remaining radio resource status of the target network device meets the QoE information required for the first terminal to perform network handover. The first network device stops switching at least one service running between the first terminal and the target network device to run on the target network device according to the second indication information.

[0202] In an alternative embodiment, the switching of at least one service running on the first terminal to run on the target network device includes at least one of the following:

[0203] According to the predicted buffer level information required for at least one first terminal to switch from the first network device to the target network device and / or the first indication information (the content indicated by this first indication information is the same as that indicated by the first indication information in the above optional embodiments and will not be elaborated here), switch at least one service run by the first terminal to run on the target network device. Specifically, the first network device obtains the buffer level collected in the past or in real time, inputs the historical buffer level and the predicted time of overload of the first network device into the AI model, and obtains the predicted buffer level information required for at least one first terminal to switch from the first network device to the target network device output by the AI model;

[0204] When the predicted buffer level information required for at least one first terminal to switch from the first network device to the target network device meets the target buffer level, according to the first indication information (the content indicated by this first indication information is the same as that indicated by the first indication information in the above optional embodiments and will not be elaborated here), switch at least one service run by the first terminal to run on the target network device, where the target buffer level is the buffer level required to switch at least one service run by the first terminal to the target network device without lagging (or interruption), and the predicted buffer level information required for at least one first terminal to switch from the first network device to the target network device is also obtained in the above manner. Exemplarily, the Source gNB compares the predicted buffer level information required for at least one first terminal to switch from the first network device to the target network device (i.e., the predicted buffer level) with the buffer level (target buffer level) that can ensure that the service is not interrupted / lagged during the handover process. If the predicted buffer level is greater than the buffer level value (i.e., the target buffer level) that can ensure that the service is not interrupted / lagged during the handover process, it is considered that the predicted buffer level can meet the target buffer level, and then according to the first indication information, switch at least one service run by the first terminal to run on the target network device.

[0205] In the case where the cache level information required when the predicted at least one first terminal is switched from the first network device to the target network device does not meet the target cache level, the cache level is allocated to the first terminal according to the target cache level, where the target cache level is the cache level required to switch at least one service running on the first terminal to run on the target network device without lag, and the cache level information required when the predicted at least one first terminal is switched from the first network device to the target network device is also obtained in the above manner. In the case where the cache level information (i.e., the predicted cache level) required when the predicted at least one first terminal is switched from the first network device to the target network device does not meet the target cache level, the Source gNB allocates the cache level to the first terminal according to the target cache level, that is, the source gNB comprehensively evaluates and re-adjusts the resource allocation situation, and allocates more resources to the UE (the first terminal) to be switched, so that the first terminal can cache more buffer levels before the handover, so that it has sufficient buffer levels to ensure that the handover service is not interrupted during the handover and does not affect the ongoing services of other UEs.

[0206] Further, after the Source gNB receives the above second information, the cache level of at least one service when the first terminal in the second information receives the RRC reconfiguration message and the cache level of at least one service when the first terminal sends the RRC reconfiguration complete message can be used to evaluate the accuracy of the above AI model, and the second information can also be used as the input information for subsequent AI model training / inference, that is, the AI model is optimized according to the second information.

[0207] Among them, optimizing the AI model according to the second information includes at least one of the following optimization methods:

[0208] Method 1: When the buffering level of at least one service is greater than a preset value (optionally, the preset value is 0) when the first terminal sends an RRC reconfiguration complete message, optimize the at least one AI model according to the buffering level of at least one service when the first terminal receives the RRC reconfiguration message and the buffering level of at least one service when the first terminal sends the RRC reconfiguration complete message. Specifically, when the buffering level of at least one service is greater than 0 when the first terminal sends the RRC reconfiguration complete message, it can be considered that at least one service of the first terminal is not interrupted during the handover process. The first network device can optimize the at least one AI model according to the difference between the buffering level of at least one service when the first terminal receives the RRC reconfiguration message and the buffering level of at least one service when the first terminal sends the RRC reconfiguration complete message. It should be noted that when the buffering level of at least one service is greater than 0 when the first terminal sends the RRC reconfiguration complete message, the at least one AI model may not be optimized and adjusted. However, to reduce the resource waste of the source node, according to the feedback buffer level difference (the difference between the buffering level of at least one service when the first terminal receives the RRC reconfiguration message and the buffering level of at least one service when the first terminal sends the RRC reconfiguration complete message), the subsequent predicted buffer level value can be reduced, that is, the value of the target buffering level is reduced.

[0209] Method 2: When the buffering level of at least one service is less than or equal to a preset value (optionally, the preset value is 0) when the first terminal sends an RRC reconfiguration complete message, optimize the at least one AI model according to the buffering level of at least one service when the first terminal receives the RRC reconfiguration message, the buffering level of at least one service when the first terminal sends the RRC reconfiguration complete message, the QoE information required for the first terminal to switch from the first network device to the target network device, and the time when the first terminal switches from the first network device to the target network device. Specifically, if the buffering level of at least one service is less than or equal to 0 when the first terminal sends the RRC reconfiguration complete message, the network needs to optimize the subsequent buffer level prediction according to the difference between the two reported buffer levels (the buffering level of at least one service when the first terminal receives the RRC reconfiguration message and the buffering level of at least one service when the first terminal sends the RRC reconfiguration complete message), the QoE information required for the first terminal to switch from the first network device to the target network device, and the time when the first terminal switches from the first network device to the target network device, etc. (that is, the buffer level (target buffering level) predicted by the current AI model is inaccurate, and more target buffering levels are required to ensure that the service is not interrupted).

