Handover method, device, and storage medium

Abstract

The present application relates to the technical field of terminals and the technical field of communications, and provides a handover method, a device, and a storage medium. When applied to a terminal device, the method comprises: sending predicted capability information, the predicted capability information indicating a handover (HO) capability predicted by an artificial intelligence (AI) model; receiving predicted configuration information, the predicted configuration information being used to indicate a trigger condition for triggering a predicted HO; and when the trigger condition is met, performing HO on the basis of a prediction result of the AI model. When applied to a network device, the method comprises: receiving predicted capability information, the predicted capability information indicating a handover (HO) capability predicted by an artificial intelligence (AI) model; and sending predicted configuration information on the basis of the predicted capability information, the predicted configuration information being used to indicate a trigger condition for triggering a predicted HO. The solution of the present application improves a handover success rate by performing handover prediction using an AI model, thereby ensuring mobile network quality, and improving user experience.

Description

Switching method, device and storage medium

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office of China on April 3, 2024, with application number 202410406809.8 and invention name “A switching method, device and storage medium”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of communication technology, and in particular to a switching method, a network device, a terminal device, and a computer storage medium. Background Art

[0003] Handover refers to the process of a terminal device moving from one base station coverage area to another, or switching to another voice channel due to external interference. Handover failure (HOF) typically occurs in three situations: premature HO, late HO, and HO to the wrong cell.

[0004] When HOF occurs, the communication service quality of the terminal device will be reduced, for example, the terminal device may be unable to send and receive information normally or make calls, etc. Therefore, a handover method with a high success rate is needed. Summary of the Invention

[0005] In order to solve the above problems, the present application provides a switching method, a network device, a terminal device and a computer storage medium, which improve the success rate of switching and ensure the quality of the mobile network.

[0006] In a first aspect, the present application provides a handover method, comprising: a terminal device transmitting prediction capability information indicating the ability of an artificial intelligence (AI) model to predict handover (HO). A network device, based on the received prediction capability information, transmits prediction configuration information to the terminal device, the prediction configuration information indicating a trigger condition for triggering the predicted HO. When the trigger condition is met, the terminal device performs HO based on the prediction result of the AI ​​model.

[0007] The solution provided by this application uses an AI model to predict handovers. For example, the AI ​​model can be used to determine the timing of handovers to avoid handovers occurring too late. Another example is that the AI ​​model can be used to predict that a handover command issued by the current network device will cause handovers to occur too early, and the AI ​​model can be used to determine the time to postpone handovers. Another example is that the AI ​​model can be used to predict that a handover command issued by the current network device will cause handovers to the wrong cell, and thus handovers can be stopped. Another example is that the AI ​​model can be used to predict the beam index of the beam after handover and the time of beam handover to ensure successful beam handovers. Compared with existing solutions, this improves the success rate of handovers, ensures the quality of mobile networks, and enhances user experience.

[0008] In a possible implementation, the triggering condition includes making a prediction according to a prediction period and / or satisfying a triggering event.

[0009] In a possible implementation, the triggering event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell, and the detection value is a detection value of a reference signal received power RSRP and / or a reference signal received quality RSRQ.

[0010] In a possible implementation, the triggering event includes one or more of the following: a HO failure rate in a first time interval is greater than a first threshold, the number of HO failures in a second time interval is greater than a first value, a probability of premature HO in a third time interval is greater than a second threshold, the number of premature HOs in a fourth time interval is greater than a second value, a probability of late HO in a fifth time interval is greater than a third threshold, the number of late HOs in a sixth time interval is greater than a third value, a probability of HO to an incorrect cell in a seventh time interval is greater than a fourth threshold, and the number of HOs to an incorrect cell in an eighth time interval is greater than a fourth value.

[0011] In one possible implementation, when a trigger condition is met, HO is performed according to a prediction result of an AI model, including: when the trigger condition is met, obtaining a HO prediction result using the AI ​​model; and sending HO prediction information, where the HO prediction information includes at least switching time information and target cell information.

[0012] In a possible implementation, the switching time information indicates a switching time corresponding to switching to at least one candidate cell, or a switching time corresponding to switching to one or more recommended cells. The target cell information indicates at least one candidate cell, or one or more recommended cells.

[0013] In one possible implementation, when a trigger condition is met, HO is performed according to a prediction result of the AI ​​model, specifically including: receiving a second HO command; triggering HO prediction and obtaining an HO prediction result using the AI ​​model, or, when the HO prediction is triggered by the trigger condition, obtaining the HO prediction result using the AI ​​model; when it is determined that a premature HO has occurred based on the switching time in the HO prediction result and the timing of receiving the second HO command, sending first indication information, where the first indication information is used to indicate that a premature HO has occurred and to request to maintain a radio resource control (RRC) connection with the source cell.

[0014] In one possible implementation, the first indication information is further used to indicate the target cell of HO; the first indication information is further used to indicate time information for postponing HO, the time information for postponing HO indicates a deferred switching time corresponding to switching to at least one candidate cell, or a deferred switching time corresponding to switching to the target cell indicated by the second HO command, or a deferred switching time corresponding to switching to the recommended target cell, and the at least one candidate cell is configured by the network device.

[0015] In one possible implementation, when the trigger condition is met, HO is performed according to the prediction result of the AI ​​model, specifically including: receiving a third HO command, the third HO command being used to at least indicate the target cell of HO; triggering HO prediction, obtaining the HO prediction result using the AI ​​model, or when the HO prediction is triggered by the trigger condition, obtaining the HO prediction result using the AI ​​model; when the target cell is determined to be an error cell according to the HO prediction result, not accessing the target cell; performing cell selection and completing the radio resource control RRC connection with the third cell, and carrying the source cell identifier and the target cell identifier in the RRC connection request message or the RRC connection establishment completion message sent to the third cell, so that the third cell sends a radio link failure RLF indication to the target cell, the RLF indication is used to enable the target cell to send a HO report to the source cell, and the HO report indicates that the type of the handover failure HOF is HO to the error cell.

[0016] In one possible implementation, the AI ​​model's ability to predict HO includes the ability to predict switching beams, and triggering events include: a detection value is lower than the threshold of the serving cell and / or higher than the threshold of the neighboring cell; or, the detection value is lower than the average beam quality of the serving cell and / or higher than the average beam quality of the neighboring cell; or, the detection value is lower than the average quality of the N best beams of the serving cell and / or higher than the average quality of the M best beams of the neighboring cell, where N and M are positive integers; the detection value is the detection value of the reference signal received power RSRP and / or the reference signal received quality RSRQ.

[0017] In one possible implementation, when the trigger condition is met, HO is performed according to the prediction result of the AI ​​model, specifically including: when the trigger condition is met, using the AI ​​model to obtain the beam switching prediction result; sending beam prediction information, the beam prediction information includes beam index information and beam switching time information.

[0018] In a possible implementation, the beam switching time information indicates a switching time to at least one beam corresponding to at least one candidate cell, or a switching time to a recommended beam corresponding to at least one candidate cell.

[0019] On the second aspect, the present application also provides a switching method applied to a network device, the method comprising: receiving prediction capability information, the prediction capability information indicating the ability of an artificial intelligence AI model to predict switching HO; based on the prediction capability information, sending prediction configuration information, the prediction configuration information is used to indicate the triggering conditions for triggering the predicted HO.

[0020] In a possible implementation, the triggering condition includes making a prediction according to a prediction period and / or satisfying a triggering event.

[0021] In a possible implementation, the triggering event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell, and the detection value is a detection value of a reference signal received power RSRP and / or a reference signal received quality RSRQ.

[0022] In one possible implementation, the triggering event includes one or more of the following:

[0023] The HO failure rate in the first time interval is greater than the first threshold, the number of HO failures in the second time interval is greater than the first value, the probability of premature HO in the third time interval is greater than the second threshold, the number of premature HOs in the fourth time interval is greater than the second value, the probability of late HO in the fifth time interval is greater than the third threshold, the number of late HOs in the sixth time interval is greater than the third value, the probability of HO to an incorrect cell in the seventh time interval is greater than the fourth threshold, and the number of HO to an incorrect cell in the eighth time interval is greater than the fourth value.

[0024] In a possible implementation manner, the method further includes: receiving HO prediction information, where the HO prediction information includes at least handover time information and target cell information; and sending a first HO command according to the HO prediction information.

[0025] In a possible implementation, the switching time information indicates a switching time corresponding to switching to at least one candidate cell, or a switching time corresponding to switching to one or more recommended cells. The target cell information indicates at least one candidate cell, or one or more recommended cells.

[0026] In a possible implementation, the method further includes: sending a second HO command; and receiving first indication information, where the first indication information is used to indicate that a premature HO occurs and to request to maintain a radio resource control (RRC) connection with the source cell.

[0027] In one possible implementation, the first indication information is further used to indicate the target cell of HO; the first indication information is further used to indicate time information for postponing HO, the time information for postponing HO indicates a deferred switching time corresponding to switching to at least one candidate cell, or a deferred switching time corresponding to switching to the target cell indicated by the second HO command, or a deferred switching time corresponding to switching to the recommended target cell, and the at least one candidate cell is configured by the network device.

[0028] In one possible implementation, the method also includes: sending a third HO command, where the third HO command is at least used to indicate the target cell of the HO; receiving a HO report sent by the target cell, where the HO report indicates that the type of the handover failure HOF is HO to an error cell, and the HO report is sent by the target cell after receiving the radio link failure RLF sent by the third cell. The third cell is the cell for the terminal device to perform cell selection and complete the radio resource control RRC connection.

[0029] In one possible implementation, the AI ​​model's ability to predict HO includes the ability to predict switching beams, and triggering events include: a detection value is lower than the threshold of the serving cell and / or higher than the threshold of the neighboring cell; or, the detection value is lower than the average beam quality of the serving cell and / or higher than the average beam quality of the neighboring cell; or, the detection value is lower than the average quality of the N best beams of the serving cell and / or higher than the average quality of the M best beams of the neighboring cell, where N and M are positive integers; the detection value is the detection value of the reference signal received power RSRP and / or the reference signal received quality RSRQ.

