Data transmission method and apparatus, and storage medium

MYPI2025004901A0Undetermined Publication Date: 2024-08-15HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
MYPI2025004901
Authority / Receiving Office
MY · MY
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-02-15
Filing Date
2024-01-19
Publication Date
2024-08-15

AI Technical Summary

Technical Problem

In wireless communication systems that apply AI technology, network equipment cannot collect required AI data in a timely manner, resulting in network equipment being unable to train, reason, or optimize AI models in a timely manner.

Method used

By carrying the event information of the AI ​​event in the AI ​​data request, the second network device can timely return the AI ​​data based on the information when detecting that the event is met, thereby realizing timely transmission and use of the AI ​​data.

Benefits of technology

It meets the diverse data needs of network devices, improves the training, inference and optimization effects of AI models, and reduces errors or unnecessary data transmission caused by changes in AI event configuration parameters.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure PI2025004901_DRAWING_001
    Figure PI2025004901_DRAWING_001
  • Figure PI2025004901_DRAWING_002
    Figure PI2025004901_DRAWING_002
  • Figure PI2025004901_DRAWING_003
    Figure PI2025004901_DRAWING_003
Patent Text Reader

Abstract

Embodiments of this application disclose a data transmission method and apparatus, and a storage medium, and pertain to the field of communication technologies. In embodiments of this application, when sending an AI data request to a second network device, a first network device may include event information of a first AI event in the AI data request, where the first AI event is an event that triggers the second network device to send AI data. In this way, when detecting that the first AI event is satisfied, the second network device can send, in a timely manner based on the event information of the first AI event, AI data needed by the first network device, thereby satisfying diversified data requirements of the first network device.
Need to check novelty before this filing date? Find Prior Art

Description

Data transmission method, device and storage medium

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on February 15, 2023, with application number 202310164089.4 and application name “Data transmission 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 data transmission method, device, and storage medium. Background Art

[0003] With the advancement of data storage and computing capabilities, artificial intelligence (AI) technology is increasingly being used. For example, the 3rd Generation Partnership Project (3GPP) proposed applying AI technology to wireless communication systems to improve network performance and user experience through intelligent data collection and analysis.

[0004] However, in wireless communication systems that apply AI technology, there is currently no method to transmit AI data between network devices, resulting in the inability of network devices to collect the AI ​​data they need in a timely manner.

[0005] Summary of the Invention

[0006] The present application provides a data transmission method, device, and storage medium, which enable network devices to collect the AI ​​data they need in a timely manner to meet diverse data needs.

[0007] To achieve the above objectives, this application adopts the following technical solutions:

[0008] In a first aspect, a data transmission method is provided, which is used for a first network device. The method includes: sending an AI data request to a second network device, where the AI ​​data request includes event information of a first AI event, where the first AI event is used to trigger the second network device to send AI data to the first network device; and receiving the AI ​​data.

[0009] In the present application, when the first network device sends an AI data request to the second network device, the AI ​​data request may include event information of a first AI event, where the first AI event is an event that triggers the second network device to send AI data. In this way, the second network device can send AI data to the first network device based on the event information of the first AI event when it detects that the first AI event is satisfied. Since the second network device can promptly return the AI ​​data required by the first network device when triggered by a trigger event specified by the first network device, it meets the diverse data needs of the first network device and can improve the effectiveness of the first network device in training, reasoning, or optimizing AI models.

[0010] In a first possible implementation, the event information of the first AI event includes an event identifier and configuration parameters of the first AI event. In this way, the second network device can directly return the AI ​​data required by the first network device based on the event identifier and configuration parameters of the first AI event.

[0011] In a second possible implementation manner, the event information of the first AI event includes an event identifier of the first AI event.

[0012] In this implementation, before sending an AI data request to the second network device, the first network device may pre-transmit an AI event information set to the second network device, instructing the second network device to store the AI ​​event information set. The AI ​​event information set includes configuration parameters and an event identifier for at least one AI event, where the at least one AI event includes the first AI event. In other words, the first network device may pre-configure the event identifier and corresponding configuration parameters of the AI ​​event into the second network device. Thus, when the first network device obtains AI data, it includes the event identifier of the first AI event in its AI data request, but not the configuration parameters. Based on the event identifier of the first AI event, the second network device obtains the configuration parameters of the first AI event from its pre-configured event information for the AI ​​event, and then sends the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event. This reduces the signaling overhead of the first network device requesting AI data.

[0013] Optionally, in the first possible implementation manner or the second possible implementation manner mentioned above, the first network device may further send a configuration update request to the second network device, where the configuration update request includes the event identifier and update information of the first AI event, and the configuration update request is used to request the second network device to update the configuration parameters of the first AI event; and receive a configuration update response sent by the second network device in response to the configuration update request, where the configuration update response is used to indicate whether the second network device has successfully or failed to update the configuration parameters of the first AI event.

[0014] In the present application, the first network device can update the configuration parameters of the AI ​​event configured by itself in the second network device through the above-mentioned configuration update request, thereby reducing errors or unnecessary data transmission caused by changes in the configuration parameters of the AI ​​event.

[0015] Optionally, the update information in the configuration update request includes at least one configuration parameter and an update operation corresponding to the at least one configuration parameter, and the update operation includes one of modification, addition and deletion.

[0016] Optionally, in the present application, the first AI event includes a prediction trigger event, and the prediction trigger event includes a trigger event based on a prediction result of the AI ​​model of the second network device. For example, the prediction trigger event may refer to an event in which the prediction result of the AI ​​model of the second network device meets a set condition of the first network device.

[0017] The prediction trigger event includes at least one of a predicted value of a first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, and the first set threshold is greater than the second set threshold.

[0018] For example, the first measurement item may be device load, device energy consumption, or terminal traffic.

[0019] When the first measurement item is device load, the prediction trigger event may include one or more of the following: the predicted value of the device load being higher than a corresponding first load threshold, or the predicted value of the device load being lower than a corresponding second load threshold, where the first load threshold is greater than the second load threshold. In this case, based on the prediction trigger event, the second network device may return AI data related to the device load to the first network device for use in implementing the AI ​​use case of load balancing on the first network device.

[0020] When the first measurement item is device energy consumption, the prediction trigger event may include one or more of the following: the predicted value of device energy consumption is higher than a corresponding first energy consumption threshold, or the predicted value of device energy consumption is higher than a corresponding second energy consumption threshold. The first energy consumption threshold is higher than the second energy consumption threshold. In this case, based on the prediction trigger event, the second network device may return AI data related to device energy consumption to the first network device, for use by the first network device in implementing the AI ​​use case of network energy saving.

[0021] When the first measurement item is terminal traffic, the prediction type trigger event may include one or more of the predicted value of the terminal traffic being higher than the corresponding first traffic threshold and the predicted value of the terminal traffic being lower than the corresponding second traffic threshold. The first traffic threshold is greater than the second traffic threshold. In this case, the second network device can return AI data related to the terminal to the first network device based on the prediction type trigger event, so that the first network device can implement the AI ​​use case of mobility optimization. Of course, the first measurement item can also be other measurement items, and accordingly, the prediction type trigger event can also include other events, which is not limited in this application.

[0022] Optionally, in the present application, the first AI event may also include a current measurement type trigger event, and the current measurement type trigger event includes a trigger event based on a current measurement result of the second network device.

[0023] Among them, the current measurement type trigger event includes at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

[0024] Optionally, the first measurement item includes any one of device load, device energy consumption and terminal traffic, the second measurement item includes the terminal's mobile path or the terminal's service, and the third measurement item includes any one of device load, device energy consumption and terminal traffic.

[0025] It should be noted that in an embodiment of the present application, when the second measurement item is the mobile path of the terminal or the service of the terminal, based on the current measurement type trigger event, the second network device can return AI data related to the terminal switching to the first network device for the first network device to implement the AI ​​use case of mobility optimization.