[0210] It should be noted that considering that the current QoE optimization mainly reflects in the optimization of aspects such as QoS and is based on past and present information for optimization, introducing an AI model to predict future QoE information and using the predicted information for resource adjustment and early optimization can effectively improve the user's service experience.

[0211] As Figure 3 shown, an embodiment of the present invention further provides a service switching method, which is applied to a target network device. The method includes:

[0212] Step 301: Receive the predicted status information sent by the first network device.

[0213] Wherein, the predicted status information includes at least one of the following:

[0214] The predicted time when the first network device is overloaded;

[0215] At least one first terminal predicted to need network switching;

[0216] At least one target network device predicted to be switched to when at least one first terminal performs network switching;

[0217] The predicted time when at least one first terminal performs network switching;

[0218] The quality of experience QoE information of at least one service run by at least one first terminal;

[0219] The RV QoE information of at least one service run by at least one first terminal;

[0220] The predicted QoE information required when the first terminal is switched from the first network device to the target network device;

[0221] The slice information of at least one service run by at least one first terminal;

[0222] The predicted cache degree information required when at least one first terminal is switched from the first network device to the target network device. Optionally, before receiving the predicted status information sent by the first network device, the method further includes:

[0223] Sending the resource status information of the target network device to the first network device. Wherein, the resource status information of the target network device is used to generate the predicted status information.

[0224] Optionally, the method further includes:

[0225] Sending a first indication information to the first network device according to the predicted status information;

[0226] Wherein, the first indication information indicates QoE information required for a target wireless resource state to satisfy a prediction that a first terminal switches from the first network device to the target network device, and the target wireless resource state is a remaining wireless resource state of the target network device corresponding to a time when the first terminal is predicted to perform network handover.

[0227] Optionally, sending the first indication information to the first network device according to the predicted state information includes:

[0228] Inputting the predicted state information and third information into a target artificial intelligence (AI) model to obtain a target wireless resource state corresponding to a predicted overload time of the first network device output by the target AI model;

[0229] When the target wireless resource state satisfies QoE information required for a prediction that the first terminal switches from the first network device to the second network device, sending the first indication information to the first network device;

[0230] Wherein, the third information includes at least one of the following:

[0231] Resource state information of the target network device;

[0232] The number of radio resource control (RRC) connections between at least one terminal running a service in the target network device and the target network device;

[0233] The number of terminals accessed by the target network device;

[0234] Slice information of at least one service of at least one terminal running a service in the target network device;

[0235] Radio access network visible quality of experience (RV QoE) information of at least one service of at least one terminal running a service in the target network device;

[0236] The traffic volume of at least one terminal running a service in the target network device.

[0237] Optionally, the method further includes:

[0238] Receiving fourth information sent by the first terminal;

[0239] Sending second information to the first network device;

[0240] Wherein, the fourth information includes at least one of the following:

[0241] The caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0242] The caching level of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0243] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message;

[0244] Wherein, the second information includes at least one of the following:

[0245] The fourth information;

[0246] The QoE information required for the first terminal to switch from the first network device to the target network device;

[0247] The time when the first terminal switches from the first network device to the target network device;

[0248] The performance information when the first terminal switches from the first network device to the target network device.

[0249] It should be noted that the service switching method applied to the target network device provided in the embodiments of the present invention corresponds to the above-mentioned service switching method applied to the first network device. All embodiments of the above-mentioned service switching method applied to the first network device are applicable to the service switching method applied to the target network device and can achieve the same or similar technical effects.

[0250] As Figure 4 shown, the embodiments of the present invention further provide a service switching method applied to a first terminal. The method includes:

[0251] Step 401: Send at least one of the following information to the first network device:

[0252] The slice information of at least one service of the first terminal;

[0253] The radio access network visible user experience quality RV QoE information of at least one service of the first terminal;

[0254] The traffic of the first terminal.

[0255] Optionally, the method further includes:

[0256] Send the fourth information to the target network device;

[0257] Wherein, the fourth information includes at least one of the following:

[0258] The caching level of at least one service when the first terminal receives an RRC reconfiguration message;

[0259] The caching level of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0260] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message.

[0261] It should be noted that the service handover method applied to the first terminal provided in the embodiments of the present invention corresponds to the service handover method applied to the first network device above. All embodiments of the service handover method applied to the first network device above are applicable to the service handover method applied to the first terminal, and can achieve the same or similar technical effects.

[0262] The following uses a specific embodiment to illustrate the specific process of the service handover method:

[0263] Step 1: There are ongoing services (at least one service) of UE1, UE2, and UE3 (at least one terminal) in the current gNB1 (first network device).

[0264] Step 2: gNB1 predicts through the AI / ML model that gNB1 will be overloaded after a period of time, and predicts and outputs the following information (prediction status information):

[0265] The time when the predicted source gNB is overloaded;

[0266] Predict to switch UE1;

[0267] Predict to switch to target gNB2;

[0268] The RV QoE of UE1's ongoing service;

[0269] Predict the QoE required when switching UE1's ongoing service;

[0270] The slice information of UE1's ongoing service.