[0030] In a possible implementation, the method further includes: receiving beam prediction information, where the beam prediction information includes beam index information and beam switching time information; and sending a fourth HO command according to the beam prediction information.

[0031] In a possible implementation, the beam switching time information indicates a switching time to at least one beam corresponding to at least one candidate cell, or a switching time to a recommended beam corresponding to at least one candidate cell.

[0032] In a third aspect, the present application also provides a terminal device, which includes a processor and a memory, and the processor is coupled to the memory; the memory is used to store computer programs and / or instructions; the processor is used to execute the computer programs and / or instructions stored in the memory to implement the switching method provided in the first aspect and any one of the implementation methods of the first aspect.

[0033] In a fourth aspect, the present application further provides a network device comprising a processor and a memory, the processor being coupled to the memory; the memory being configured to store instructions. The processor is configured to execute a computer program or instruction stored in the memory to implement the switching method provided in the second aspect and any one of the implementations of the second aspect.

[0034] In a fifth aspect, the present application also provides a computer storage medium for storing a computer program. When the computer program is executed, it is used to implement the switching method provided by the first aspect and any one of the implementation methods of the first aspect, or to implement the switching method provided by the second aspect and any one of the implementation methods of the second aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] FIG1 is a schematic diagram of a communication system provided in an embodiment of the present application;

[0036] FIG2 is a schematic diagram of a process of premature HO leading to HOF according to an embodiment of the present application;

[0037] FIG3 is a schematic diagram of a process of causing HOF due to late HO according to an embodiment of the present application;

[0038] FIG4 is a schematic diagram of a process of HOF caused by HO to an incorrect cell according to an embodiment of the present application;

[0039] FIG5 is a flow chart of a switching method provided in an embodiment of the present application;

[0040] FIG6 is a schematic diagram of a management architecture of an AI model provided in an embodiment of the present application;

[0041] FIG7 is a flow chart of another switching method provided in an embodiment of the present application;

[0042] FIG8 is a flowchart of another switching method provided in an embodiment of the present application;

[0043] FIG9 is a flowchart of another switching method provided in an embodiment of the present application;

[0044] FIG10 is a flowchart of another switching method provided in an embodiment of the present application;

[0045] FIG11 is a schematic diagram of a switching device provided in an embodiment of the present application;

[0046] FIG12 is a schematic diagram of another switching device provided in an embodiment of the present application;

[0047] FIG13 is a schematic diagram of an electronic device provided in an embodiment of the present application;

[0048] FIG14 is a schematic diagram of another electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The "multiple" involved in the embodiments of the present application means greater than or equal to two. It should be noted that in the description of the embodiments of the present application, the words "first" and "second" are only used for the purpose of distinguishing the description and cannot be understood as indicating or implying relative importance or order.

[0050] In the embodiments of this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of this application should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0051] The embodiments of the present application are applied to communication systems, which may be second-generation (2G) communication systems, third-generation (3G) communication systems, LTE systems, fifth-generation (5G) communication systems, LTE and 5G hybrid architectures, 5G New Radio (5G NR) systems, and new communication systems that may emerge in future communication developments.

[0052] The network devices in the following embodiments of the present application are sometimes also referred to as network elements. The network devices may generally be base stations (including functional units of base stations, or a combination of functional units of base stations) or core network units, wherein the core network units may be functional units in the core network, including but not limited to access and mobility management function (AMF) units or session management function (SMF) units.

[0053] Refer to Figure 1, which is a schematic diagram of a communication system provided in an embodiment of the present application.

[0054] The communication system shown in FIG1 includes a base station 11 and a terminal device 12 .

[0055] In the embodiments provided in the present application, the base station can be any device with wireless transceiver functions, including but not limited to: an evolved base station (NodeB or eNB or e-NodeB, evolutionary Node B) in long term evolution (LTE), a base station (gNodeB or gNB) or a transmission receiving point (TRP) in new radio (NR), a base station of subsequent evolution of 3GPP, an access node in a Wi-Fi system, a wireless relay node, a wireless backhaul node, etc. The base station can be: a macro base station, a micro base station, a pico base station, a small station, a relay station, or a balloon station, etc. The base station can include one or more co-site or non-co-site transmission points (Transmission Reception Point, TRP). The base station can also be a wireless controller, a centralized unit (CU), and / or a distributed unit (DU) in a cloud radio access network (CRAN) scenario. The base station can communicate with a terminal device, or communicate with the terminal device through a relay station. The terminal device can communicate with multiple base stations of different technologies. For example, the terminal device can communicate with a base station that supports the LTE network, and can also communicate with a base station that supports the 5G network. It can also establish dual connections with a base station that supports the LTE network and a base station that supports the 5G network.

[0056] In the embodiments provided herein, the terminal device may be in various forms, such as a mobile phone, a tablet computer, a computer with wireless transceiver function, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal device in industrial control, a vehicle-mounted terminal device, a wireless terminal device in self-driving, a wireless terminal device in remote medical, a wireless terminal device in smart grid, a wireless terminal device in transportation safety, a wireless terminal device in smart city, a wireless terminal device in smart home, a wearable terminal device, etc. The terminal device may also be sometimes referred to as a terminal device, a user equipment, an access terminal device, a vehicle-mounted terminal device, an industrial control terminal device, a UE unit, a UE station, a mobile station, a mobile station, a remote station, a remote terminal device, a mobile device, a UE terminal device, a terminal device, a wireless communication device, a UE agent, or a UE device, etc. The terminal device may also be a fixed terminal device or a mobile terminal device.

[0057] Cell: also known as cellular cell, refers to the area covered by an access network device or a part of the antenna of the access network device in a cellular mobile communication system, in which the terminal can reliably communicate with the access network device through a wireless channel. The access network device in the embodiment of the present application may be a base station, a relay node (RN), an integrated access and backhaul (IAB) node, etc. The method provided in the embodiment of the present application is exemplified below by taking the access network device as an example of a base station, that is, the base stations in the following can be replaced by access network devices.

[0058] An Ad Hoc Network is a network that combines mobile communications and computer networks. It is a type of mobile computer network in which terminal devices can move freely within the network while maintaining communication.

[0059] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the following first describes the HO process and the specific reasons for the occurrence of HOF. When a terminal device switches between base stations, the base station before the switch is the source base station (source gNB, S-gNB), and the base station after the switch is the target base station (target gNB, T-gNB). Generally speaking, one base station corresponds to one cell, that is, a coverage area centered on the base station, identified by cellId. Therefore, S-gNB can also represent the source cell, and T-gNB can also represent the destination cell, and no distinction will be made in the following description.

[0060] The most basic HO is a soft handover, which occurs when a mobile phone in active communication switches from its current cell (called the source cell) and currently used channel to another available cell (called the destination cell) and channel. (The source and destination cells can be different cells or different sectors of the same cell.) This handover process is called an inter-cell handover. Another special case is when the source and destination cells are the same, or even the same sector, but the handover occurs only between used channels. This type of handover is called an intra-cell handover.

[0061] Refer to FIG2 , which is a flow chart illustrating how premature HO leads to HOF according to an embodiment of the present application.

[0062] S11: The terminal device completes HO.

[0063] S12: A radio link failure (RLF) occurs in the terminal device.

[0064] An RLF triggers a Radio Resource Control (RRC) re-establishment. This triggers cell selection, which results in the S-gNB being selected for the original cell.

[0065] S13: The terminal device and S-gNB complete RRC re-establishment.

[0066] This process is called premature HO.

[0067] S14: The S-gNB sends an RLF indication to the T-gNB.

[0068] In an ad hoc network, failure reporting can be used to optimize handover parameter configuration. After reestablishment is completed in S13, the source cell (source base station) sends an RLF indication to the target cell (target base station), indicating that the terminal device has experienced an RLF in the T-gNB.

[0069] S14: The T-gNB sends a HO report to the S-gNB.

[0070] After receiving the RLF indication, the T-gNB sends a HO report to the S-gNB. The HO report is used to report the HOF type. In this case, the reported HOF type is premature HO.

[0071] The S-gNB can modify and optimize the handover configuration parameters based on the HO report.

[0072] Refer to FIG3 , which is a flow chart illustrating how a too-late HO leads to a HOF according to an embodiment of the present application.

[0073] S21A: The terminal device performs HO.

[0074] S21B: During the HO process of the terminal device, RLF occurs between the terminal device and the S-gNB.

[0075] S22: The terminal device and T-gNB complete RRC re-establishment.

[0076] The terminal device performs RRC re-establishment, that is, performs cell selection, and then completes RRC re-establishment on the T-gNB. This process is called late HO.

[0077] S23: The T-gNB sends an RLF indication to the S-gNB.

[0078] In an ad hoc network, the target cell T-gNB that completes RRC re-establishment sends an RLF indication to the source cell S-gNB, indicating that an RLF has occurred between the terminal and the source cell S-gNB. This indicates that the handover failure was caused by a late handover. After receiving the RLF indication, the S-gNB can modify or optimize handover configuration parameters.

[0079] Refer to FIG4 , which is a schematic diagram of a process of HOF caused by HO to an incorrect cell according to an embodiment of the present application.

[0080] S31: The terminal device HO is completed.

[0081] When making a handover decision, the S-gNB determines an inappropriate T-gNB (second cell), causing the terminal device to switch to the T-gNB.

[0082] S32: Terminal device RLF.

[0083] RLF occurs shortly after the terminal device switches to the T-gNB.

[0084] S33: The terminal device and the third gNB complete RRC connection re-establishment.

[0085] At this time, the terminal device performs RRC re-establishment, that is, the terminal device performs cell selection, and the result of the cell selection is the third gNB.

[0086] S34: The third gNB sends an RLF indication to the T-gNB.

[0087] In an ad hoc network, after the terminal device completes re-establishment on the third gNB, the third gNB sends an RLF indication to the T-gNB to indicate that an RLF has occurred between the terminal device and the T-gNB.