[0026] Optionally, when the first AI event includes the prediction trigger event or the current measurement trigger event, the configuration parameters of the first AI event include indication information of the measurement item and / or a set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the duration of the first AI event. The indication information of the measurement item is used to indicate the measurement item. For example, the indication information of the measurement item may be an identifier of the measurement item.

[0027] Optionally, in the present application, the first AI event may also include a current action type trigger event, and the current action type trigger event includes a trigger event based on the current action of the second network device.

[0028] Among them, the current action trigger event may include at least one of cell activation, terminal switching success, terminal switching in progress, and execution of AI use case.

[0029] When the first AI event includes cell activation, the configuration parameters of the first AI event may include a cell identifier list and the duration of the first AI event. Based on the cell activation AI event, the second network device may return AI data related to the target cell for the first network device to implement AI use cases such as mobility optimization or load balancing, for example, implementing cell handover based on an AI model.

[0030] When the first AI event includes at least one of a successful terminal handover and a handover in progress, the configuration parameters of the first AI event may include a terminal identification list and the duration. Based on the AI ​​event of a successful terminal handover or a handover in progress, the second network device may return AI data related to the terminal handover for use by the first network device in implementing an AI use case for mobility optimization, for example, implementing terminal handover based on an AI model.

[0031] When the first AI event includes executing an AI use case, the configuration parameters of the first AI event include the AI ​​use case and the duration. Based on the AI ​​event in executing the AI ​​use case, the second network device may return AI data related to the executed AI use case for the first network device to implement the corresponding AI use case.

[0032] In a second aspect, a data transmission method is provided, which is used for a second network device. The method includes: receiving an AI data request sent by a first network device, the AI ​​data request including event information of a first AI event, and the first AI event is used to trigger the second network device to send AI data to the first network device; and sending the AI ​​data to the first network device based on the event information of the first AI event.

[0033] In a first possible implementation manner, the event information of the first AI event includes the event identifier and configuration parameters of the first AI event; accordingly, the implementation process of the second network device sending the AI ​​data to the first network device based on the event information of the first AI event may be: the second network device sends the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event.

[0034] In a second possible implementation, the event information of the first AI event includes an event identifier of the first AI event; accordingly, the implementation process of the second network device sending the AI ​​data to the first network device based on the event information of the first AI event includes: obtaining configuration parameters of the first AI event based on the event identifier of the first AI event; and sending the AI ​​data to the first network device based on the event identifier of the first AI event and the configuration parameters.

[0035] In this implementation, before receiving the AI ​​data request sent by the first network device, the second network device may also receive an AI event information set sent by the first network device, wherein the AI ​​event information set includes configuration parameters and event identifiers of at least one AI event, and the at least one AI event includes the first AI event; and store the AI ​​event information set to complete the configuration of the AI ​​event.

[0036] Optionally, in the first possible implementation or the second possible implementation described above, the second network device may further receive a configuration update request sent by the first network device, the configuration update request including the event identifier and update information of the first AI event; update the configuration parameters of the first AI event based on the event identifier and update information of the first AI event; and send a configuration update response to the first network device, the configuration update response being used to indicate the success or failure of the configuration parameter update for the first AI event. The update information includes at least one configuration parameter and an update operation corresponding to each of the at least one configuration parameter, and the update operation includes one of modification, addition, and deletion. By interacting with the first network device to update the event information of the configured AI event, erroneous or unnecessary AI data transmission caused by changes in the configuration parameters of the AI ​​event is avoided.

[0037] Optionally, in the present application, the first AI event includes a prediction-type trigger event, and the prediction-type trigger event includes a trigger event based on a prediction result of an AI model of the second network device.

[0038] The prediction trigger event includes at least one of a predicted value of a first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, and the first set threshold being higher than the second set threshold. The first measurement item may be any one of device load, device energy consumption, and terminal traffic.

[0039] Optionally, in the present application, the first AI event includes a current measurement type trigger event, and the current measurement type trigger event includes a trigger event based on a current measurement result of the second network device.

[0040] Among them, the current measurement type trigger event includes at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

[0041] The first measurement item includes any one of device load, device energy consumption and terminal traffic, the second measurement item includes the terminal's mobile path or the terminal's service, and the third measurement item includes any one of device load, device energy consumption and terminal traffic.

[0042] Optionally, when the first AI event includes the prediction-type trigger event or the current measurement-type trigger event, the configuration parameters of the first AI event include indication information of the measurement item and / or a set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the duration of the first AI event. In this case, the implementation process of the second network device sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event may include: detecting the first AI event based on the event identifier of the first AI event, the indication information of the measurement item, and / or the set threshold corresponding to the measurement item; and sending the AI ​​data to the first network device if the first AI event is continuously detected within the duration.

[0043] Optionally, in the present application, the first AI event includes a current action type trigger event, and the current action type trigger event includes a trigger event based on a current action of the second network device.

[0044] Among them, the current action trigger event includes at least one of cell activation, terminal switching success, terminal switching in progress, and execution of AI use case.

[0045] When the first AI event includes cell activation, the configuration parameters of the first AI event include a cell identifier list and a duration of the first AI event; the implementation process of sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event may include: detecting the on / off state of the target cell based on the event identifier of the first AI event, where the cell identifier of the target cell is any cell identifier in the cell identifier list; and sending the AI ​​data to the first network device when it is detected that the target cell is in the on state for the duration.

[0046] When the first AI event includes a successful terminal switching, the configuration parameters of the first AI event include a terminal identification list and a duration of the first AI event; the implementation process of sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event may include: detecting the resident status of the target terminal based on the event identifier of the first AI event, where the identifier of the target terminal is any identifier in the terminal identification list, and the target terminal is a terminal switched to the second network device; and sending the AI ​​data to the first network device when it is detected that the target terminal continues to reside on the second network device within the duration.

[0047] When the first AI event includes executing an AI use case, the configuration parameters of the first AI event include the AI ​​use case and the duration of the first AI event; the implementation process of sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event may include: detecting the execution of the AI ​​use case by the second network device based on the event identifier of the first AI event; and sending the AI ​​data to the first network device when it is detected that the second network device executes the AI ​​use case within the duration.

[0048] According to a third aspect, a data transmission device is provided, comprising at least one module, wherein the at least one module is configured to execute the data transmission method according to the first or second aspect.

[0049] In a fourth aspect, a data transmission device is provided, comprising a processor, wherein the processor is configured to execute at least one program instruction or code stored in a memory to implement the data transmission method described in the first or second aspect above.

[0050] The data transmission device described in the third aspect or the fourth aspect above may be a network device, or a device used for a network device such as a chip, or a device used to implement some functions of a network device.

[0051] In a fifth aspect, a data transmission system is provided, which includes a first network device and a second network device, wherein the first network device is used to execute the data transmission method described in the first aspect, and the second network device is used to execute the data transmission method described in the second aspect.

[0052] In a sixth aspect, a computer-readable storage medium is provided, wherein instructions are stored in the computer-readable storage medium. When the instructions are executed on a network device, the network device executes the data transmission method described in the first or second aspect above.

[0053] In a seventh aspect, a computer program product comprising instructions is provided, which, when executed on a computer, enables the computer to execute the data transmission method described in the first or second aspect above.

[0054] The technical effects obtained in the above-mentioned second to seventh aspects are similar to the technical effects obtained by the corresponding technical means in the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] FIG1 is a system architecture diagram of a data transmission method provided in an embodiment of the present application;

[0056] FIG2 is a schematic structural diagram of a base station with a centralized unit and a distributed unit separation architecture provided by an embodiment of the present application;

[0057] FIG3 is a schematic structural diagram of a data transmission device provided in an embodiment of the present application;

[0058] FIG4 is a flow chart of a data transmission method provided in an embodiment of the present application;

[0059] FIG5 is a flowchart of another data transmission method provided in an embodiment of the present application;

[0060] FIG6 is a flowchart of another data transmission method provided in an embodiment of the present application;

[0061] FIG7 is a schematic structural diagram of another data transmission device provided in an embodiment of the present application;

[0062] FIG8 is a schematic structural diagram of another data transmission device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0063] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0064] Before explaining the embodiments of the present application in detail, the application scenarios involved in the embodiments of the present application are first introduced.