[0271] Step 3: gNB1 and gNB2 (target network device) interact with the predicted information (prediction status information) of the expected switched UE1;

[0272] Step 4: After receiving the information, gNB2 judges:

[0273] (1) If the QoE required for UE1's ongoing service can be satisfied, feedback the indication information that the required QoE for the handover can be satisfied, and then perform steps a - d:

[0274] Step a: Before the handover, gNB1 predicts whether the buffer level of the ongoing service of UE1 can meet the requirement of uninterrupted / smooth handover of services.

[0275] Step b: If it cannot be met, after comprehensive evaluation by gNB1, resources are reallocated to UE1 to ensure that UE1 has sufficient buffer level to ensure uninterrupted services during the handover process without affecting the ongoing services of UE2 and UE3.

[0276] Step c: If it can be met, the resource allocation situation is not adjusted accordingly.

[0277] Step d: UE1 records one or more of the following information and reports it after successfully accessing gNB2: the buffer level of the ongoing service when receiving the RRC reconfiguration message; the buffer level of the ongoing service when sending the RRC reconfiguration complete message; the handover time (the time difference between the handover request and the RRC reconfiguration complete message).

[0278] (2) It cannot meet the QoE required by the ongoing service of UE1, and feedbacks that the indicated information cannot meet the QoE required for handover. Reuse Steps 2 - 4 to predict another candidate target gNB for handover.

[0279] Step 5: gNB2 receives the UE-reported information and feeds it back to gNB1.

[0280] Step 6: gNB1 receives the feedback information from gNB2 and uses it for subsequent AI / ML-related predictions.

[0281] In the embodiment of the present invention, AI technology is combined with the existing QoE. By predicting future relevant information through AI, a solution is given in advance for the problem that may occur in the future, that is, due to the overload of the current base station and insufficient resources, the QoE required by the UE cannot be guaranteed. And based on the problems such as insufficient resources predicted by AI at future time points, information interaction between nodes and resource preparation of the target node are carried out in advance to avoid the situation that the resources of the target base station are also insufficient to meet the QoE required by the UE after handover. Before the handover, it is predicted whether the buffer level of the UE ongoing service during the handover can meet the requirements of uninterrupted and smooth services during the handover process, and resource reallocation and adjustment are carried out in advance. During the handover process based on AI prediction, UE records the changes in the buffer level before and after the handover and the handover time, and feeds them back to the source base station through the target base station to collect effective input information for subsequent AI prediction, so as to ensure that the UE service is smooth during the handover process and guarantee the user experience.

[0282] As shown Figure 5 in the figure, an embodiment of the present invention further provides a service switching device, which is applied to a first network device. The device includes:

[0283] A first processing module 501, configured to generate prediction status information according to first information, where the first information is information related to Quality of Experience (QoE);

[0284] A second processing module 502, configured to switch at least one service running on a first terminal to run on a target network device according to the prediction status information;

[0285] A first receiving module 503, configured to receive second information sent by the target network device, where the second information is information related to service switching of the first terminal.

[0286] Optionally, the device further includes:

[0287] A first obtaining module, configured to obtain first information from a target device;

[0288] where the target device includes at least one of the following: at least one terminal, the first network device, and at least one second network device;

[0289] where the first information includes at least one of the following:

[0290] Resource status information of the first network device;

[0291] Resource status information of at least one second network device;

[0292] The number of Radio Resource Control (RRC) connections between at least one terminal and the first network device;

[0293] The number of terminals accessed by the first network device;

[0294] Slice information of at least one service of at least one terminal;

[0295] Radio Visible Quality of Experience (RV QoE) information of at least one service of at least one terminal;

[0296] The traffic of at least one terminal.

[0297] Optionally, the prediction status information includes at least one of the following:

[0298] The predicted time when the first network device is overloaded;

[0299] At least one first terminal predicted to need network switching;

[0300] At least one target network device to which the at least one first terminal is predicted to switch when performing network switching;

[0301] The time when the at least one first terminal is predicted to perform network switching;

[0302] Quality of Experience (QoE) information of at least one service running on the at least one first terminal;

[0303] RV QoE information of at least one service running on the at least one first terminal;

[0304] Predicted QoE information required when the first terminal switches from the first network device to the target network device;

[0305] Slice information of at least one service running on the at least one first terminal;

[0306] Predicted cache level information required when the at least one first terminal switches from the first network device to the target network device.

[0307] Optionally, the second information includes at least one of the following:

[0308] The cache level of at least one service when the first terminal receives an RRC reconfiguration message;

[0309] The cache level of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0310] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message;

[0311] The time when the first terminal switches from the first network device to the target network device;

[0312] The QoE information required when the first terminal switches from the first network device to the target network device;

[0313] The performance information when the first terminal switches from the first network device to the second network device.

[0314] Optionally, the first processing module 501 includes:

[0315] A first processing unit, configured to input the first information into at least one artificial intelligence (AI) model to obtain the predicted state information;

[0316] Among them, the predicted state information output by different AI models is different.

[0317] Optionally, the device further includes:

[0318] A first information sending module, configured to send the predicted status information to the target network device;

[0319] A first information receiving module, configured to receive first indication information sent by the target network device according to the predicted status information;

[0320] Wherein, the first indication information indicates that the target radio resource status meets the QoE information required for the first terminal predicted to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device at the time predicted for the first terminal to perform network handover.