[0088] S35: The T-gNB sends a HO report to the S-gNB.

[0089] After receiving the RLF indication, the T-gNB sends a HO report to the S-gNB. The HO report reports the type of HOF. In this case, the reported HOF type indicates handover to an incorrect cell. The S-gNB modifies and optimizes the handover configuration parameters based on the HO report.

[0090] When the above types of HOF occur, the communication service quality of the terminal device will be reduced, for example, the terminal device may not be able to send or receive information normally or make calls. To solve the above technical problems, this application proposes a method for assisted switching using an artificial intelligence (AI) model, a network device, a terminal device, and a computer storage medium, which can reduce the probability of switching failure.

[0091] The technical solutions provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings. The time information in the following embodiments may also be referred to as opportunity information.

[0092] See Figure 5, which is a flowchart of a switching method provided in an embodiment of the present application.

[0093] S51: The terminal device sends prediction capability information.

[0094] The solution provided in the embodiment of the present application can assist HO through an AI model. The AI ​​model is deployed on the terminal device, that is, HO prediction is performed through terminal device behavior and radio resource management (RRM) measurements.

[0095] In one possible implementation, the AI ​​model can be built and trained on a computing device cluster and then downloaded or updated locally by the terminal device.

[0096] In another possible implementation, the AI ​​model can be built and trained directly on the terminal device.

[0097] See Figure 6, which is a schematic diagram of the management architecture of an AI model provided in an embodiment of the present application.

[0098] When the terminal device performs HO prediction, the main prediction capabilities may include: the ability to predict the type of HOF and the ability to predict the HOF level.

[0099] The ability to predict the type of HOF specifically includes: the ability to predict the timing of HO to avoid late handover, the ability to predict premature HO and determine the time to postpone handover, and the ability to predict HO to a wrong cell.

[0100] The ability to predict the HOF level includes: cell-level capability and beam-level capability.

[0101] When the HOF-level capabilities of the AI ​​model include beam-level capabilities, the terminal device can use the AI ​​model to predict information such as the beam index and the timing of beam switching. The switched beam can be a beam in the same cell or a beam in a different cell.

[0102] The capabilities of the AI ​​model may include one or more of the above capabilities. For example, the first AI model may be capable of predicting the timing of HO to avoid late handover and the ability to predict HO to the wrong cell. For another example, the second AI model may be capable of predicting premature HO and determining a delayed handover time, as well as beam-level capabilities. The embodiments of this application do not limit the specific capabilities of the AI ​​model.

[0103] It is understandable that in order to enable the AI ​​model to have the corresponding capabilities, the AI ​​model needs to be trained first.

[0104] The training process is tailored to the specific purpose of the AI ​​model. The following example illustrates the AI ​​model's ability to predict HO to an incorrect cell. The training process for other capabilities is similar and will not be detailed here.

[0105] The data collection module collects training data, which is then used by the model training module to train the AI ​​model. During AI model training, the cell selection process is repeated numerous times, and the AI ​​model parameters are continuously adjusted until the cell selection results meet preset requirements. Model training is considered complete. The trained model can be stored, for example, on the computing device used to train the model or deployed on a terminal device.

[0106] During the use of AI models, the AI ​​models may no longer be applicable due to factors such as changes in the scenario of the terminal device (from a low-speed mobile environment to a high-speed mobile environment, or from a high-speed mobile environment to a low-speed mobile environment) or changes in the performance of the terminal device (such as aging or replacement of the terminal device antenna or RF circuit). In this case, model management is required.

[0107] When the model management module performs model management, it can perform model management based on the monitoring data obtained by the data collection module. The monitoring data may include data reflecting the characteristics of the scenario in which the terminal device is located and data reflecting the performance of the terminal device, etc. When the monitoring data exceeds the scope of use of the AI ​​model, corresponding management operations can be performed. The model management module can also determine the current prediction accuracy of the AI ​​model based on the reasoning output results of the reasoning module, and then feedback the results to the model training module, and the model training module determines whether to update and retrain the model. Alternatively, the model management module can determine the current prediction accuracy of the AI ​​model based on the reasoning output results of the reasoning module. When the prediction accuracy of the AI ​​model does not meet the requirements, a retraining request is sent to the model training module to enable the model training module to determine whether to update and retrain the model.

[0108] The management operations of the model management module on the AI ​​model may include but are not limited to activation of the AI ​​model, deactivation of the AI ​​model, selection of the AI ​​model, activation of the AI ​​model function, cessation of the AI ​​model function, and rollback of the AI ​​model.

[0109] Activating an AI model refers to enabling it, while deactivating it refers to terminating its activation. Selecting an AI model refers to choosing an appropriate one from multiple available AI models. Enabling and disabling AI model functions refers to enabling or disabling some or all of the functions of an AI model with multiple capabilities. AI model fallback refers to deactivating an AI model and implementing HO using traditional HO methods.

[0110] The inference module is the AI ​​model's operational phase. It can be deployed on terminal devices, which use the acquired inference data and AI models to implement corresponding functions. The inference module can request the scheduling of AI models from the model management module.

[0111] In one possible implementation, the data collection module in Figure 6 can collect training data using a computing device cluster, collect detection data using a computing device cluster and / or a terminal device, and collect inference data using a terminal device. The model management module can also be implemented by a computing device cluster. The model management module and inference module can also be implemented by a terminal device. The model storage module can be a module that stores models on the computing device cluster.

[0112] When an AI model is installed on a terminal device, the terminal device first reports the prediction capability information.

[0113] The prediction capability information indicates the capability of the AI ​​model to predict handover HO.

[0114] In one possible implementation, the prediction capability information can be carried in radio resource control (RRC) signaling sent from the terminal device to the network device. RRC signaling is signaling exchanged between the network device and the terminal device through the RRC layer, which can implement radio resource management, connection management, and measurement.

[0115] S52: The network device sends prediction configuration information to the terminal device based on the received prediction capability information.

[0116] After receiving the prediction capability information, the network device enables the AI ​​model to predict HO.

[0117] The network device sends the predicted configuration information to the terminal device. In one possible implementation, the trigger condition is generated by the network device based on the predicted HO capability of the terminal device. The predicted configuration information is used to indicate the trigger condition for triggering the HO.

[0118] The embodiments of the present application do not specifically limit the triggering conditions for triggering HO. For example, the triggering conditions may be: a prediction period configured for the AI ​​model of the terminal device, and / or a triggering event.

[0119] Different AI models have different prediction capabilities, and the corresponding prediction periods can be different. Different AI models have different prediction capabilities, and the corresponding trigger events can be different.

[0120] The trigger conditions configured by the network device for the terminal device can include both a trigger event and a prediction period. In this case, the terminal device's AI model predicts when a trigger event occurs, and predicts according to the prediction period when no trigger event occurs.

[0121] S53: When the trigger condition is met, the terminal device performs HO according to the prediction result of the AI ​​model.

[0122] When the trigger conditions are met, the terminal device uses the AI ​​model to obtain prediction results. The prediction results can indicate specific HO information, such as:

[0123] When the AI ​​model has the ability to predict the timing of HO, the HO information can include HO time information and target cell information;

[0124] When the AI ​​model has the ability to predict premature HO and determine the delayed HO time, the HO information may include the delayed HO time information;

[0125] When the AI ​​model has the ability to predict HO to the wrong cell, the HO information can include the identifier of the wrong cell;

[0126] When the AI ​​model has the ability to predict beam switching, the HO information may include beam index information and beam switching time information.

[0127] At this point, the terminal device can perform HO based on the prediction results of the AI ​​model.

[0128] In summary, using the solution provided by the embodiment of the present application, an AI model is used to predict handovers. For example, the AI ​​model can be used to determine the timing of handovers to avoid handovers being too late. For another example, the AI ​​model can be used to predict that the handover command issued by the current network device will cause handovers to be too early, and the AI ​​model can be used to determine the time to postpone handovers. For another example, the AI ​​model can be used to predict that the handover command issued by the current network device will cause handovers to the wrong cell, and thus handovers can be stopped. For another example, the AI ​​model can be used to predict the beam index of the beam after handover and the time of beam switching to ensure that beam switching is successful. Compared with existing solutions, this improves the success rate of handovers, ensures the quality of mobile networks, and enhances user experience.

[0129] The following describes the method of using AI model to assist switching with specific implementation methods.

[0130] The following first describes a method for using an AI model to predict switching timing.

[0131] See FIG. 7 , which is a flowchart of another switching method provided in an embodiment of the present application.

[0132] The method comprises the following steps:

[0133] S61: The terminal device sends prediction capability information.

[0134] An AI model is deployed on the terminal device, and the terminal device reports prediction capability information to the network device. The prediction capability information is used to indicate the ability of the AI ​​model to predict HO. In the embodiment of the present application, the AI ​​model has the ability to predict the timing of HO.

[0135] In one possible implementation, the prediction capability information may be carried in RRC signaling sent by the terminal device to the network device. That is, when the terminal device establishes an RRC connection with the network device, the prediction capability information is sent to the network device via RRC signaling.

[0136] The network device in the embodiment of the present application may be a base station of a source cell.

[0137] S62: The network device sends prediction configuration information to the terminal device based on the received prediction capability information.

[0138] The prediction configuration information is used to indicate a triggering condition for triggering the predictive HO.

[0139] In one possible implementation, the trigger condition may include making a prediction based on a prediction period and / or satisfying a trigger event. When the trigger condition includes both a trigger event and a prediction period, the AI ​​model of the terminal device makes a prediction when a trigger event occurs, and makes a prediction based on the prediction period when no trigger event occurs.