[0065] In a wireless communication system using AI technology, a first network device can collect operational data from access network devices, core network devices, terminals, or other management devices to train an AI model. The trained AI model can then make predictions based on the operational data of a second network device and adjust the policy of the second network device based on the prediction results. Subsequently, the first network device can obtain the adjusted AI data from the second network device to further optimize the AI ​​model.

[0066] For example, in the first possible application scenario, AI technology can be used to achieve network energy conservation. In this scenario, a first network device can collect data from the RAN to train an AI model. The trained AI model can then be used to predict the energy efficiency and load status of a second network device in the RAN. Based on these predicted energy efficiency and load status, an energy conservation policy can be configured for the second network device to reduce energy consumption while maintaining a balance between network performance and energy efficiency. Subsequently, the second network device configured with the energy conservation policy can provide feedback to the first network device for AI model optimization.

[0067] In a second possible application scenario, AI technology can be used to achieve network load balancing. For example, a first network device can collect measurement data, feedback data, and historical data from network devices such as terminals and base stations to train an AI model. The trained AI model is then used to predict the cell load of a second network device. Based on the predicted cell load, the terminal handover process or handover parameters are optimized to improve load balancing performance and user experience. Subsequently, the second network device can provide feedback to the first network device for optimization of the AI ​​model.

[0068] In a third possible application scenario, AI technology can be used to achieve mobility optimization. For example, a first network device collects terminal data from the RAN and uses it to train an AI model. The trained AI model predicts the terminal's mobility path, and then performs radio resource management actions based on the predicted terminal's mobility path, such as selecting a target cell for handover. Subsequently, the second network device corresponding to the selected target cell can feed back data to the first network device for AI model optimization.

[0069] The above application scenarios are several possible application scenarios for applying AI technology to NR as shown in the embodiments of this application. Obviously, in NR, AI technology can also be used to achieve other network optimizations. The data transmission method provided in the embodiments of this application can be used to realize the transmission of AI data between network devices in NR using AI technology. Among them, AI data refers to data collected by network devices in various scenarios such as those described above for AI model training, reasoning, and optimization.

[0070] Figure 1 is a diagram of a system architecture involved in a data transmission method provided in an embodiment of the present application. As shown in Figure 1, the system architecture includes a first network device 101 and a second network device 102. The first network device 101 and the second network device 102 have established a communication connection.

[0071] The system architecture is applicable to a variety of communication networks, for example, fifth-generation (5G) mobile communication systems such as long-term evolution (LTE) networks, enhanced long-term evolution (eLTE) networks, and new radio (NR) networks, as well as sixth-generation (6G) communication systems or other future-oriented communication systems. In addition, the communication system provided in this application can be used in terrestrial networks (TN) and / or non-terrestrial networks (NTN), without limitation.

[0072] The first network device 101 can collect AI data from the second network device 102 and use the collected AI data for AI model training, inference, or optimization. For example, the first network device 101 can send an AI data request to the second network device 102, where the AI ​​data request includes event information about a first AI event. The first AI event triggers the second network device to send AI data to the first network device. After receiving the AI ​​data request, the second network device 102 sends AI data to the first network device based on the event information about the first AI event.

[0073] In one possible implementation, first network device 101 may be deployed with an AI model. In this case, when its own AI model requires training data, inference data, or optimization data, first network device 101 may send an AI data request to second network device 102. After receiving the AI ​​data request, second network device 102 may detect a first AI event based on the event information of the first AI event in the AI ​​data request. If the first AI event is detected, the second network device 102 returns the AI ​​data required by the first network device to first network device 101.

[0074] Optionally, an AI model may also be deployed on the second network device 102. In this case, the second network device 102 may detect the first AI event based on the prediction result of the AI ​​model deployed on the second network device 102.

[0075] In one example, the first network device 101 or the second network device 102 may be an access network device in a radio access network (RAN). In the present application, the access network device may be various types of base stations, such as an evolved NodeB (eNodeB), a transmission reception point (TRP), a next generation NodeB (gNB), or base stations in other future-oriented communication systems. In addition, when the eNB is connected to the 5G Core network (5GC), the LTE eNB may also be referred to as an eLTE eNB, and the eLTE eNB may also be considered as a base station device in the NR system; the access network device may also be a wireless relay node, a wireless backhaul node, or other devices. The base station may be a macro base station, a micro base station, an indoor station, or the like. The access network device may also be a module or unit that performs some of the functions of a base station. For example, the access network device may also include a central unit (CU) and / or a distributed unit (DU). Exemplarily, the first network device 101 or the second network device 102 may be a base station with a centralized unit (CU) and a distributed unit (DU) separated architecture. For example, FIG2 shows a schematic diagram of the structure of a base station with a CU and DU separated architecture. As shown in FIG2, the base station includes a CU 201 and a DU 202.

[0076] CU 201 includes a central unit control plane (CU-CP) 2011 and a central unit user plane (CU-UP) 2012. The CU-CP 2011 and CU-UP 2012 communicate over the E1 interface. The CU-CP 2011 implements RRC and the control plane's packet data convergence protocol-control (PDCP-C) functions. The CU-UP 2012 implements the service data adaptation protocol (SDAP) and the user plane's packet data convergence protocol-user (PDCP-U) functions.

[0077] The DU 202 can communicate with the CU-CP 2011 via the F1-C interface and with the CU-UP 2012 via the F1-U interface. The DU 202 implements radio link control (RLC), medium access control (MAC), and physical layer (PHY) functions.

[0078] In different systems, CU (including CU-CP or CU-UP) or DU may have different names, but those skilled in the art can understand their meanings. For example, in an open radio access network (O-RAN) system, CU may also be called O-CU (open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, and CU-UP may also be called O-CU-UP. For the convenience of description, this application uses CU, CU-CP, CU-UP, and DU as examples for description. It should be noted that O-RAN is an intelligent, open access network. The O-RAN architecture integrates a modular base station software stack on existing hardware, allowing baseband and wireless unit components from discrete suppliers to run seamlessly together. The main feature of the O-RAN architecture is the separation of software and hardware, which realizes the virtualization of network functions and the standardization of hardware.

[0079] In one example, the first network device 101 or the second network device 102 may also be a terminal. In the present application, the terminal may be any type of device that provides voice and / or data connectivity to users and can communicate with one or more core networks via a radio access network (RAN). The terminal may also be referred to as a terminal device, user equipment (UE), mobile station, mobile terminal, etc. The terminal can be applied to various communication scenarios, such as device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IOT), virtual reality (VR), augmented reality (AR), industrial control, self-driving, telemedicine, smart grid, smart furniture, smart office, smart wearable, smart transportation, smart city, etc. Terminal devices can be mobile phones, tablet computers, computers with wireless transceiver functions, wearable devices, aerospace equipment, drone equipment, etc. The chips used in the above devices can also be called terminals.

[0080] In an example, the first network device 101 or the second network device 102 may also be a core network device or an operation administration and maintenance (OAM) device or other network device with AI functions, which is not specifically limited in the embodiments of the present application. Among them, the core network device may include one or more core network network elements. Taking the 5G core network as an example, the 5G core network includes an access and mobility management function (AMF) network element responsible for services such as mobility management and access management, a session management function (SMF) network element responsible for session management, a user plane function (UPF) network element responsible for data packet routing and forwarding and quality of service (QoS) control on the user plane, a policy control function (PCF) network element, etc. The above-mentioned core network network elements can work independently or be combined to implement certain control functions, such as: AMF, SMF and PCF can be combined together as a core network device.