[0321] Optionally, the second processing module 502 includes at least one of the following:

[0322] A second processing unit, configured to switch at least one service run by the first terminal to run on the target network device according to the caching degree information and / or the first indication information required when at least one first terminal predicted to switch from the first network device to the target network device;

[0323] A third processing unit, configured to switch at least one service run by the first terminal to run on the target network device according to the first indication information when the caching degree information required when at least one first terminal predicted to switch from the first network device to the target network device meets the target caching degree;

[0324] A fourth processing unit, configured to allocate a caching degree for the first terminal according to the target caching degree when the caching degree information required when at least one first terminal predicted to switch from the first network device to the target network device does not meet the target caching degree.

[0325] Wherein, the target caching degree is the caching degree required to switch at least one service run by the first terminal to run on the target network device without stuttering;

[0326] The first indication information indicates that the target radio resource status meets the QoE information required for the first terminal predicted to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device at the time predicted for the first terminal to perform network handover.

[0327] Optionally, the device further includes:

[0328] An optimization module, configured to optimize at least one AI model according to the second information;

[0329] Wherein, the optimization module includes at least one of the following:

[0330] A first optimization unit, configured to optimize the at least one AI model according to the caching degree of at least one service when the first terminal receives an RRC reconfiguration message and the caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message, in a case where the caching degree of at least one service is greater than a preset value when the first terminal sends the RRC reconfiguration complete message;

[0331] A second optimization unit, configured to optimize the at least one AI model according to the caching degree of at least one service when the first terminal receives an RRC reconfiguration message, the caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message, the QoE information required when the first terminal switches from a first network device to the target network device, and the time when the first terminal switches from the first network device to the target network device, in a case where the caching degree of at least one service is less than or equal to a preset value when the first terminal sends the RRC reconfiguration complete message.

[0332] It should be noted that the service switching device applied to the first network device provided in the embodiments of the present invention is a device capable of executing the above-mentioned service switching method applied to the first network device. Therefore, all embodiments of the above-mentioned service switching method applied to the first network device are applicable to this device and can achieve the same or similar technical effects.

[0333] As Figure 6 shown, the embodiments of the present invention further provide a service switching device, which is applied to a target network device. The device includes:

[0334] A second receiving module 601, configured to receive prediction status information sent by a first network device.

[0335] Optionally, the prediction status information includes at least one of the following:

[0336] The predicted time when the first network device is overloaded;

[0337] At least one first terminal predicted to need network switching;

[0338] At least one target network device predicted to be switched to when the at least one first terminal performs network switching;

[0339] The predicted time when the at least one first terminal performs network switching;

[0340] The QoE information of at least one service run by the at least one first terminal;

[0341] The RV QoE information of at least one service run by the at least one first terminal;

[0342] The QoE information required for the predicted first terminal to be switched from the first network device to the target network device;

[0343] The slice information of at least one service run by the at least one first terminal;

[0344] The cache level information required for the predicted at least one first terminal to be switched from the first network device to the target network device.

[0345] Optionally, the device further includes:

[0346] A second information sending module, configured to send the resource status information of the target network device to the first network device.

[0347] Optionally, the device further includes:

[0348] A third information sending module, configured to send first indication information to the first network device according to the predicted status information;

[0349] Wherein, the first indication information indicates that the target radio resource status meets the QoE information required for the predicted first terminal to be switched from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device corresponding to the time when the predicted first terminal performs network handover.

[0350] Optionally, the third information sending module includes:

[0351] A fifth processing unit, configured to input the predicted status information and the third information into a target artificial intelligence (AI) model, and obtain the target radio resource status corresponding to the predicted overload time of the first network device output by the target AI model;

[0352] A first information sending unit, configured to send the first indication information to the first network device when the target radio resource status meets the QoE information required for the predicted first terminal to be switched from the first network device to the second network device;

[0353] Wherein, the third information includes at least one of the following:

[0354] The resource status information of the target network device;

[0355] The radio resource control (RRC) connection number between at least one terminal running a service in the target network device and the target network device;

[0356] The number of terminals accessed by the target network device;

[0357] Slice information of at least one service of at least one terminal running services in the target network device;

[0358] Radio Access Network visible User Experience Quality (RV QoE) information of at least one service of at least one terminal running services in the target network device;

[0359] Traffic of at least one terminal running services in the target network device.

[0360] Optionally, the device further includes:

[0361] A second information receiving module, configured to receive fourth information sent by a first terminal;

[0362] A fourth information sending module, configured to send second information to the first network device;

[0363] Wherein, the fourth information includes at least one of the following:

[0364] The caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0365] The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0366] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message;

[0367] Wherein, the second information includes at least one of the following:

[0368] The fourth information;

[0369] The QoE information required when the first terminal switches from the first network device to the target network device;

[0370] The time when the first terminal switches from the first network device to the target network device;

[0371] The performance information when the first terminal switches from the first network device to the target network device.

[0372] It should be noted that the service switching device provided in the embodiments of the present invention applicable to the target network device is a device capable of executing the above-mentioned service switching method applicable to the target network device. Then, all embodiments of the above-mentioned service switching method applicable to the target network device are applicable to this device and can achieve the same or similar technical effects.