[0140] In one possible implementation, the triggering event includes a detection value being lower than a threshold of the serving cell and / or higher than a threshold of an adjacent cell. The detection value and the threshold may be reference signal receiving power (RSRP) and / or reference signal received quality (RSRQ). For example, the threshold of the serving cell is configured as a first RSRP threshold, and the triggering event is that the detection value of the RSRP of the serving cell is less than the first RSRP threshold; for another example, the threshold of the adjacent cell is configured as a second RSRP threshold, and the triggering condition is that the detection value of the RSRP of the adjacent cell is greater than the second RSRQ threshold; for another example, the threshold of the serving cell is configured as a first RSRQ threshold, and the triggering event is that the detection value of the RSRQ of the serving cell is less than the first RSRQ threshold; for another example, the threshold of the adjacent cell is configured as a second RSRQ threshold, and the triggering condition is that the detection value of the RSRQ of the adjacent cell is greater than the second RS RQ threshold; for another example, the threshold of the serving cell is configured as a third RSRP threshold, and the threshold of the neighboring cell is configured as a fourth RSRP threshold, and the trigger event is that the detected RSRP value of the serving cell is less than the third RSRP threshold, and the detected RSRP value of the neighboring cell is greater than the fourth RSRQ threshold; for another example, the threshold of the serving cell is configured as a third RSRQ threshold, and the threshold of the neighboring cell is configured as a fourth RSRQ threshold, and the trigger event is that the detected RSRQ value of the serving cell is less than the third RSRQ threshold, and the detected RSRQ value of the neighboring cell is greater than the fourth RSRQ threshold. In addition, other configuration methods can also be used, which are not detailed here.

[0141] In another possible implementation, the triggering time can be configured based on the number of HO failures or the HO failure rate within a certain period of time. For example, the triggering event includes one or more of the following:

[0142] The HO failure rate in the first time interval is greater than the first threshold, the number of HO failures in the second time interval is greater than the first value, the probability of premature HO in the third time interval is greater than the second threshold, the number of premature HOs in the fourth time interval is greater than the second value, the probability of late HO in the fifth time interval is greater than the third threshold, the number of late HOs in the sixth time interval is greater than the third value, the probability of HO to an incorrect cell in the seventh time interval is greater than the fourth threshold, and the number of HO to an incorrect cell in the eighth time interval is greater than the fourth value.

[0143] In another possible implementation, the terminal device may also use the HO command received from the network device as a trigger event. That is, after receiving the HO command from the network device, the terminal device uses the AI ​​model to obtain the HO prediction result.

[0144] Prediction configuration information can be carried in Media Access Control Element (MAC CE) signaling. MAC CE is an alternative method for exchanging control information between user equipment and network devices, in addition to RRC signaling. It is signaling between network devices and user equipment over the MAC layer. It enables uplink synchronization adjustments, activation, and deactivation.

[0145] The prediction configuration information may also be carried in RRC signaling.

[0146] S63: When the terminal device meets the triggering conditions, it uses the AI ​​model to obtain the HO prediction result.

[0147] When the AI ​​model of the terminal device meets the trigger conditions, that is, when a trigger event occurs or a periodic prediction is made, the AI ​​model is used to obtain the prediction results.

[0148] The prediction result includes at least switching time information and target cell information.

[0149] In one possible implementation, the handover time information indicates the handover time corresponding to handover to at least one candidate cell. A candidate cell is a mobile communication system in which the signal of a channel within the serving cell gradually weakens before handover, while the signal of a channel within certain adjacent cells strengthens to a certain threshold to maintain normal communication. These adjacent cells are referred to as candidate cells. In other words, the handover time information indicates the handover time corresponding to all selectable candidate cells, which are then further selected by the network device. The handover time information may include a correspondence between the cell ID of at least one candidate cell and the handover time.

[0150] In another possible implementation, the switching time information indicates the switching time corresponding to switching to one or more recommended cells. The recommended cells may be one or more, and the embodiment of the present application does not specifically limit the number of recommended cells. In a preferred implementation, the number of recommended cells is smaller than the number of candidate cells, thereby achieving convergence in the number of cells.

[0151] The target cell information indicates at least one candidate cell, or indicates one or more recommended cells.

[0152] The switching time in the embodiment of the present application can be an absolute time or a relative time, which is not specifically limited in the embodiment of the present application. The time involved in the following embodiments can be an absolute time or a relative time, which will not be described in detail.

[0153] The absolute time may be global navigation satellite system (GNSS) time or coordinated universal time (UTC).

[0154] GNSS can be the global positioning system (GPS), the global navigation satellite system (GLONASS), the Beidou navigation satellite system (BDS), the quasi-zenith satellite system (QZSS) or the satellite based augmentation system (SBAS).

[0155] The relative time may be a system frame number (SFN), a timeslot number, an orthogonal frequency division multiplexing (OFDM) symbol number, or an OFDM symbol offset value, etc.

[0156] S64: The terminal device sends HO prediction information.

[0157] In one possible implementation, the terminal device generates HO prediction information according to the HO prediction result of the AI ​​model.

[0158] The HO prediction information includes at least switching time information and target cell information, that is, the HO prediction information is used to indicate the switching time corresponding to switching to at least one candidate cell, or the switching time corresponding to switching to one or more recommended cells, and indicates at least one candidate cell, or indicates one or more recommended cells.

[0159] In a possible implementation, the HO prediction information may be carried in RRC signaling.

[0160] S65: The network device sends a first HO command according to the HO prediction information.

[0161] After obtaining the HO prediction information, the network device parses it to obtain the handover time information and target cell information, and uses the handover time information and target cell information to assist in decision-making. For example, if the handover time information indicates a first moment T1 for the handover time, and the network device determines a second moment T2 for the handover time, and the first moment T1 is earlier than the second moment T2, the network device will determine the first moment T1 as the handover time to avoid handover failure due to a late handover. For another example, in another possible implementation, the handover time information indicates a first moment T1 for the handover time, and the network device determines a second moment T2 for the handover time, and the first moment T1 is earlier than the second moment T2, but the time difference between T1 and T2 is less than a preset threshold, indicating that the handover time determined by the network device has a high degree of credibility. In this case, the network device may still determine the second moment T2 as the handover time. If the time difference between T1 and T2 is greater than the preset threshold, the network device may determine the first moment T1 as the handover time.

[0162] In summary, the method provided by the embodiments of this application uses an AI model to determine the handover timing to avoid handover failures caused by handover being too late. Compared with existing solutions, this improves the handover success rate, ensures the quality of the mobile network, and enhances the user experience.

[0163] The following describes a method for predicting premature switching using an AI model.

[0164] See FIG8 , which is a flowchart of another switching method provided in an embodiment of the present application.

[0165] S71: The network device sends a second HO command to the terminal device.

[0166] The network device in the embodiment of the present application may be a base station of a source cell.

[0167] The base station device and the terminal device establish an RRC connection. Before S71, the terminal device sends prediction capability information to the network device, and the network device sends prediction configuration information to the terminal device. For specific steps, please refer to the description in S61-S62 above, and this embodiment will not be repeated here.

[0168] After determining that switching is required, the network device sends a second HO command to the terminal device, where the second HO command is at least used to indicate a switching time.

[0169] In a possible implementation, the second HO command may be specifically used to indicate a target cell for handover. The target cell may be identified by a cell ID.

[0170] S72: The terminal obtains a HO prediction result using the AI ​​model.

[0171] The triggering condition may be prediction according to the prediction period and / or satisfying a triggering event. Detailed description of the triggering condition can be found in the above embodiments and will not be repeated here.

[0172] After the terminal device receives the second HO command sent by the network device, it triggers the AI ​​model to perform HO prediction, or when the periodic prediction is just completed or a triggering event occurs, the AI ​​model is used to obtain the HO prediction result.

[0173] The HO prediction result includes switching time information, or includes switching postponement time information.

[0174] S73: The terminal device determines that a premature HO occurs according to the switching time in the HO prediction result and the timing of receiving the second HO command.

[0175] The handover time in the HO prediction result is the third time T3, and the timing of receiving the second HO command is the fourth time T4. When the terminal device determines that T4 is earlier than T3, it determines that a premature HO has occurred. If the terminal device still performs handover according to the fourth time T4, the handover may fail.

[0176] S74: The terminal device sends a first indication message to the network device.

[0177] In one possible implementation, the first indication information is used to indicate that a premature HO has occurred and to request that an RRC connection with the source cell be maintained. The first indication information may be carried in RRC signaling. In this case, maintaining an RRC connection with the source cell is also maintaining an RRC connection with the network device.

[0178] In another possible implementation, the terminal device may send first indication information indicating that a premature HO has occurred to the network device, and also send request information to the network device requesting to maintain the RRC connection with the source cell. That is, the terminal device sends two pieces of information to the network device.

[0179] Optionally, the first indication information is further used to indicate time information for postponing HO, so that the network device can adjust the handover configuration parameters according to the time information for postponing HO after receiving the first indication information.

[0180] Specifically, the deferred HO time information may indicate a deferred switching time corresponding to switching to at least one candidate cell, that is, the switching time information indicates the switching times corresponding to all optional candidate cells. The deferred HO time information may include a correspondence between the cell ID of at least one candidate cell and the deferred switching time.

[0181] The time information for postponing HO may also indicate the postponed switching time corresponding to the target cell indicated by the second HO command. When the AI ​​model of the terminal device does not have the ability to predict HO to the wrong cell, the postponed switching time is only determined specifically for the target cell indicated by the second HO command. When the AI ​​model of the terminal device has the ability to predict HO to the wrong cell, it determines the postponed switching time corresponding to switching to the target cell indicated by the second HO command, and also indicates that the AI ​​model determines that the target cell is the correct cell at this time, that is, after switching to the target cell, HOF will not be caused because the target cell is the wrong cell.

[0182] The HO deferral time information may also indicate the deferred handover time corresponding to handover to the recommended target cell. The recommended cells may be one or more, and the present embodiment does not specifically limit the number of recommended cells. In a preferred implementation, the number of recommended cells is smaller than the number of candidate cells, thereby achieving convergence in the number of cells.

[0183] In a possible implementation, the first indication information is further used to indicate a target cell for HO.

[0184] At this time, the terminal device will not directly switch according to the switching time indicated by the second HO command, but will determine the switching time based on the HO prediction result obtained by the AI ​​model to achieve delayed switching.

[0185] S75: The network device adjusts the switching configuration parameters.