[0081] FIG3 is a schematic structural diagram of a data transmission device shown in an embodiment of the present application.

[0082] The data transmission device may be a network device, or a device for a network device, such as a chip, or a device for implementing part of the functions of a network device, such as a CU or DU in a split base station. The network device in this application may refer to an access network device, a terminal, or a core network device.

[0083] As shown in Figure 3, the data transmission device may include at least one processor 301, a communication bus 302, a memory 303, and at least one communication interface 304. It should be noted that the device structure shown in Figure 3 does not constitute a limitation on the data transmission device, and may include more or fewer components than shown, or combine certain components, or arrange components differently, which is not limited in this embodiment of the present application. The following is a detailed introduction to the various components of the data transmission device in conjunction with Figure 3:

[0084] The processor 301 is the control center of the data transmission device, which can be a processor or a general term for multiple processing elements. For example, the processor 301 can be a general-purpose central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the present application, such as one or more microprocessors (digital signal processors, DSPs), or one or more field programmable gate arrays (FPGAs). Among them, the processor 301 can perform various functions of the data transmission device by running or executing the software program stored in the memory 303 and calling the data stored in the memory 303. For example, the actions of the first network device or the second network device in the embodiments shown in Figures 4 to 6 can be executed by calling the data in the memory by the corresponding processor.

[0085] As an embodiment, the processor 301 may include one or more CPUs, such as CPU0 and CPU1 shown in FIG. 3 .

[0086] As an embodiment, the data transmission apparatus may include multiple processors, such as processor 301 and processor 305 shown in FIG3 . Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. A processor herein may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0087] Communication bus 302 may include a path for transmitting information between the aforementioned components. Communication bus 302 may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. Such buses may be classified as address buses, data buses, or control buses. For ease of illustration, FIG3 shows a single thick line, but this does not imply a single bus or type of bus.

[0088] The memory 303 can be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited to this. The memory 303 can be independent and connected to the processor 301 via the communication bus 302. The memory 303 can also be integrated with the processor 301. Among them, the memory 303 is used to store the software program that executes the solution provided in the embodiment of the present application, and is controlled by the processor 301 to execute.

[0089] The communication interface 304 is used to communicate with other devices or communication networks, such as Ethernet, RAN, wireless local area networks (WLAN), etc. The communication interface 304 may include a receiving unit to implement a receiving function, and a sending unit to implement a sending function.

[0090] When the data transmission device is a network device, the network device may be a general network device or a dedicated network device, for example, a desktop computer, a portable computer, a network server, etc., which is not limited in the present embodiment.

[0091] Next, the data transmission method provided in the embodiment of the present application is explained in detail. It is understood that the method provided in the present application can be executed by a network device or by a module (such as a chip) for a network device. The following describes the data transmission method provided in the present application using the network device (the first network device and the second network device) as the execution subject.

[0092] FIG4 is a flow chart of a data transmission method according to an embodiment of the present application. As shown in FIG4 , the method includes the following steps:

[0093] Step 401: A first network device sends an AI data request to a second network device, where the AI ​​data request includes event information of a first AI event.

[0094] The first network device and the second network device may be applicable to the system architecture shown in Figure 1. For example, the first network device and the second network device may both be access network devices, or the first network device may be an access network device and the second network device may be a core network device or a terminal, or the first network device may be a core network device and the second network device may be an access network device, which is not limited in this embodiment of the present application.

[0095] In an embodiment of the present application, data related to various operations of AI model application such as AI model training, reasoning or optimization can be collectively referred to as AI data. Corresponding to different operations, for example, it can be divided into training data corresponding to AI model training, reasoning data corresponding to AI model reasoning, and feedback data corresponding to AI model optimization. When the first network device needs to obtain training data to train the AI ​​model, or when the first network device needs to obtain reasoning data and use the AI ​​model for reasoning, or when the first network device needs to obtain feedback data from the second network device to optimize the AI ​​model, the first network device can send an AI data request to the second network device to request the above-mentioned required AI data.

[0096] The AI ​​data request may carry event information of a first AI event. The first AI event refers to an event that triggers the second network device to send AI data. Furthermore, the first AI event may be associated with an AI use case executed by the first network device. An AI use case refers to a matter of applying the reasoning results of an AI model to adjust a policy. For example, the AI ​​use case may include at least one of network energy saving, load balancing, and mobility optimization introduced in the aforementioned application scenario.

[0097] In some embodiments, the first AI event may include a prediction trigger event. The prediction trigger event refers to a trigger event based on a prediction result of an AI model of the second network device. For example, the prediction trigger event may refer to an event in which the prediction result of the AI ​​model of the second network device meets a condition set by the first network device.

[0098] Illustratively, a prediction trigger event may include at least one of a predicted value of a first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, where the first set threshold is greater than the second set threshold. The first measurement item is an object specified by the first network device and actually measured by the second network device.

[0099] For example, the first measurement item includes device load, and accordingly, the prediction-type trigger event may include the predicted value of the device load being higher than the corresponding first load threshold, and / or the predicted value of the device load being lower than the corresponding second load threshold, wherein the first load threshold is greater than the second load threshold. For another example, the first measurement item includes device energy consumption, and the corresponding prediction-type trigger event may include the predicted value of the device energy consumption being higher than the corresponding first energy consumption threshold, and / or the predicted value of the device energy consumption being higher than the corresponding second energy consumption threshold, wherein the first energy consumption threshold is greater than the second energy consumption threshold. For another example, the first measurement item includes terminal traffic, and the corresponding prediction-type trigger event may include the predicted value of the terminal traffic being higher than the corresponding first traffic threshold and the predicted value of the terminal traffic being lower than the corresponding second traffic threshold, wherein the first traffic threshold is greater than the second traffic threshold. Of course, the first measurement item may also be other measurement items, and accordingly, the prediction-type trigger event may also include other events, which is not limited in the embodiments of the present application.

[0100] In some embodiments, the first AI event may also include a current measurement trigger event, where the current measurement trigger event refers to a trigger event based on a current measurement result of the second network device. For example, the current measurement trigger event may refer to a current measurement value of the second network device satisfying a condition set by the first network device.

[0101] Exemplarily, the current measurement type trigger event may include at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, wherein the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

[0102] The predicted value of the measurement item before the current time may refer to the value at the current time predicted before the current time. For example, assuming the current time is T1, the predicted value before the current time may refer to the value at T1 predicted at T2. Where T2 is earlier than T1.

[0103] In addition, the above-mentioned first measurement item may include any one of device load, device energy consumption, and terminal traffic, the second measurement item may include the terminal's mobile path or the terminal's business, and the third measurement item may include any one of device load, device energy consumption, and terminal traffic.

[0104] For example, the first measurement item may include device load, and accordingly, the current measurement type trigger event may include the current measurement value of the device load being higher than a corresponding third load threshold, and / or the current measurement value of the device load being lower than a corresponding fourth load threshold, wherein the third load threshold is greater than the fourth load threshold. For another example, the first measurement item may include device energy consumption, and accordingly, the current measurement type trigger event may include the current measurement value of the device energy consumption being higher than a corresponding third energy consumption threshold, and / or the current measurement value of the device energy consumption being lower than a corresponding fourth energy consumption threshold, wherein the third energy consumption threshold is greater than the fourth energy consumption threshold. For another example, the second measurement item may include the terminal's moving path, and accordingly, the current measurement type trigger event may include the current measurement value of the terminal's moving path being different from a predicted value of the terminal's moving path before the current moment. For another example, the second measurement item may include the terminal's service, and accordingly, the current measurement type trigger event may include the current measurement value of the terminal's service being different from a predicted value of the terminal's service before the current moment. For another example, the third measurement item may include terminal traffic, and accordingly, the current measurement type trigger event may include the error ratio between the current measurement value of the terminal traffic and the predicted value before the current moment being higher than the corresponding fifth set threshold, and / or the error ratio between the current measurement value of the terminal and the predicted value before the current moment being lower than the corresponding sixth set threshold, wherein the fifth set threshold is greater than the sixth set threshold.