[0373] As Figure 7 shown, the embodiments of the present invention further provide a service switching device, applicable to a first terminal. The device includes:

[0374] A first sending module 701, configured to send at least one of the following information to a first network device:

[0375] Slice information of at least one service of the first terminal;

[0376] Radio access network visible user experience quality RV QoE information of at least one service of the first terminal;

[0377] Traffic of the first terminal.

[0378] Optionally, the apparatus further includes:

[0379] A second sending module, configured to send fourth information to a target network device;

[0380] Wherein, the fourth information includes at least one of the following:

[0381] The caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0382] The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0383] A first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message.

[0384] It should be noted that the service switching apparatus applied to the first terminal provided in the embodiments of the present invention is a device capable of executing the above-mentioned service switching method applied to the terminal. Therefore, all embodiments of the above-mentioned service switching method applied to the first terminal are applicable to this device and can achieve the same or similar technical effects.

[0385] As Figure 8 shown, an embodiment of the present invention further provides a network device. The network device is a first network device, including: a processor 801; and a memory 803 connected to the processor 801 through a bus interface 802. The memory 803 is used to store programs and data used by the processor 801 when executing operations. The processor 801 calls and executes the programs and data stored in the memory 803.

[0386] Wherein, a transceiver 804 is connected to the bus interface 802 and is configured to receive and send data under the control of the processor 801. Specifically, the processor 801 is configured to read a program in the memory 803 and execute the following processes:

[0387] Generate predicted state information according to first information, where the first information is information related to user experience quality QoE;

[0388] According to the predicted status information, switch at least one service running on the first terminal to run on the target network device;

[0389] The transceiver 804 performs the following process:

[0390] Receive second information sent by the target network device, where the second information is information related to service switching of the first terminal.

[0391] Optionally, the processor 801 is further configured to:

[0392] Obtain first information from a target device;

[0393] Wherein, the target device includes at least one of the following: at least one terminal, the first network device, at least one second network device;

[0394] Wherein, the first information includes at least one of the following:

[0395] Resource status information of the first network device;

[0396] Resource status information of the at least one second network device;

[0397] The number of radio resource control (RRC) connections between the at least one terminal and the first network device;

[0398] The number of terminals accessed by the first network device;

[0399] Slice information of at least one service of the at least one terminal;

[0400] Radio access network visible radio access network quality of experience (RV QoE) information of at least one service of the at least one terminal;

[0401] Traffic of the at least one terminal.

[0402] Optionally, the predicted status information includes at least one of the following:

[0403] Predicted time of overload of the first network device;

[0404] At least one first terminal predicted to need network switching;

[0405] At least one target network device to be switched to when the at least one first terminal predicted to perform network switching;

[0406] Predicted time for the at least one first terminal to perform network switching;

[0407] Quality of experience (QoE) information of at least one service running on the at least one first terminal;

[0408] RV QoE information of at least one service run by the at least one first terminal;

[0409] Predicted QoE information required when the first terminal is switched from the first network device to the target network device;

[0410] Slice information of at least one service run by the at least one first terminal;

[0411] Predicted caching degree information required when the at least one first terminal is switched from the first network device to the target network device.

[0412] Optionally, the second information includes at least one of the following:

[0413] Caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0414] Caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0415] First duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message;

[0416] Time when the first terminal is switched from the first network device to the target network device;

[0417] QoE information required when the first terminal is switched from the first network device to the target network device;

[0418] Performance information when the first terminal is switched from the first network device to the second network device.

[0419] Optionally, the processor 801 is configured to:

[0420] Input the first information into at least one artificial intelligence (AI) model to obtain the predicted status information;

[0421] Among them, the predicted status information output by different AI models is different.

[0422] Optionally, the transceiver 804 is further configured to:

[0423] Send the predicted status information to the target network device;

[0424] Receive first indication information sent by the target network device according to the predicted status information;

[0425] Wherein, the first indication information indicates that the target radio resource status meets the QoE information required for the predicted first terminal to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device corresponding to the time when the first terminal is predicted to perform network switching.

[0426] Optionally, the processor 801 is specifically configured to perform at least one of the following:

[0427] According to the predicted caching degree information and / or the first indication information required for at least one first terminal to switch from the first network device to the target network device, switch at least one service running on the first terminal to run on the target network device;

[0428] In the case where the predicted caching degree information required for at least one first terminal to switch from the first network device to the target network device meets the target caching degree, according to the first indication information, switch at least one service running on the first terminal to run on the target network device;

[0429] In the case where the predicted caching degree information required for at least one first terminal to switch from the first network device to the target network device does not meet the target caching degree, allocate a caching degree for the first terminal according to the target caching degree.

[0430] Wherein, the target caching degree is the caching degree required to switch at least one service running on the first terminal to the target network device without lagging.

[0431] The first indication information indicates that the target radio resource status meets the QoE information required for the predicted first terminal to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device corresponding to the time when the first terminal is predicted to perform network switching.