[0186] The network device receives first indication information, where the first indication information is used to indicate that a premature HO occurs and to request to maintain a radio resource control (RRC) connection with a source cell. The network device can then adjust a handover configuration parameter.

[0187] In summary, the method provided by the embodiments of this application uses an AI model to determine the handover time or postpone the handover time to avoid handover failures caused by premature handover. Compared with existing solutions, this method improves the handover success rate, ensures the quality of the mobile network, and enhances the user experience.

[0188] The following describes a method for using an AI model to predict switching to an incorrect cell.

[0189] See Figure 9, which is a flowchart of another switching method provided in an embodiment of the present application.

[0190] The method comprises the following steps:

[0191] S81: The S-gNB sends a third HO command.

[0192] The S-gNB in ​​the embodiment of the present application is a network device of the source cell, that is, a base station of the source cell.

[0193] The S-gNB and the terminal device establish an RRC connection. The terminal device sends prediction capability information to the S-gNB, and the S-gNB sends prediction configuration information to the terminal device. For specific steps, please refer to the description in S61-S62 above. This embodiment will not be repeated here.

[0194] After the S-gNB determines that cell switching is required, it sends a third HO command to the terminal device.

[0195] The third HO command is used to at least indicate the target cell of the HO, that is, the T-gNB. Specifically, the third HO command may carry the cell ID of the target cell.

[0196] S82: The terminal device obtains a HO prediction result using the AI ​​model.

[0197] After receiving the third HO command, the terminal device triggers HO prediction and obtains the HO prediction result by using the AI ​​model, or when the HO prediction is triggered by a trigger condition, obtains the HO prediction result by using the AI ​​model.

[0198] After the terminal device receives the third HO command sent by the network device, it triggers the AI ​​model to perform HO prediction, or when the periodic prediction is just completed or a triggering event occurs, the AI ​​model is used to obtain the HO prediction result.

[0199] For detailed description of the triggering conditions, please refer to the above embodiments and will not be repeated here.

[0200] The HO prediction result may be used to indicate an erroneous cell. For example, the HO prediction result may include a cell ID of the erroneous cell.

[0201] S83: When the terminal device determines that the target cell is an incorrect cell based on the HO prediction result, the terminal device does not access the target cell.

[0202] After receiving the third HO command, the terminal device parses and obtains the cell ID of the target cell. When the cell ID of the target cell matches the cell ID of the error cell indicated by the prediction result, it is determined that the target cell is an error cell.

[0203] At this point, the terminal device does not access the target cell. Specifically, the terminal device stops executing the handover process, that is, stops initiating random access to the target cell indicated in the third HO command; or the terminal device does not initiate the handover process, that is, the terminal device can directly ignore the third HO command to avoid handover failure due to handover to the wrong cell.

[0204] S84: Perform cell selection and complete RRC connection with the third gNB.

[0205] Cell selection, also known as cell reselection, is the process by which a terminal device selects the best cell for service by monitoring the signal strength or quality of neighboring cells and the current cell. Cell reselection maximizes the chance that the terminal device will reside in the appropriate cell.

[0206] After the terminal performs cell reselection, the selected cell is the third cell, that is, the third gNB. The terminal initiates the RRC re-establishment process.

[0207] The terminal device carries the identifier of the source cell S-gNB and the identifier of the target cell T-gNB in ​​the RRC re-establishment request message or the RRC connection establishment complete message, that is, it carries the cell ID of the source cell S-gNB and the cell ID of the target cell T-gNB.

[0208] S85: The third gNB sends an RLF indication to the T-gNB.

[0209] After the third gNB completes the re-establishment with the terminal device, the source cell before the re-establishment of the terminal device and the target cell indicated by the third HO command are determined based on the obtained identifier of the source cell S-gNB and the identifier of the target cell T-gNB.

[0210] In an ad hoc network, failure reporting can be used to optimize handover parameter configuration. After the third gNB and the terminal device are re-established, the third gNB sends an RLF indication to the T-gNB to indicate that an RLF has occurred between the terminal device and the T-gNB. It is understood that although an actual RLF does not occur between the terminal device and the T-gNB in ​​the solution of the embodiment of the present application, this is because the AI ​​model's predictive function avoids this. That is, if the terminal device switches to the T-gNB in ​​accordance with the third HO command, an RLF will occur in the terminal device. Therefore, in the embodiment of the present application, it is considered that an RLF has occurred between the terminal device and the T-gNB.

[0211] Another function of the RLF indication is to enable the target cell T-gNB to send a HO report to the source cell S-gNB after receiving the RLF indication.

[0212] The HO report is used to report the type of HOF. In this case, the reported HOF type is a handover to an incorrect cell.

[0213] S86: The T-gNB sends a HO report to the S-gNB.

[0214] The S-gNB modifies and optimizes the handover configuration parameters based on the HO report.

[0215] In summary, the method provided in the embodiment of the present application uses an AI model to determine whether the target cell indicated in the third HO command is an incorrect cell. If it is not an incorrect cell, the terminal device can switch to the target cell according to the third HO command. If it is an incorrect cell, the terminal device will not establish an RRC connection with the target cell, but will perform cell reselection, thereby avoiding handover failure caused by the terminal device switching to the incorrect cell. Compared with existing solutions, this improves the success rate of handovers, ensures the quality of the mobile network, and enhances the user experience.

[0216] The following describes a method for using an AI model to predict switching beams.

[0217] See Figure 10, which is a flowchart of another switching method provided in an embodiment of the present application.

[0218] The method comprises the following steps:

[0219] S91: The terminal device sends prediction capability information.

[0220] The prediction capability information indicates the AI ​​model's ability to predict handover (HO). The AI ​​model in the embodiments of the present application has beam-level capabilities, that is, it can predict information such as the beam index of the handover beam and the timing of beam switching. The beam after switching can be a beam in the same cell or a beam in a different cell.

[0221] In a possible implementation, the prediction capability information may be carried in the RRC signaling sent by the terminal device to the network device.

[0222] S92: The network device sends prediction configuration information to the terminal device based on the received prediction capability information.

[0223] After receiving the prediction capability information, the network device enables the AI ​​model to predict HO.

[0224] The network device sends prediction configuration information to the terminal device, where the prediction configuration information is used to indicate a triggering condition for triggering the predictive HO. In a possible implementation, the triggering condition is generated by the network device according to the predictive HO capability of the terminal device.

[0225] The trigger conditions include making predictions according to a prediction period and / or satisfying a trigger event.

[0226] The trigger conditions configured by the network device for the terminal device can include both a trigger event and a prediction period. In this case, the terminal device's AI model predicts when a trigger event occurs, and predicts according to the prediction period when no trigger event occurs.

[0227] In one possible implementation, the triggering event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell. The detection value and threshold may be reference signal receiving power (RSRP) and / or reference signal received quality (RSRQ).

[0228] For example, the threshold of the serving cell is configured as the first RSRP threshold, and the trigger event is that the detection value of the RSRP of the serving cell is less than the first RSRP threshold; for another example, the threshold of the adjacent cell is configured as the second RSRP threshold, and the trigger condition is that the detection value of the RSRP of the adjacent cell is greater than the second RSRQ threshold; for another example, the threshold of the serving cell is configured as the first RSRQ threshold, and the trigger event is that the detection value of the RSRQ of the serving cell is less than the first RSRQ threshold; for another example, the threshold of the adjacent cell is configured as the second RSRQ threshold, and the trigger condition is that the detection value of the RSRQ of the adjacent cell is greater than the second RS RQ threshold; for another example, the threshold of the serving cell is configured as a third RSRP threshold, and the threshold of the neighboring cell is configured as a fourth RSRP threshold, and the trigger event is that the detected RSRP value of the serving cell is less than the third RSRP threshold, and the detected RSRP value of the neighboring cell is greater than the fourth RSRQ threshold; for another example, the threshold of the serving cell is configured as a third RSRQ threshold, and the threshold of the neighboring cell is configured as a fourth RSRQ threshold, and the trigger event is that the detected RSRQ value of the serving cell is less than the third RSRQ threshold, and the detected RSRQ value of the neighboring cell is greater than the fourth RSRQ threshold. In addition, other configuration methods can also be used, which are not detailed here.

[0229] In another possible implementation, the triggering event may include a detection value being lower than an average beam quality value of a serving cell and / or higher than an average beam quality value of a neighboring cell, wherein the detection value and the average quality value are RSRP values ​​and / or RSRQ values.

[0230] For example, the average beam quality value of the serving cell is configured to be a first RSRP value, and the triggering event is that the average RSRP value of each beam of the serving cell is less than the first RSRP value;

[0231] For another example, the average beam quality of the neighboring cell is configured as a second RSRP value, and the trigger condition is that the average RSRP value of each beam of the neighboring cell is greater than the second RSRQ value;

[0232] For another example, the average beam quality value of the serving cell is configured to be a first RSRQ value, and the triggering event is that the average RSRQ value of each beam of the serving cell is less than the first RSRQ value;

[0233] For another example, the average beam quality of the adjacent cell is configured as the second RSRQ value, and the trigger condition is that the average RSRQ of each beam of the adjacent cell is greater than the second RSRQ threshold;

[0234] For another example, the average beam quality of the serving cell is configured as a third RSRP value, and the average beam quality of the neighboring cell is configured as a fourth RSRP value. The triggering event is that the average RSRP value of each beam of the serving cell is less than the third RSRP threshold, and the average RSRP value of each beam of the neighboring cell is greater than the fourth RSRQ value.

[0235] For another example, the average beam quality value of the serving cell is configured as a third RSRQ value, and the average beam quality value of the adjacent cell is configured as a fourth RSRQ value. The triggering event is that the average RSRQ value of each beam of the serving cell is less than the third RSRQ threshold, and the average RSRQ value of each beam of the adjacent cell is greater than the fourth RSRQ threshold.

[0236] In addition, other configurations may also be used, which will not be described in detail here.