[0105] In some embodiments, the first AI event may also include a current action trigger event. The current action trigger event refers to a trigger event based on the current action of the second network device. For example, the current action trigger event may refer to an event in which the current action of the second network device meets a condition set by the first network device.

[0106] Exemplarily, the current action-type triggering event may include one or more of cell activation, successful terminal handover, terminal in handover, and execution of an AI use case.

[0107] In one possible implementation, the event information of the first AI event in the AI ​​data request may include an event identifier and configuration parameters for the first AI event, where the event identifier is used to identify the AI ​​event, and the configuration parameters are used to configure parameters required for detecting the corresponding event. The following examples illustrate different AI event configuration parameters.

[0108] Case 1: The first AI event includes any of the above-mentioned prediction trigger events.

[0109] In this case, the configuration parameters of the first AI event may include a set threshold corresponding to the measurement item and the duration of the first AI event. The duration of the first AI event refers to the duration required by the first network device for the second network device to detect the first AI event. In this embodiment of the application, the duration of the first AI event is also referred to as the trigger duration.

[0110] Optionally, the configuration parameters of the first AI event also include the prediction accuracy corresponding to the measurement item, and the prediction accuracy refers to the prediction accuracy that the second network device must meet for the predicted value of the corresponding measurement item. The prediction accuracy can be accuracy, precision, or recall, etc. Accuracy refers to the ratio between the number of samples correctly predicted by the AI ​​model and the total number of samples in the sample set, which is used to characterize how many samples in the sample set the AI ​​model predicts correctly. Precision refers to the ratio between the number of positive samples correctly predicted by the AI ​​model and the number of all positive samples predicted by the AI ​​model, which is used to characterize how many samples in the positive samples predicted by the AI ​​model are correctly predicted. Recall refers to the ratio between the number of positive samples correctly predicted by the AI ​​model and the number of positive samples included in the sample set, which is used to indicate how many samples in the positive samples included in the sample set are correctly predicted.

[0111] Optionally, the configuration parameters of the first AI event may further include a hysteresis parameter, which is used to indicate a hysteresis range allowed for the predicted value of the measurement item. For example, when the hysteresis parameter is a and the set threshold corresponding to the measurement item is b, the hysteresis range allowed for the predicted value of the measurement item is (b a, b + a). When the second network device detects that the predicted value of the measurement item is within the hysteresis range, it can determine that the first AI event has been detected.

[0112] For example, when the first AI event includes a predicted value of the device load that is higher than the corresponding first load threshold, the configuration parameters of the first AI event may include the first load threshold corresponding to the device load, trigger duration, hysteresis parameter, and prediction accuracy.

[0113] Among them, since the device load can be characterized by different sub-load items, the configuration parameters of the first AI event may include load thresholds corresponding to different sub-load items. For example, the device load can be characterized by one or more sub-load items including throughput, physical resource block (PRB) utilization, number of activated terminals, number of radio resource control (RRC) connections, comprehensive available capacity group, transport network layer (TNL) capacity index, and slice available capacity. Accordingly, the configuration parameters of the first AI event may include load thresholds of the above one or more sub-load items. Among them, the number of activated terminals refers to the number of terminals in an activated state.

[0114] In the case where the configuration parameters of the first AI event include the load thresholds of the above-mentioned multiple sub-load items, assuming that the predicted value of the device load is higher than the corresponding first load threshold, the event identifier of this AI event is AI Event A1, and the event information of the first AI event carried in the AI ​​data request can be as shown in the example in Table 1 below.

[0115] Table 1 Event information of the first AI event

[0116] Among them, the symbols ">", ">>" and ">>>" in Table 1 represent the hierarchy of the information cells. For example, ">" represents the first-level information cell, ">>" represents the second-level information cell, ">>>" represents the third-level information cell, and so on. The information cells at each level have a nested relationship according to the level. For example, the second-level information cell is included in the first-level information cell, and the third-level information cell is included in the second-level information cell. It can be understood that the second-level information cell and the third-level information cell are both included in the first-level information cell. If the above-mentioned symbols such as ">" appear in the examples of information cell design in this application, the meaning is the same as in Table 1 and will not be repeated.

[0117] The second case: the first AI event includes the above-mentioned current measurement trigger event.

[0118] In this case, the configuration parameters of the first AI event may include a set threshold and / or measurement item corresponding to the measurement item, and the configuration parameters of the first AI event may also include a triggering duration.

[0119] For example, when the first AI event includes any one or more current measurement type trigger events corresponding to the aforementioned first measurement item or third measurement item, the configuration parameters of the first AI event may include the set threshold and trigger duration corresponding to the first measurement item or the third measurement item.

[0120] Optionally, the configuration parameters of the first AI event further include a hysteresis parameter. For the relevant description of the hysteresis parameter, reference may be made to the relevant introduction in the aforementioned first case, and details will not be repeated here.

[0121] Optionally, when the measurement item of the first AI event involves other devices besides the second network device, for example, a terminal, the configuration parameter of the first AI event may further include an identifier of a measurement object, where the measurement object refers to an object to be measured by the second network device.

[0122] For example, when the first AI event includes a current measurement value of the device load that is higher than a corresponding third load threshold, the configuration parameters of the first AI event may include the third load threshold, trigger duration, and hysteresis parameter corresponding to the device load. The device load may include one or more sub-load items described in the first case above, and accordingly, the configuration parameters of the first AI event may include load thresholds for one or more sub-load items.

[0123] In the case where the configuration parameters of the first AI event include load thresholds for multiple sub-load items, assuming that the current measured value of the device load is higher than the corresponding third load threshold, the event identifier of this AI event is AI Event B1, and the event information of the first AI event carried by the AI ​​data request can be as shown in the example in Table 2.

[0124] Table 2 Event information of the first AI event

[0125] For another example, when the first AI event includes a terminal flow rate whose current measurement value and the predicted value before the current moment have an error ratio higher than the corresponding fifth set threshold, the configuration parameters of the first AI event may include a terminal identification list, a fifth set threshold, a trigger duration, and a hysteresis parameter. The terminal identification list includes at least one terminal identification, and the at least one terminal identification is used to indicate the terminal for which the measurement is being performed. That is, the terminal identified by the at least one terminal identification is the measurement object corresponding to the terminal flow rate.

[0126] The terminal traffic may include at least one of uplink traffic and downlink traffic. Based on this, the fifth set threshold may include at least one of an uplink traffic threshold and a downlink traffic threshold. In addition, the error ratio between the current measurement value of the terminal traffic and the predicted value before the current moment may be equal to (current measurement value - predicted value before the current moment) / predicted value before the current moment, or may be equal to (current measurement value - predicted value before the current moment) / current measurement value.

[0127] In the case where the configuration parameters of the first AI event include an uplink traffic threshold and a downlink traffic threshold, assuming that the error ratio between the current measured value of the terminal traffic and the predicted value before the current moment is higher than the corresponding fifth set threshold, the event identifier of this AI event is AI Event C3, and the event information of the first AI event carried in the AI ​​data request can be as shown in the example in Table 3 below.

[0128] Table 3 Event information of the first AI event

[0129] In other embodiments, when the first AI event includes a current measurement value of a second measurement item in a current measurement type trigger event that is different from a predicted value before the current moment, the configuration parameters of the first AI event may include indication information of the measurement item and a trigger duration.