[0432] Optionally, the processor 801 is further configured to:

[0433] Optimize at least one AI model according to the second information;

[0434] Wherein, optimizing at least one AI model according to the second information includes at least one of the following:

[0435] In the case where the caching degree of at least one service is greater than a preset value when the first terminal sends an RRC reconfiguration complete message, optimize the at least one AI model according to the caching degree of at least one service when the first terminal receives the RRC reconfiguration message and the caching degree of at least one service when the first terminal sends the RRC reconfiguration complete message;

[0436] When the buffering degree of at least one service is less than or equal to a preset value when the first terminal sends an RRC reconfiguration complete message, optimize the at least one AI model according to the buffering degree of at least one service when the first terminal receives the RRC reconfiguration message, the buffering degree of at least one service when the first terminal sends the RRC reconfiguration complete message, the QoE information required for the first terminal to switch from the first network device to the target network device, and the time when the first terminal switches from the first network device to the target network device.

[0437] Among them, in Figure 8 In, the bus architecture may include any number of interconnected buses and bridges, specifically, various circuits of one or more processors represented by the processor 801 and the memory represented by the memory 803 are linked together. The bus architecture can also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art, so they will not be further described herein. The bus interface provides an interface. The transceiver 804 can be multiple elements, that is, including a transmitter and a receiver, and provides a unit for communicating with various other devices on the transmission medium. The processor 801 is responsible for managing the bus architecture and general processing, and the memory 803 can store the data used by the processor 801 when performing operations.

[0438] An embodiment of the present invention further provides a network device, where the network device is a target network device, including: a processor; and a memory connected to the processor through a bus interface, where the memory is used to store programs and data used by the processor when performing operations, and the processor calls and executes the programs and data stored in the memory.

[0439] Among them, the transceiver is connected to the bus interface and is used to receive and send data under the control of the processor.

[0440] It should be noted that the target network device provided by the embodiment of the present invention is similar to the structure of the first network device shown in Figure 8 and will not be elaborated herein.

[0441] Specifically, the processor is used to read the program in the memory, and the transceiver performs the following processes:

[0442] Receive the predicted status information sent by the first network device.

[0443] Optionally, the predicted status information includes at least one of the following:

[0444] The predicted time when the first network device is overloaded;

[0445] At least one first terminal predicted to need network handover;

[0446] At least one target network device to which the at least one first terminal is predicted to be switched when performing network handover;

[0447] Predicted time for the at least one first terminal to perform network handover;

[0448] Quality of Experience (QoE) information of at least one service running on the at least one first terminal;

[0449] RV QoE information of at least one service running on the at least one first terminal;

[0450] Predicted QoE information required for the first terminal to switch from the first network device to the target network device;

[0451] Slice information of at least one service running on the at least one first terminal;

[0452] Predicted cache level information required for the at least one first terminal to switch from the first network device to the target network device.

[0453] Optionally, the transceiver is further configured to:

[0454] Send resource status information of the target network device to the first network device.

[0455] Optionally, the processor is configured to:

[0456] Send a first indication message to the first network device according to the predicted status information;

[0457] Wherein, the first indication message indicates that the target radio resource status meets the predicted QoE information required for the first terminal to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device corresponding to the predicted time of network handover of the first terminal.

[0458] Optionally, the processor is specifically configured to:

[0459] Input the predicted status information and third information into a target artificial intelligence (AI) model to obtain the target radio resource status corresponding to the predicted overload time of the first network device output by the target AI model;

[0460] When the target radio resource status meets the predicted QoE information required for the first terminal to switch from the first network device to the second network device, send the first indication message to the first network device;

[0461] Among them, the third information includes at least one of the following:

[0462] Resource status information of the target network device;

[0463] The number of Radio Resource Control (RRC) connections between at least one terminal operating a service in the target network device and the target network device;

[0464] The number of terminals accessed by the target network device;

[0465] Slice information of at least one service of at least one terminal operating a service in the target network device;

[0466] Radio Access Network Visible Radio Quality of Experience (RV QoE) information of at least one service of at least one terminal operating a service in the target network device;

[0467] The traffic of at least one terminal operating a service in the target network device.

[0468] Optionally, the transceiver is further configured to:

[0469] Receive fourth information sent by a first terminal;

[0470] Send second information to the first network device;

[0471] Among them, the fourth information includes at least one of the following:

[0472] The buffering degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0473] The buffering degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0474] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message;

[0475] Among them, the second information includes at least one of the following:

[0476] The fourth information;

[0477] The QoE information required for the first terminal to switch from the first network device to the target network device;

[0478] The time when the first terminal switches from the first network device to the target network device;

[0479] The performance information when the first terminal switches from the first network device to the target network device.

[0480] Such as Figure 9As shown in the figure, an embodiment of the present invention further provides a terminal. The terminal is a first terminal and includes: a processor 901; and a memory 903 connected to the processor 901 through a bus interface 902. The memory 903 is used to store programs and data used by the processor 901 when performing operations. The processor 901 calls and executes the programs and data stored in the memory 903.

[0481] Wherein, the transceiver 904 is connected to the bus interface 902 and is used to receive and send data under the control of the processor 901. Specifically, the processor 901 is used to read the program in the memory 903, and the transceiver 904 performs the following processes:

[0482] Send at least one of the following information to a first network device:

[0483] Slice information of at least one service of the first terminal;

[0484] Radio Access Network visible Radio Video Quality of Experience (RV QoE) information of at least one service of the first terminal;

[0485] Traffic of the first terminal.