[0237] In another possible implementation, the triggering event includes a detection value being lower than the quality average of the N optimal beams of the serving cell and / or higher than the quality average of the M optimal beams of the adjacent cell, where N and M are positive integers, and the embodiment of the present application does not specifically limit the size relationship between M and N. The detection value and the quality average are RSRP values ​​and / or RSRQ values. The N optimal beams refer to the N beams with the highest quality after all beams of the serving cell are sorted in order of quality from high to low. The M optimal beams refer to the M beams with the highest quality after all beams of the adjacent cells are sorted in order of quality from high to low.

[0238] For example, the average quality value of the N best beams of the serving cell is configured to be a first RSRP value, and the triggering event is that the average RSRP value of the current N best beams of the serving cell is less than the first RSRP value;

[0239] For another example, the average quality value of the M best beams of the neighboring cell is configured to be the second RSRP value, and the trigger condition is that the average RSRP value of the current M best beams of the neighboring cell is greater than the second RSRQ value;

[0240] For another example, the quality average of the N best beams of the serving cell is configured to be a first RSRQ value, and the triggering event is that the RSRQ average of the current N best beams of the serving cell is less than the first RSRQ value;

[0241] For another example, the average quality of M optimal beams of the neighboring cell is configured as the second RSRQ value, and the trigger condition is that the average RSRQ of the current M optimal beams of the neighboring cell is greater than the second RSRQ threshold;

[0242] For another example, the average quality of the N best beams of the serving cell is configured as a third RSRP value, and the average quality of the M best beams of the neighboring cell is configured as a fourth RSRP value. The triggering event is that the average RSRP value of the current N best beams of the serving cell is less than the third RSRP threshold, and the average RSRP value of the current M best beams of the neighboring cell is greater than the fourth RSRQ value.

[0243] For another example, the average quality value of the N optimal beams of the serving cell is configured as a third RSRQ value, and the average quality value of the M optimal beams of the adjacent cell is configured as a fourth RSRQ value. The triggering event is that the average RSRQ value of the current N optimal beams of the serving cell is less than the third RSRQ threshold, and the average RSRQ value of the current M optimal beams of the adjacent cell is greater than the fourth RSRQ threshold.

[0244] In addition, other configurations may also be used, which will not be described in detail here.

[0245] In another possible implementation, the triggering time can be configured based on the number of HO failures or the HO failure rate within a certain period of time. For example, the triggering event includes one or more of the following:

[0246] The HO failure rate in the first time interval is greater than the first threshold, the number of HO failures in the second time interval is greater than the first value, the probability of premature HO in the third time interval is greater than the second threshold, the number of premature HOs in the fourth time interval is greater than the second value, the probability of late HO in the fifth time interval is greater than the third threshold, the number of late HOs in the sixth time interval is greater than the third value, the probability of HO to an incorrect cell in the seventh time interval is greater than the fourth threshold, and the number of HO to an incorrect cell in the eighth time interval is greater than the fourth value.

[0247] In another possible implementation, the terminal device may trigger the AI ​​model to perform prediction when receiving the HO command issued by the network device.

[0248] S93: When the trigger condition is met, the AI ​​model is used to obtain the beam switching prediction result.

[0249] When the AI ​​model of the terminal device meets the trigger condition, that is, when a trigger event occurs or a periodic prediction is performed, the AI ​​model is used to obtain the prediction result. The prediction result may include beam index information and beam switching time information.

[0250] To identify candidate beams, the system assigns an index to each candidate beam, called a beam index. There is a one-to-one correspondence between the beam index and the candidate beam. The beam index is used to identify the beam after switching.

[0251] S94: The terminal device sends beam prediction information to the network device.

[0252] The beam prediction information includes beam index information and beam switching time information.

[0253] In one possible implementation, the beam switching time information indicates the switching time to at least one beam corresponding to at least one candidate cell. That is, the beam switching time information indicates the switching time of the beams corresponding to all optional candidate cells, which are further selected by the network device. The beam switching time information may include the switching time of one or more beams corresponding to each cell ID. For example, the current candidate cells are the first cell and the second cell, the first cell includes the first beam and the second beam, and the second cell includes the third beam and the fourth beam. The beam switching time information may include the switching time of the first beam of the first cell, the switching time of the second beam of the first cell, the switching time of the third beam of the second cell, and the switching time of the fourth beam of the second cell.

[0254] In another possible implementation, the beam switching time information indicates the switching time of the recommended beam corresponding to at least one candidate cell, thereby achieving convergence in the number of beams. For example, if the current candidate cells are a first cell and a second cell, the first cell includes a first beam and a second beam, and the second cell includes a third beam and a fourth beam, the first beam of the first cell is the recommended beam, and the third beam of the second cell is the recommended beam, then the beam switching time information may include the switching time of the first beam of the first cell and the switching time of the third beam of the second cell.

[0255] S95: The network device sends a fourth HO command according to the beam prediction information.

[0256] After receiving the beam prediction information, the network device parses it to obtain beam index information and beam switching time information as auxiliary information for decision-making. For example, if the network device determines the beam switching time to be T1, and the beam after switching is the first beam of the first cell, the network device receives beam prediction information including the switching time T2 for the first beam of the first cell and the switching time T3 for the third beam of the second cell. The network device can then send a fourth HO command, which instructs the terminal device to switch to the first beam of the first cell, with the switching time being T2.

[0257] In summary, the method provided in the embodiments of this application uses an AI model to determine the beam index after beam switching and the timing of beam switching, thereby avoiding beam switching failures caused by incorrect beam switching, premature beam switching, or late beam switching. Compared with existing solutions, this improves the success rate of beam switching, ensures the quality of the mobile network, and enhances the user experience.

[0258] Based on the switching method provided in the above embodiment, the embodiment of the present application further provides a switching device, which is described in detail below with reference to the accompanying drawings.

[0259] See Figure 11, which is a schematic diagram of a switching device provided in an embodiment of the present application.

[0260] The switching device can be applied to a terminal device, and includes: a first sending unit 11 , a first receiving unit 12 , and a prediction unit 13 .

[0261] The first sending unit 11 is configured to send prediction capability information indicating the capability of the artificial intelligence (AI) model to predict handover (HO).

[0262] The first receiving unit 12 is configured to receive prediction configuration information. The prediction configuration information is used to indicate a triggering condition for triggering HO.

[0263] The prediction unit 13 is configured to perform HO according to the prediction result of the AI ​​model when a trigger condition is met.

[0264] In a possible implementation, the triggering condition includes making a prediction according to a prediction period and / or satisfying a triggering event.

[0265] In a possible implementation, the triggering event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell, and the detection value is a detection value of a reference signal received power RSRP and / or a reference signal received quality RSRQ.

[0266] In one possible implementation, the triggering event includes one or more of the following:

[0267] The HO failure rate in the first time interval is greater than the first threshold, the number of HO failures in the second time interval is greater than the first value, the probability of premature HO in the third time interval is greater than the second threshold, the number of premature HOs in the fourth time interval is greater than the second value, the probability of late HO in the fifth time interval is greater than the third threshold, the number of late HOs in the sixth time interval is greater than the third value, the probability of HO to an incorrect cell in the seventh time interval is greater than the fourth threshold, and the number of HO to an incorrect cell in the eighth time interval is greater than the fourth value.

[0268] In a possible implementation, the prediction unit 13 is configured to obtain a HO prediction result using the AI ​​model when a trigger condition is met; and send HO prediction information, where the HO prediction information includes at least switching time information and target cell information.

[0269] In one possible implementation, the switching time information indicates the switching time corresponding to switching to at least one candidate cell, or the switching time corresponding to switching to one or more recommended cells; the target cell information indicates at least one candidate cell, or one or more recommended cells.

[0270] In one possible implementation, the prediction unit 13 is configured to receive the second HO command; trigger HO prediction and obtain an HO prediction result using the AI ​​model, or, when the HO prediction is triggered by a trigger condition, obtain the HO prediction result using the AI ​​model; and, when it is determined that a premature HO has occurred based on the switching time in the HO prediction result and the timing of receiving the second HO command, send first indication information, where the first indication information is used to indicate the occurrence of a premature HO and to request to maintain a radio resource control (RRC) connection with the source cell.

[0271] In one possible implementation, the first indication information is further used to indicate the target cell of HO; the first indication information is further used to indicate time information for postponing HO, the time information for postponing HO indicates a deferred switching time corresponding to switching to at least one candidate cell, or a deferred switching time corresponding to switching to the target cell indicated by the second HO command, or a deferred switching time corresponding to switching to the recommended target cell, and the at least one candidate cell is configured by the network device.

[0272] In a possible implementation, the prediction unit 13 is used to receive a third HO command, where the third HO command is at least used to indicate a target cell for HO; trigger HO prediction, and obtain a HO prediction result using an AI model, or when HO prediction is triggered by a trigger condition, obtain a HO prediction result using the AI ​​model; when the target cell is determined to be an error cell according to the HO prediction result, not accessing the target cell; performing cell selection and completing a radio resource control (RRC) connection with the third cell, and carrying an identifier of the source cell and an identifier of the target cell in an RRC connection request message or an RRC connection establishment completion message sent to the third cell, so that the third cell sends a radio link failure (RLF) indication to the target cell, where the RLF indication is used to enable the target cell to send a HO report to the source cell, and the HO report indicates that the type of the handover failure (HOF) is HO to an error cell.

[0273] In one possible implementation, the AI ​​model's ability to predict HO includes the ability to predict switching beams, and triggering events include: a detection value is lower than the threshold of the serving cell and / or higher than the threshold of the neighboring cell; or, the detection value is lower than the average beam quality of the serving cell and / or higher than the average beam quality of the neighboring cell; or, the detection value is lower than the average quality of the N best beams of the serving cell and / or higher than the average quality of the M best beams of the neighboring cell, where N and M are positive integers; the detection value is the detection value of the reference signal received power RSRP and / or the reference signal received quality RSRQ.

[0274] In one possible implementation, the prediction unit 13 is used to obtain a beam switching prediction result using an AI model when a trigger condition is met; and send beam prediction information, which includes beam index information and beam switching time information.