[0130] Optionally, if the measurement item of the first AI event involves other devices besides the second network device, for example, a terminal, the configuration parameter of the first AI event may further include an identifier of a measurement object, where the measurement object refers to an object to be measured by the second network device.

[0131] For example, when the first AI event includes that the current measured value of the terminal's service is different from the predicted value before the current moment, the configuration parameters of the first AI event may include a terminal identification list, a service type identification, and a trigger duration. The terminal identification list includes at least one terminal identification, and the at least one terminal identification is used to indicate the terminal for which the measurement is targeted, that is, the measurement object. The service type identification is used to indicate the service to be measured, that is, the service type identification is indication information of the measurement item.

[0132] The terminal may have various service types, such as conversational voice, real-time gaming, automated process monitoring, mission-critical video user plane, low-latency enhanced mobile broadband (eMBB) applications for augmented reality, and intelligent transportation systems. Based on this, the configuration parameters of the first AI event may include identifiers of one or more of the aforementioned service types.

[0133] In the case where the configuration parameters of the first AI event include identifiers of the above-mentioned multiple service types, assuming that the current measured value of the terminal's service is different from the predicted value before the current moment, the event identifier of this AI event is AI Event C2, and the event information of the first AI event carried in the AI ​​data request may be as shown in the example in Table 4 below:

[0134] Table 4 Event information of the first AI event

[0135] In the third case, the first AI event includes the aforementioned current action triggering event.

[0136] Exemplarily, when the first AI event includes cell on, the configuration parameters of the first AI event may include a cell identifier list and a trigger duration. The cell identifier list includes at least one cell identifier, which indicates the cell targeted for measurement. That is, the at least one cell identifier indicates the cell to be measured by the second network device. In other words, the second network device can detect the first AI event by detecting the on / off state of the cell identified by the at least one cell identifier.

[0137] When the first AI event includes at least one of a successful terminal handover or a terminal in the process of handover, the configuration parameters of the first AI event include a terminal identifier list and a trigger duration. The terminal identifier list includes at least one terminal identifier, which indicates the terminal to be measured. That is, the at least one terminal identifier indicates the terminal to be measured by the second network device. The second network device can detect the first AI event by detecting the handover state of the terminal identified by the at least one terminal identifier.

[0138] For example, when the first AI event includes a successful terminal switching, assuming that the event identifier of the AI ​​event of successful terminal switching is AI Event E1, the event information of the first AI event included in the AI ​​data request may be as shown in Table 5 below.

[0139] Table 5 Event information of the first AI event

[0140] When the first AI event includes executing an AI use case, the configuration parameters of the first AI event may include the AI ​​use case and a trigger duration. The AI ​​use case may include at least one of network energy saving, load balancing, and mobility optimization described in the aforementioned application scenario.

[0141] For example, when the configuration parameters of the first AI event include the aforementioned multiple AI use cases, assuming that the event identifier of the AI ​​event of executing the AI ​​use case is AI Event F1, the event information of the first AI event carried in the AI ​​data request may be as shown in the example in Table 6 below.

[0142] Table 6 Event information of the first AI event

[0143] It should be noted that after the first network device sends an AI data request carrying the event identifier and configuration parameters of the first AI event to the second network device, the second network device may store the received event identifier and configuration parameters of the first AI event in correspondence to implement event information configuration for the first AI event. Subsequently, the second network device may send an AI data response to the first network device, where the AI ​​data response carries first indication information, indicating whether the second network device has successfully or failed to configure the event information for the first AI event. If configuration failure is indicated, the first indication information may also indicate the reason for the configuration failure.

[0144] It can be understood that the examples in Tables 1 to 6 above are merely possible implementations of the cell structure when the event information of the first AI event is transmitted using an AI data request, and do not constitute any limitation on the transmission method or cell structure of the event information of the first AI event.

[0145] Optionally, when storing the event identifier and configuration parameters of the first AI event, the second network device may store the event identifier and configuration parameters of the first AI event in correspondence with the device identifier of the first network device, so as to distinguish different AI events configured by different network devices.

[0146] In another possible implementation, the event information of the first AI event in the AI ​​data request may include an event identifier of the first AI event, but does not include configuration parameters of the first AI event.

[0147] In this case, before sending the AI ​​data request to the second network device, the first network device may also send an AI event information set to the second network device to instruct the second network device to store the AI ​​event information set. The AI ​​event information set includes an event identifier and configuration parameters of at least one AI event, where the at least one AI event includes the event identifier of the first AI event.

[0148] For example, the first network device may carry the AI ​​event information set through an AI event configuration request. The AI ​​event configuration request may be a request specifically for transmitting the AI ​​event information set. Alternatively, the AI ​​event configuration request may also reuse other requests between the first network device and the second network device. For example, the AI ​​event configuration request may be an Xn link establishment request (Xn setup request), that is, the first network device may carry the AI ​​event information set in the Xn link establishment request sent to the second network device. The implementation method of the event identifier and configuration parameters of at least one AI event in the AI ​​event information set can refer to the aforementioned related introduction, and the embodiments of the present application will not be repeated here.

[0149] After receiving the AI ​​event information set, the second network device may store the event identifiers and configuration parameters for each AI event in the AI ​​event information set, thereby completing the configuration of the event information for the AI ​​events. Subsequently, the second network device may send second indication information to the first network device, indicating whether the second network device has successfully or failed to configure the event information for some or all of the AI ​​events in the AI ​​event information set. Optionally, if there is an AI event for which configuration failed, the second indication information may also indicate the reason for the configuration failure.

[0150] For example, when the first network device uses an AI event configuration request to carry an AI event information set, the second network device may send an AI event configuration response to the first network device, where the AI ​​event configuration response carries the above-mentioned second indication information.

[0151] Among them, when the AI ​​event configuration request is a request specifically for transmitting an AI event information set, the AI ​​event configuration response is also a response specifically for the AI ​​event configuration request. When the AI ​​event configuration request reuses other requests between the first network device and the second network device, the AI ​​event configuration response is a response to the corresponding request. For example, when the AI ​​event configuration request is an Xn link establishment request, the AI ​​event configuration response can be an Xn link establishment response (Xn setup response).

[0152] Optionally, when storing the AI ​​event information set, the second network device may store the device identifier of the first network device in correspondence with the AI ​​event information set, so as to distinguish AI event information sets from different network devices.

[0153] Optionally, when sending an AI data request to the second network device, the first network device may further send third indication information to the second network device, where the third indication information is used to indicate the AI ​​data item requested by the first network device. The AI ​​data item requested by the first network device may include one or more of the predicted value, current measurement value, and other relevant data of the measurement item included in the first AI event, and may also include other data of the second network device, which is not limited in this embodiment of the present application.

[0154] It should be noted that the AI ​​data item requested by the first network device can be associated with the AI ​​use case executed by the first network device. For example, when the first network device performs load balancing based on an AI model, the AI ​​data item requested by the first network device can be AI data related to the device load predicted or currently measured by the second network device.

[0155] In addition, the third indication information may be carried by other messages different from the AI ​​data request, or may be carried by the AI ​​data request, and the third indication information and the AI ​​data request may be sent in any order.

[0156] In addition, the AI ​​data request can be a message specifically used to request AI data, or it can reuse other requests between the first network device and the second network device. For example, the AI ​​data request can be a handover request sent by the first network device to the second network device. The first network device can include event information of the first AI event in the handover request sent to the second network device.

[0157] Step 402: The second network device sends AI data to the first network device based on the event information of the first AI event.

[0158] After receiving the AI ​​data request sent by the first network device, the second network device may send AI data to the first network device based on the event information of the first AI event carried in the AI ​​data request.

[0159] In one possible implementation, when the event information of the first AI event carried in the AI ​​data request is the event identifier and configuration parameters of the first AI event, the second network device can send AI data to the first network device based on the event identifier and configuration parameters of the first AI event.