[0486] Optionally, the transceiver 904 is further used to:

[0487] Send fourth information to a target network device;

[0488] Wherein, the fourth information includes at least one of the following:

[0489] The caching degree of at least one service when the first terminal receives an RRC reconfiguration message;

[0490] The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message;

[0491] The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message.

[0492] Wherein, in Figure 9Among them, the bus architecture may include any number of interconnected buses and bridges, specifically, various circuits of one or more processors represented by processor 901 and memory represented by memory 903 are linked together. The bus architecture can also link together various other circuits such as peripheral devices, voltage regulators, and power management circuits, etc., which are well known in the art, so they will not be further described herein. The bus interface provides user interface 905. The transceiver 904 can be multiple components, that is, including a transmitter and a receiver, and provides a unit for communicating with various other devices on the transmission medium. The processor 901 is responsible for managing the bus architecture and general processing, and the memory 903 can store the data used by the processor 901 when executing operations.

[0493] In addition, a specific embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps in the service switching method applied to the first network device described in any one of the above, or implements the steps in the service switching method applied to the first network device described in any one of the above, or implements the steps in the service switching method applied to the first terminal described in any one of the above.

[0494] In several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point, the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces, and the indirect coupling or communication connection of the device or unit can be in an electrical, mechanical or other form.

[0495] In addition, in each embodiment of the present invention, each functional unit can be integrated in a processing unit, or each unit can be physically included separately, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0496] The integrated unit implemented in the form of software functional units can be stored in a computer-readable storage medium. The above-mentioned software functional units are stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute some steps of the transceiver method described in various embodiments of the present invention. The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs.

[0497] The above are the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A service switching method, characterized in that, applied to a first network device, the method includes: generating prediction status information according to first information, where the first information is information related to Quality of Experience (QoE); switching at least one service running on a first terminal to run on a target network device according to the prediction status information; receiving second information sent by the target network device, where the second information is information related to service switching of the first terminal.

2. The method according to claim 1, characterized in that, before generating the prediction status information according to the first information, the method further includes: obtaining the first information from a target device; where the target device includes at least one of the following: at least one terminal, the first network device, at least one second network device; where the first information includes at least one of the following: resource status information of the first network device; resource status information of at least one second network device; the number of Radio Resource Control (RRC) connections between at least one terminal and the first network device; the number of terminals accessed by the first network device; slice information of at least one service of at least one terminal; Radio Visible Quality of Experience (RV QoE) information of at least one service of at least one terminal; traffic of at least one terminal.

3. The method according to claim 1, characterized in that, the prediction status information includes at least one of the following: the predicted time of overload of the first network device; at least one first terminal predicted to need network switching; at least one target network device predicted to be switched to when at least one of the first terminals performs network switching; the predicted time when at least one of the first terminals performs network switching; QoE information of at least one service running on at least one of the first terminals; RV QoE information of at least one service running on at least one of the first terminals; predicted QoE information required when the first terminal is switched from the first network device to the target network device; slice information of at least one service running on at least one of the first terminals; predicted cache degree information required when at least one of the first terminals is switched from the first network device to the target network device.

4. The method according to claim 1, characterized in that, the second information includes at least one of the following: the cache degree of at least one service when the first terminal receives an RRC reconfiguration message; the cache degree of at least one service when the first terminal sends an RRC reconfiguration complete message; the first duration between when the first terminal receives an RRC reconfiguration message and when the first terminal sends an RRC reconfiguration complete message; the time when the first terminal is switched from the first network device to the target network device; QoE information required when the first terminal is switched from the first network device to the target network device; performance information when the first terminal is switched from the first network device to a second network device.

5. The method according to claim 1, characterized in that, generating the prediction status information according to the first information includes: Input the first information into at least one artificial intelligence (AI) model to obtain the predicted status information; Among them, the predicted status information output by different AI models is different.

6. The method according to claim 1, characterized in that, before switching at least one service running on the first terminal to run on the target network device according to the predicted status information, the method further includes: sending the predicted status information to the target network device; receiving a first indication information sent by the target network device according to the predicted status information; wherein, the first indication information indicates that the target radio resource status meets the QoE information required when the predicted first terminal switches from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device at the time corresponding to the predicted network switching of the first terminal.

7. The method according to claim 1, characterized in that, switching at least one service running on the first terminal to run on the target network device includes at least one of the following: switching at least one service running on the first terminal to run on the target network device according to the predicted caching degree information and / or the first indication information required when at least one first terminal switches from the first network device to the target network device; when the predicted caching degree information required when at least one first terminal switches from the first network device to the target network device meets the target caching degree, switching at least one service running on the first terminal to run on the target network device according to the first indication information; when the predicted caching degree information required when at least one first terminal switches from the first network device to the target network device does not meet the target caching degree, allocating a caching degree for the first terminal according to the target caching degree; wherein, the target caching degree is the caching degree required to switch at least one service running on the first terminal to run on the target network device without lagging; the first indication information indicates that the target radio resource status meets the QoE information required when the predicted first terminal switches from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device at the time corresponding to the predicted network switching of the first terminal.