[0275] In a possible implementation, the beam switching time information indicates a switching time to at least one beam corresponding to at least one candidate cell, or a switching time to a recommended beam corresponding to at least one candidate cell.

[0276] Based on the switching method provided in the above embodiment, the embodiment of the present application further provides a terminal device, which is described in detail below with reference to the accompanying drawings.

[0277] See FIG. 12 , which is a schematic diagram of another switching device provided in an embodiment of the present application.

[0278] The switching device includes a second receiving unit 21 and a configuration unit 22 .

[0279] The second receiving unit 21 is configured to receive prediction capability information indicating the capability of the artificial intelligence (AI) model to predict handover (HO).

[0280] The configuration unit 22 is configured to send prediction configuration information according to the prediction capability information.

[0281] The prediction configuration information is used to indicate a triggering condition for triggering the predictive HO.

[0282] In a possible implementation, the triggering condition includes making a prediction according to a prediction period and / or satisfying a triggering event.

[0283] In a possible implementation, the triggering event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell, and the detection value is a detection value of a reference signal received power RSRP and / or a reference signal received quality RSRQ.

[0284] In a possible implementation, the triggering event includes one or more of the following: a HO failure rate in a first time interval is greater than a first threshold, the number of HO failures in a second time interval is greater than a first value, a probability of premature HO in a third time interval is greater than a second threshold, the number of premature HOs in a fourth time interval is greater than a second value, a probability of late HO in a fifth time interval is greater than a third threshold, the number of late HOs in a sixth time interval is greater than a third value, a probability of HO to an incorrect cell in a seventh time interval is greater than a fourth threshold, and the number of HOs to an incorrect cell in an eighth time interval is greater than a fourth value.

[0285] In one possible implementation, the second receiving unit 21 is further configured to receive HO prediction information, where the HO prediction information includes at least handover time information and target cell information. The apparatus further includes a second sending unit configured to send a first HO command based on the HO prediction information.

[0286] In a possible implementation, the switching time information indicates a switching time corresponding to switching to at least one candidate cell, or a switching time corresponding to switching to one or more recommended cells.

[0287] The target cell information indicates at least one candidate cell, or one or more recommended cells.

[0288] In a possible implementation, the second sending unit is further configured to send a second HO command. The second receiving unit 21 is further configured to receive first indication information. The first indication information is used to indicate that a premature HO has occurred and to request to maintain a radio resource control (RRC) connection with the source cell.

[0289] In one possible implementation, the first indication information is further used to indicate a target cell for HO. The first indication information is further used to indicate time information for postponing HO, where the time information for postponing HO indicates a postponed switching time corresponding to switching to at least one candidate cell, or a postponed switching time corresponding to switching to a target cell indicated by the second HO command, or a postponed switching time corresponding to switching to a recommended target cell, where the at least one candidate cell is configured by the network device.

[0290] In one possible implementation, the second sending unit is further configured to send a third HO command. The third HO command is configured to at least indicate a target cell for HO. The second receiving unit 21 is further configured to receive an HO report sent by the target cell, where the HO report indicates that the type of the handover failure (HOF) is HO to an incorrect cell, and the HO report is sent by the target cell after receiving a radio link failure (RLF) sent by a third cell, where the third cell is the cell for which the terminal device performs cell selection and completes a radio resource control (RRC) connection.

[0291] In one possible implementation, the AI ​​model's ability to predict HO includes the ability to predict switching beams, and triggering events include: a detection value is lower than the threshold of the serving cell and / or higher than the threshold of the neighboring cell; or, the detection value is lower than the average beam quality of the serving cell and / or higher than the average beam quality of the neighboring cell; or, the detection value is lower than the average quality of the N best beams of the serving cell and / or higher than the average quality of the M best beams of the neighboring cell, where N and M are positive integers; the detection value is the detection value of the reference signal received power RSRP and / or the reference signal received quality RSRQ.

[0292] In a possible implementation, the second receiving unit 21 is further configured to receive beam prediction information, where the beam prediction information includes beam index information and beam switching time information. The second sending unit is further configured to send a fourth HO command.

[0293] In a possible implementation, the beam switching time information indicates a switching time to at least one beam corresponding to at least one candidate cell, or a switching time to a recommended beam corresponding to at least one candidate cell.

[0294] The embodiment of the present application also provides a network device, which is described in detail below with reference to the accompanying drawings.

[0295] See Figure 13, which is a schematic diagram of another switching device provided in an embodiment of the present application.

[0296] The electronic equipment network equipment includes but is not limited to network equipment such as a base station, a core network unit, etc. The network equipment is described as a base station as an example.

[0297] The base station includes part 1110 , part 1120 and part 1130 .

[0298] Part 1110 is mainly used for baseband processing, base station control, etc.; Part 1110 is usually the control center of the base station, which can usually be called a processor, and is used to control the base station to execute the switching method of the network device in the above method embodiment.

[0299] Section 1120 is primarily used to store computer program code and data. Section 1130 is primarily used for transmitting and receiving radio frequency signals and converting radio frequency signals into baseband signals. Section 1130 is often referred to as a transceiver module, transceiver, transceiver circuit, or transceiver.

[0300] The transceiver module in section 1130, also known as a transceiver or transceiver, includes an antenna 1133 and a radio frequency circuit (not shown), which primarily performs radio frequency processing. Alternatively, the device used for receiving in section 1130 can be considered a receiver, and the device used for transmitting can be considered a transmitter. Specifically, section 1130 includes a receiver 1132 and a transmitter 1131. A receiver can also be referred to as a receiving module, receiver, or receiving circuit, and a transmitter can be referred to as a transmitting module, transmitter, or transmitting circuit.

[0301] Sections 1110 and 1120 may include one or more boards, each of which may include one or more processors and one or more memories. The processor is used to read and execute programs in the memory to implement baseband processing functions and control the base station. If multiple boards are present, the boards may be interconnected to enhance processing capabilities. As an optional implementation, multiple boards may share one or more processors, multiple boards may share one or more memories, or multiple boards may simultaneously share one or more processors.

[0302] For example, in one implementation, the transceiver module in section 1130 is used to execute the process of sending and receiving related information executed by the network device in the aforementioned method embodiment. The processor in section 1110 is used to execute the process of processing related information executed by the network device in the aforementioned method embodiment.

[0303] It should be understood that FIG13 is merely an example and not a limitation, and the network device including the processor, memory, and transceiver may not rely on the structure shown in FIG13 .

[0304] See FIG. 14 , which is a schematic diagram of another electronic device provided in an embodiment of the present application.

[0305] The electronic device can be a terminal device. Taking the terminal device as a mobile phone as an example, the terminal device may include a processor 310, an external memory interface 320, an internal memory 321, a display screen 330, a camera 340, an antenna 1, an antenna 2, a mobile communication module 350, and a wireless communication module 360, etc.

[0306] It should be understood that the structure illustrated in this embodiment does not constitute a specific limitation on the electronic device. In other embodiments, the electronic device may include more or fewer components than shown, or some components may be combined or separated, or the components may be arranged differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0307] The processor 310 may include one or more processing units. For example, the processor 310 may include an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU). The different processing units may be independent devices or integrated into one or more processors.

[0308] It is understood that the interface connection relationship between the modules illustrated in this embodiment is only a schematic illustration and does not constitute a structural limitation of the electronic device. In other embodiments of the present application, the electronic device may also adopt different interface connection methods in the above embodiments, or a combination of multiple interface connection methods.

[0309] External memory interface 320 can be used to connect an external memory card, such as a Micro SD card, to expand the storage capacity of the electronic device. The external memory card communicates with processor 310 via external memory interface 320 to implement data storage functions. For example, files such as music and videos can be stored on the external memory card.

[0310] The internal memory 321 can be used to store computer executable program code, and the executable program code includes instructions. The processor 310 executes various functional applications and data processing of the electronic device by running the instructions stored in the internal memory 321. The internal memory 321 may include a program storage area and a data storage area. Among them, the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc. The data storage area can store data created during the use of the electronic device (such as audio data, a phone book, etc.), etc. In addition, the internal memory 321 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc. The processor 310 executes various functional applications and data processing of the electronic device by running the instructions stored in the internal memory 321, and / or the instructions stored in the memory provided in the processor.

[0311] The wireless communication function of the electronic device can be implemented through antenna 1, antenna 2, mobile communication module 350, wireless communication module 360, modem processor and baseband processor.

[0312] Antenna 1 and Antenna 2 are used to transmit and receive electromagnetic wave signals. Each antenna in an electronic device can be used to cover a single or multiple communication frequency bands. Different antennas can also be reused to improve antenna utilization. For example, antenna 1 can be reused as a diversity antenna for a wireless local area network. In other embodiments, the antennas can be used in conjunction with a tuning switch.

[0313] The mobile communication module 350 can provide solutions for wireless communications including 2G / 3G / 4G / 5G applied to electronic devices. The mobile communication module 350 may include at least one filter, a switch, a power amplifier, a low noise amplifier (LNA), etc. The mobile communication module 350 can receive electromagnetic waves from the antenna 1, and filter, amplify, and process the received electromagnetic waves, and transmit them to the modulation and demodulation processor for demodulation. The mobile communication module 350 can also amplify the signal modulated by the modulation and demodulation processor, and convert it into electromagnetic waves for radiation through the antenna 1. In some embodiments, at least some of the functional modules of the mobile communication module 350 can be set in the processor 310. In some embodiments, at least some of the functional modules of the mobile communication module 350 can be set in the same device as at least some of the modules of the processor 310.

[0314] In some embodiments, the electronic device initiates or receives a call request via the mobile communication module 350 and the antenna 1 .

[0315] The present application also provides a communication system, which may include a network device as shown in FIG13 and a terminal device as shown in FIG14 .

[0316] In this application, a terminal device or network device may include a hardware layer, an operating system layer running on the hardware layer, and an application layer running on the operating system layer. The hardware layer may include hardware such as a central processing unit (CPU), a memory management unit (MMU), and memory (also known as main memory). The operating system of the operating system layer may be any one or more computer operating systems that implement business processing through processes, such as the Linux operating system, Unix operating system, Android operating system, iOS operating system, or Windows operating system. The application layer may include applications such as browsers, address books, word processing software, and instant messaging software.