[0160] Among them, it can be seen from the above introduction that when the first AI event includes a prediction-type trigger event or a current measurement-type trigger event, the configuration parameters of the first AI event will include the measurement item and / or the set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the trigger duration. In this case, after receiving the event identifier and configuration parameters of the first AI event, the second network device can detect the first AI event based on the event identifier of the first AI event, the indication information of the measurement item and / or the set threshold corresponding to the measurement item, and send AI data to the first network device if the first AI event is continuously detected within the trigger duration.

[0161] As an example, when a first AI event includes a prediction trigger event in which the predicted value of the first measurement item is higher than a corresponding first set threshold, the second network device may detect whether the predicted value of the first measurement item is higher than the corresponding first set threshold based on the event identifier of the first AI event. If the predicted value of the first measurement item is higher than the first set threshold and the predicted value of the first measurement item remains higher than the first set threshold within the trigger duration, the second network device may send AI data to the first network device. The predicted value of the first measurement item refers to the value predicted by the second network device using the AI ​​model.

[0162] Optionally, if the configuration parameters of the first AI event also include a hysteresis parameter and prediction accuracy, the second network device may detect whether the predicted value of the first measurement item meets the prediction accuracy based on the event identifier of the first AI event. If the predicted value meets the prediction accuracy, the second network device detects whether the predicted value of the first measurement item is within the hysteresis range based on the hysteresis parameter and a first set threshold corresponding to the first measurement item. If it is detected that the predicted value of the first measurement item remains within the hysteresis range within the trigger duration, the AI ​​data is sent to the first network device.

[0163] The second network device may return all of its own AI data to the first network device. Alternatively, if the first network device sends third indication information indicating the data item requested by itself to the second network device, the second network device may return the corresponding AI data based on the third indication information.

[0164] For example, when the first AI event includes a predicted value of the device load that is higher than the corresponding first load threshold, and the configuration parameters of the AI ​​event are as shown in step 401, the second network device can detect that the predicted value of one or more of the throughput, PRB utilization, number of activated terminals, number of RRC connections, comprehensive available capacity group, TNL capacity index, and slice available capacity meets the prediction accuracy and is within the corresponding hysteresis range within the triggering time, and then return AI data to the first network device under the instruction of the second indication information. For example, when the first network device performs load balancing based on the AI ​​model, then under the instruction of the third indication information, the second network device can return the predicted value of each of the above-mentioned sub-load items predicted at the current moment, or the current measured value of each of the above-mentioned sub-load items, so that the first network device can train, infer or optimize the AI ​​model based on these AI data.

[0165] As another example, when the first AI event includes a second measurement item whose current measurement value is different from the predicted value before the current moment, the second network device can compare the current measurement value of the second measurement item with the predicted value before the current moment based on the event identifier of the first AI event to see if they are the same. If the current measurement value of the second measurement item is inconsistent with the predicted value before the current moment within the trigger duration, the second network device returns AI data to the first network device. The implementation method of returning AI data to the first network device can refer to the above example.

[0166] For example, when the first AI event includes that the current measured value of the terminal's service is different from the predicted value before the current moment, and the configuration parameters of the AI ​​event are as shown in the aforementioned step 401, the second network device can detect whether the current measured value of the service of the terminal identified by at least one terminal identifier in the terminal identification list is consistent with the predicted value before the current moment, that is, detect whether the service of the at least one terminal at the current moment is consistent with the service of the at least one terminal predicted before the current moment. If the second network device detects that the two are always inconsistent within the triggering time, the second network device can return the AI ​​data associated with the at least one terminal to the first network device. For example, the AI ​​data may include the service type, throughput, packet loss rate, and transmission delay currently executed by the terminal.

[0167] As another example, when the first AI event includes a current action trigger event, for example, when the first AI event includes cell activation, and the configuration parameters of the first AI event include a cell identifier list and a trigger duration, the second network device can detect the on / off state of the target cell based on the event identifier of the first AI event, where the cell identifier of the target cell is any cell identifier in the cell identifier list; and when it is detected that the target cell is in the on state for a continuous period of time, the second network device sends AI data to the first network device. The AI ​​data can be data related to the target cell measured by the second network device, and the AI ​​data can be used by the first network device to implement network energy saving based on the AI ​​model.

[0168] For another example, when the first AI event includes a successful terminal handover, and the configuration parameters of the AI ​​event include a terminal identifier list and a trigger duration, the second network device can detect the resident status of the target terminal based on the event identifier of the first AI event, where the identifier of the target terminal is any identifier in the terminal identifier list, and the target terminal is a terminal that has switched to the second network device; if it is detected that the target terminal continues to reside on the second network device within the trigger duration, AI data is sent to the first network device. The AI ​​data can be data associated with the target terminal, such as the handover parameters of the target terminal. In addition, the AI ​​data can be used by the first network device to implement mobility optimization based on the AI ​​model.

[0169] For another example, when the first AI event includes executing an AI use case, and the configuration parameters of the AI ​​event include the AI ​​use case and trigger duration introduced in the aforementioned step 401, the second network device can detect the execution of the AI ​​use case by the second network device based on the event identifier of the first AI event; and when it is detected that the second network device executes any of the above-mentioned AI use cases within the trigger duration, send AI data to the first network device. The AI ​​data can be data related to the AI ​​use case executed by the second network device. For example, if the second network device performs load balancing within the trigger duration, the second network device can return data related to load balancing to the first network device, such as the values ​​of different sub-load items predicted or currently measured by the second network device.

[0170] In another possible implementation, if the event information of the first AI event carried in the AI ​​data request includes an event identifier of the first AI event, the second network device may obtain configuration parameters for the first AI event based on the event identifier of the first AI event. Thereafter, the second network device may send the AI ​​data to the first network device based on the event identifier and the configuration parameters of the first AI event.

[0171] When the AI ​​data request carries the event identifier of the first AI event but does not carry configuration parameters, the first network device will pre-send an AI event information set including the event identifier of the first AI event and the configuration parameters to the second network device, instructing the second network device to store the AI ​​event information set. Based on this, after receiving the AI ​​data request, the second network device can search and obtain the configuration parameters for the first AI event from the stored AI event information set based on the event identifier of the first AI event.

[0172] After obtaining the configuration parameters of the first AI event, the second network device may send AI data to the first network device based on the event identifier of the first AI event and the configuration parameters. For related implementation methods, please refer to the relevant introduction in the aforementioned step 402.

[0173] Optionally, in this implementation, the first network device sends an AI event information set to the second network device in advance to configure the AI ​​event. Subsequently, the first network device may also update the configuration parameters of the AI ​​event in the pre-configured AI event information set.

[0174] Exemplarily, a first network device may send a configuration update request to a second network device. The configuration update request includes an event identifier and update information for a first AI event. The configuration update request is used to request the second network device to update the configuration parameters of the first AI event. After receiving the configuration update request, the second network device updates the configuration parameters of the first AI event based on the event identifier and update information of the first AI event carried in the configuration update request. Thereafter, the second network device sends a configuration update response to the first network device. The configuration update response is used to indicate the success or failure of the configuration parameter update for the first AI event. If the configuration parameter update for the first AI event fails, the configuration update response may also indicate the reason for the update failure.

[0175] The update information carried in the configuration update request may include at least one configuration parameter and an update operation corresponding to the at least one configuration parameter, and the update operation includes any one of modification, addition and deletion.

[0176] Correspondingly, the second network device can search for the configuration parameters of the first AI event stored in itself according to the event identifier of the first AI event, and then perform corresponding update operations on the configuration parameters of the first AI event stored in itself according to the configuration parameters carried in the update information and the corresponding update operations.

[0177] For example, when the predicted value of the device load of the first AI event is higher than the corresponding first set threshold, the first network device wants to modify the throughput threshold in the configuration parameters of the first AI event, then the information carried in the configuration update request can be as shown in the example in Table 7 below.