8. The method according to claim 1, characterized in that, the method further includes: optimizing at least one AI model according to the second information; wherein, optimizing at least one AI model according to the second information includes at least one of the following: when the caching degree of at least one service is greater than a preset value when the first terminal sends an RRC reconfiguration complete message, optimizing the at least one AI model according to the caching degree of at least one service when the first terminal receives the RRC reconfiguration message and the caching degree of at least one service when the first terminal sends the RRC reconfiguration complete message; When the buffering degree of at least one service is less than or equal to a preset value when the first terminal sends an RRC reconfiguration complete message, optimize the at least one AI model according to the buffering degree of at least one service when the first terminal receives the RRC reconfiguration message, the buffering degree of at least one service when the first terminal sends the RRC reconfiguration complete message, the QoE information required for the first terminal to switch from the first network device to the target network device, and the time when the first terminal switches from the first network device to the target network device.

9. A service switching method Characterized in that Applied to a target network device, the method includes: Receiving prediction status information sent by a first network device.

10. The method according to claim 9, Characterized in that The prediction status information includes at least one of the following: The predicted time of overload of the first network device; At least one first terminal predicted to need to perform network switching; At least one target network device to which the at least one first terminal is predicted to switch when performing network switching; The predicted time for the at least one first terminal to perform network switching; The user experience quality QoE information of at least one service run by the at least one first terminal; The RV QoE information of at least one service run by the at least one first terminal; The predicted QoE information required for the first terminal to switch from the first network device to the target network device; The slice information of at least one service run by the at least one first terminal; The predicted buffering degree information required for the at least one first terminal to switch from the first network device to the target network device.

11. The method according to claim 10, Characterized in that Before receiving the prediction status information sent by the first network device, the method further includes: Sending the resource status information of the target network device to the first network device.

12. The method according to claim 9, Characterized in that The method further includes: Sending first indication information to the first network device according to the prediction status information; Wherein, the first indication information indicates that the target radio resource status meets the QoE information required for the predicted first terminal to switch from the first network device to the target network device, and the target radio resource status is the remaining radio resource status of the target network device corresponding to the predicted time when the first terminal performs network switching.

13. The method according to claim 12, Characterized in that Sending the first indication information to the first network device according to the prediction status information includes: Inputting the prediction status information and third information into a target artificial intelligence AI model to obtain the target radio resource status corresponding to the predicted time of overload of the first network device output by the target AI model; Sending the first indication information to the first network device when the target radio resource status meets the QoE information required for the predicted first terminal to switch from the first network device to the second network device; Wherein, the third information includes at least one of the following: The resource status information of the target network device; The number of Radio Resource Control (RRC) connections between at least one terminal running services in the target network device and the target network device; The number of terminals accessed by the target network device; The slice information of at least one service of at least one terminal running services in the target network device; The Radio Access Network visible Radio View Quality of Experience (RV QoE) information of at least one service of at least one terminal running services in the target network device; The traffic of at least one terminal running services in the target network device.

14. The method according to claim 9, wherein, the method further comprises: receiving fourth information sent by a first terminal; sending second information to the first network device; wherein the fourth information comprises at least one of the following: The caching degree of at least one service when the first terminal receives an RRC reconfiguration message; The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message; The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message; wherein the second information comprises at least one of the following: The fourth information; The QoE information required when the first terminal switches from the first network device to the target network device; The time when the first terminal switches from the first network device to the target network device; The performance information when the first terminal switches from the first network device to the target network device.

15. A service switching method, wherein, applied to a first terminal, the method comprises: sending at least one of the following information to a first network device: The slice information of at least one service of the first terminal; The Radio Access Network visible Radio View Quality of Experience (RV QoE) information of at least one service of the first terminal; The traffic of the first terminal.

16. The method according to claim 15, wherein, the method further comprises: sending fourth information to a target network device; wherein the fourth information comprises at least one of the following: The caching degree of at least one service when the first terminal receives an RRC reconfiguration message; The caching degree of at least one service when the first terminal sends an RRC reconfiguration complete message; The first duration between the first terminal receiving an RRC reconfiguration message and the first terminal sending an RRC reconfiguration complete message.

17. A service switching device, wherein, applied to a first network device, the device comprises: A first processing module, configured to generate predicted status information according to first information, where the first information is information related to Quality of Experience (QoE); A second processing module, configured to switch at least one service run by a first terminal to run on a target network device according to the predicted status information; A first receiving module, configured to receive second information sent by the target network device, where the second information is information related to service switching of the first terminal.

18. A service switching device, wherein, applied to a target network device, the device comprises: A second receiving module, configured to receive predicted status information sent by a first network device.

19. A service switching device, wherein, Applied to a first terminal, the device includes: A first sending module, configured to send at least one of the following information to a first network device: Slice information of at least one service of the first terminal; Radio Access Network visible Radio Visible Quality of Experience (RV QoE) information of at least one service of the first terminal; Traffic of the first terminal.

20. A network device, characterized in that it includes: A processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, it implements the service switching method according to any one of claims 1 to 8, or implements the service switching method according to any one of claims 9 to 14 when executed.

21. A terminal, characterized in that it includes: A processor, a memory, and a program stored on the memory and executable on the processor. When the program is executed by the processor, it implements the service switching method according to claim 15 or 16.

22. A readable storage medium, characterized in that A program is stored on the readable storage medium. When the program is executed by a processor, it implements the steps of the service switching method according to any one of claims 1 to 8, or implements the steps of the service switching method according to any one of claims 9 to 14, or implements the steps of the service switching method according to claim 15 or 16.

Images