[0317] An embodiment of the present application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that can be stored by a computing device or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc. The computer-readable storage medium includes instructions that instruct the electronic device to perform the above-mentioned switching method. An embodiment of the present application also provides another computer-readable storage medium. The computer-readable storage medium includes instructions that instruct the electronic device to perform the above-mentioned switching method.

[0318] The embodiment of the present application further provides a computer program product comprising instructions. The computer program product may be software or a program product comprising instructions that can be run on an electronic device or stored in any available medium. When the computer program product is run on at least one electronic device, the at least one electronic device executes the above-mentioned switching method. The embodiment of the present application further provides a computer program product comprising instructions. When the computer program product is run on at least one electronic device, the at least one electronic device executes the above-mentioned switching method.

[0319] It should be understood that in this application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0320] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A switching method, characterized in that: The method comprises: Sending prediction capability information, where the prediction capability information indicates the ability of the artificial intelligence (AI) model to predict handover (HO); receiving prediction configuration information, where the prediction configuration information is used to indicate a triggering condition for triggering the predicted HO; When the trigger condition is met, HO is performed according to the prediction result of the AI ​​model.

2. The method according to claim 1, characterized in that The triggering conditions include making predictions according to a prediction period and / or satisfying a triggering event.

3. The method according to claim 2, characterized in that The trigger event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell, and the detection value is a detection value of a reference signal received power RSRP and / or a reference signal received quality RSRQ.

4. The method according to claim 2, characterized in that The triggering event includes one or more of the following: The HO failure rate in the first time interval is greater than the first threshold, the number of HO failures in the second time interval is greater than the first value, the probability of premature HO in the third time interval is greater than the second threshold, the number of premature HOs in the fourth time interval is greater than the second value, the probability of late HO in the fifth time interval is greater than the third threshold, the number of late HOs in the sixth time interval is greater than the third value, the probability of HO to an incorrect cell in the seventh time interval is greater than the fourth threshold, and the number of HO to an incorrect cell in the eighth time interval is greater than the fourth value.

5. The method according to any one of claims 1 to 4, characterized in that When the trigger condition is met, performing HO according to the prediction result of the AI ​​model includes: When the trigger condition is met, obtaining a HO prediction result using the AI ​​model; Sending HO prediction information, where the HO prediction information includes at least switching time information and target cell information.

6. The method according to claim 5, characterized in that The switching time information indicates a switching time corresponding to switching to at least one candidate cell, or a switching time corresponding to switching to one or more recommended cells; The target cell information indicates the at least one candidate cell, or the recommended one or more cells.

7. The method according to any one of claims 1 to 4, characterized in that When the trigger condition is met, performing HO according to the prediction result of the AI ​​model specifically includes: receiving a second HO command; triggering HO prediction and obtaining a HO prediction result using the AI ​​model, or, when HO prediction is triggered by a trigger condition, obtaining a HO prediction result using the AI ​​model; When it is determined that a premature HO occurs according to the switching time in the HO prediction result and the second HO command reception timing, first indication information is sent, where the first indication information is used to indicate the occurrence of a premature HO and request to maintain a radio resource control RRC connection with the source cell.

8. The method according to claim 7, characterized in that The first indication information is further used to indicate a target cell for HO; The first indication information is further used to indicate time information for postponing HO, where the time information for postponing HO indicates a deferred switching time corresponding to switching to at least one candidate cell, or a deferred switching time corresponding to switching to a target cell indicated by the second HO command, or a deferred switching time corresponding to switching to a recommended target cell, where the at least one candidate cell is configured by a network device.

9. The method according to any one of claims 1 to 4, characterized in that When the trigger condition is met, performing HO according to the prediction result of the AI ​​model specifically includes: receiving a third HO command, where the third HO command is used to at least indicate a target cell for the HO; triggering HO prediction, obtaining a HO prediction result using the AI ​​model, or obtaining a HO prediction result using the AI ​​model when HO prediction is triggered by a trigger condition; When the target cell is determined to be an incorrect cell according to the HO prediction result, not accessing the target cell; Perform cell selection and complete radio resource control RRC connection with the third cell, and carry the source cell identifier and the target cell identifier in the RRC connection request message or RRC connection establishment completion message sent to the third cell, so that the third cell sends a radio link failure RLF indication to the target cell, and the RLF indication is used to enable the target cell to send an HO report to the source cell, and the HO report indicates that the type of the handover failure HOF is HO to an error cell.

10. The method according to claim 2, characterized in that The AI ​​model's ability to predict HO includes the ability to predict handover beams, and the triggering events include: The detection value is lower than the threshold of the serving cell and / or higher than the threshold of the adjacent cell; Alternatively, the detected value is lower than the average beam quality of the serving cell and / or higher than the average beam quality of the neighboring cell; Alternatively, the detected value is lower than the average quality of N optimal beams of the serving cell and / or higher than the average quality of M optimal beams of the neighboring cell, where N and M are positive integers; The detection value is a detection value of reference signal received power RSRP and / or reference signal received quality RSRQ.

11. The method according to any one of claims 1-2, 4, and 10, characterized in that: When the trigger condition is met, performing HO according to the prediction result of the AI ​​model specifically includes: When the trigger condition is met, using the AI ​​model to obtain a beam switching prediction result; Send beam prediction information, where the beam prediction information includes beam index information and beam switching time information.

12. The method according to claim 11, characterized in that The beam switching time information indicates a switching time to at least one beam corresponding to at least one candidate cell, or a switching time to a recommended beam corresponding to at least one candidate cell.

13. A switching method, characterized in that: The method comprises: receiving prediction capability information indicating an ability of an artificial intelligence (AI) model to predict handover (HO); Prediction configuration information is sent according to the prediction capability information, where the prediction configuration information is used to indicate a triggering condition for triggering the predicted HO.

14. The method according to claim 13, characterized in that The triggering conditions include making predictions according to a prediction period and / or satisfying a triggering event.

15. The method according to claim 14, characterized in that The trigger event includes a detection value being lower than a threshold of a serving cell and / or higher than a threshold of a neighboring cell, and the detection value is a detection value of a reference signal received power RSRP and / or a reference signal received quality RSRQ.

16. The method according to claim 14, characterized in that The triggering event includes one or more of the following: The HO failure rate in the first time interval is greater than the first threshold, the number of HO failures in the second time interval is greater than the first value, the probability of premature HO in the third time interval is greater than the second threshold, the number of premature HOs in the fourth time interval is greater than the second value, the probability of late HO in the fifth time interval is greater than the third threshold, the number of late HOs in the sixth time interval is greater than the third value, the probability of HO to an incorrect cell in the seventh time interval is greater than the fourth threshold, and the number of HO to an incorrect cell in the eighth time interval is greater than the fourth value.

17. The method according to any one of claims 13 to 16, characterized in that The method further comprises: receiving HO prediction information, where the HO prediction information includes at least handover time information and target cell information; A first HO command is sent according to the HO prediction information.

18. The method according to claim 17, characterized in that The switching time information indicates a switching time corresponding to switching to at least one candidate cell, or a switching time corresponding to switching to one or more recommended cells. The target cell information indicates the at least one candidate cell, or the recommended one or more cells.

19. The method according to any one of claims 13 to 16, characterized in that The method further comprises: Send the second HO command; First indication information is received, where the first indication information is used to indicate that a premature HO occurs and to request to maintain a radio resource control (RRC) connection with a source cell.

20. The method according to claim 19, wherein The first indication information is further used to indicate a target cell for HO; The first indication information is further used to indicate time information for postponing HO, where the time information for postponing HO indicates a deferred switching time corresponding to switching to at least one candidate cell, or a deferred switching time corresponding to switching to a target cell indicated by the second HO command, or a deferred switching time corresponding to switching to a recommended target cell, where the at least one candidate cell is configured by a network device.

21. The method according to any one of claims 13 to 16, characterized in that The method further comprises: sending a third HO command, where the third HO command is used to at least indicate a target cell for the HO; Receive the HO report sent by the target cell, where the HO report indicates that the type of the handover failure HOF is HO to the wrong cell, and the HO report is sent by the target cell after receiving the radio link failure RLF sent by the third cell, where the third cell is the cell for the terminal device to perform cell selection and complete the radio resource control RRC connection.

22. The method according to claim 14, wherein The AI ​​model's ability to predict HO includes the ability to predict handover beams, and the triggering events include: The detection value is lower than the threshold of the serving cell and / or higher than the threshold of the adjacent cell; Alternatively, the detected value is lower than the average beam quality of the serving cell and / or higher than the average beam quality of the neighboring cell; Alternatively, the detected value is lower than the average quality of N optimal beams of the serving cell and / or higher than the average quality of M optimal beams of the neighboring cell, where N and M are positive integers; The detection value is a detection value of reference signal received power RSRP and / or reference signal received quality RSRQ.

23. The method according to any one of claims 13-14, 16 and 22, characterized in that The method further comprises: receiving beam prediction information, where the beam prediction information includes beam index information and beam switching time information; Send a fourth HO command according to the beam prediction information.

24. The method according to claim 23, wherein The beam switching time information indicates a switching time to at least one beam corresponding to at least one candidate cell, or a switching time to a recommended beam corresponding to at least one candidate cell.

25. A terminal device, characterized in that: including processor and memory; The processor is coupled to the memory; The memory is used to store computer programs and / or instructions; The processor is configured to execute the computer program and / or instructions stored in the memory to implement the switching method according to any one of claims 1 to 12.

26. A network device, characterized in that: including processor and memory; The processor is coupled to the memory; The memory is used to store instructions The processor is configured to execute the computer program or instructions stored in the memory to implement the switching method according to any one of claims 13 to 24.

27. A computer storage medium, characterized in that Used to store a computer program, which, when executed, is used to implement the switching method according to any one of claims 1 to 12, or to implement the switching method according to any one of claims 13 to 24.

Images