[0178] Table 7 Information in a configuration update request

[0179] The throughput threshold shown in Table 7 refers to the changed throughput threshold determined by the first network device. After receiving the configuration update request, the second network device may modify its stored throughput threshold to the throughput threshold in Table 7 above.

[0180] For another example, when the predicted value of the device load of the first AI event is higher than the corresponding first set threshold, the configuration parameters of the first AI event stored by the second network device do not include the activation terminal number threshold, and the first network device wants to add the activation terminal number threshold to the configuration parameters of the first AI event, then the information carried in the configuration update request can be as shown in the example in Table 8 below.

[0181] Table 8 Information in the configuration update request

[0182] Among them, the activation terminal number threshold in Table 8 above is the threshold item to be added by the first network device. In this case, the second network device can add the activation terminal number threshold carried in the configuration update request to the configuration parameters of the first AI event stored in itself.

[0183] For another example, when the predicted value of the device load of the first AI event is higher than the corresponding first set threshold, the first network device wants to delete the RRC connection number threshold in the configuration parameters of the first AI event stored by the second network device, then the information carried in the configuration update request can be as shown in the example of Table 9 below.

[0184] Table 9 Information in the configuration update request

[0185] Among them, the RRC connection number threshold in Table 9 above is the threshold item to be deleted by the first network device. In this case, the second network device can delete the RRC connection number threshold in the configuration parameters of the first AI event stored in itself.

[0186] Exemplarily, the configuration update request may further carry the device identification of the first network device, thereby instructing the second network device to update the configuration parameters of the first AI event pre-sent by the first network device.

[0187] It can be understood that the examples in Tables 7 to 9 above are merely possible implementations of the cell structure when the update information of the first AI event is transmitted using a configuration update request, and do not constitute any limitation on the transmission method or cell structure of the update information of the first AI event.

[0188] It should be noted that the configuration update request and configuration update response can be signaling specifically used to update the configuration parameters of the AI ​​event. Optionally, the configuration update request and configuration update response can also reuse other requests and corresponding responses between the first network device and the second network device. For example, the configuration update request can be an Xn link establishment request, and the configuration update response can be an Xn link establishment response. Alternatively, the configuration update request can be a handover request, and the configuration update response can be a handover request response.

[0189] The above description uses the example of updating the configuration parameters of the first AI event. Obviously, the first network device can also update the configuration parameters of the corresponding AI event pre-configured on the second network device by carrying the event identifier and corresponding update information of other AI events in the configuration update request. No further details will be given.

[0190] It should be noted that, when the AI ​​data request directly carries the event identifier and configuration parameters of the first AI event, the configuration parameters of the AI ​​event on the second network device can also be updated using the method described above. For example, after the first network device receives the AI ​​data returned by the second network device based on the AI ​​data request, it can also send a configuration update request to the second network device to update the configuration parameters of the AI ​​event previously sent to the second network device. This will not be further described.

[0191] Using the data transmission method provided in this application, when a first network device sends an AI data request to a second network device, the first network device can carry event information of a first AI event in the AI ​​data request. The first AI event is an event that triggers the second network device to send AI data. In this way, based on the event information of the first AI event, the second network device can send AI data to the first network device when it detects that the first AI event is met. Because the second network device can promptly return the AI ​​data required by the first network device when triggered by a trigger event specified by the first network device, it meets the diverse data needs of the first network device and can improve the effectiveness of the first network device in training, reasoning, or optimizing AI models.

[0192] Based on the above description, when the AI ​​data request carries the event identifier and configuration parameters of the first AI event, as an example, as shown in FIG5 , the data transmission method may include the following steps:

[0193] Step 501: A first network device sends an AI data request to a second network device. The AI ​​data request includes an event identifier and configuration parameters of a first AI event.

[0194] Step 502: The second network device sends an AI data response to the first network device. The AI ​​data response includes first indication information. The first indication information is used to indicate whether the second network device has successfully configured the event information of the first AI event or failed.

[0195] Step 503: The second network device sends AI data to the first network device based on the event identifier and configuration parameters of the first AI event.

[0196] The implementation of the above steps 501 to 503 can refer to the implementation of the relevant steps in the embodiment shown in FIG4 , and will not be described in detail.

[0197] When adopting the data transmission method shown in Figure 5, the first network device can directly carry the event identifier and configuration parameters of the first AI event in the AI ​​data request. In this way, the configuration of the event information of the AI ​​event and the request for AI data can be realized at the same time, reducing the signaling overhead between network devices.

[0198] When the AI ​​data request carries the event identifier of the first AI event but does not carry the configuration parameters of the first AI event, as an example, as shown in FIG6 , the data transmission method may include the following steps:

[0199] Step 601: A first network device sends an AI event configuration request to a second network device, where the AI ​​event configuration request includes an AI event information set.

[0200] Step 602: The second network device stores the AI ​​event information set.

[0201] Step 603: The second network device sends an AI event configuration response to the first network device. The AI ​​event configuration response includes second indication information. The second indication information is used to indicate whether the configuration of storing event information of the AI ​​event is successful or failed.

[0202] Step 604: The first network device sends an AI data request to the second network device, where the AI ​​data request includes an event identifier of the first AI event.

[0203] Step 605: The second network device obtains configuration parameters of the first AI event based on the event identifier of the first AI event.

[0204] Step 606: The second network device sends AI data to the first network device based on the event identifier and configuration parameters of the first AI event.

[0205] Step 607: The first network device sends a configuration update request to the second network device. The configuration update request includes the event identifier of the first AI event and update information.

[0206] Step 608: The second network device updates the configuration parameters of the first AI event based on the event identifier and update information of the first AI event.

[0207] Step 609: The second network device sends a configuration update response to the first network device. The configuration update response is used to indicate whether the configuration parameters of the first AI event are updated successfully or failed.

[0208] The implementation of the above steps 601 to 609 can refer to the relevant introduction in the embodiment shown in Figure 4 and will not be repeated here.

[0209] When the data transmission method provided in Figure 6 is adopted, the first network device can pre-configure the event information of the AI ​​event on the second network device through steps 601 to 603. Subsequently, when the first network device needs to obtain AI data, the event identifier of the AI ​​event can be carried in the AI ​​data request to instruct the second network device to return the AI ​​data when the corresponding AI event is met, thereby reducing signaling overhead.

[0210] In addition, the first network device and the second network device can update the configuration parameters of the AI ​​event pre-configured by the first network device on the second network device through the above steps 607 to 609, thereby reducing errors or unnecessary data transmission caused by changes in the configuration parameters of the AI ​​event.

[0211] Next, the data transmission device provided in the embodiment of the present application is introduced.

[0212] FIG7 is a schematic diagram of the structure of a data transmission device 700 provided in an embodiment of the present application, which can be deployed in the first network device of the aforementioned embodiment. As shown in FIG7 , the data transmission device includes a sending module 701 and a receiving module 702 .

[0213] The sending module 701 is used to execute step 401 in the above embodiment. In some examples, the sending module is used to execute step 501 or step 604 in the above embodiment.

[0214] The receiving module 702 is configured to receive the AI ​​data returned by the second network device.

[0215] Exemplarily, the event information of the first AI event includes an event identifier and configuration parameters of the first AI event.

[0216] Exemplarily, the event information of the first AI event includes an event identifier of the first AI event.

[0217] Exemplarily, the sending module 701 is further configured to:

[0218] An AI event information set is sent to the second network device to instruct the second network device to store the AI ​​event information set, where the AI ​​event information set includes configuration parameters and an event identifier of at least one AI event, and the at least one AI event includes the first AI event.

[0219] Exemplarily, the sending module 701 is also used to send a configuration update request to the second network device, the configuration update request includes the event identifier and update information of the first AI event, and the configuration update request is used to request the second network device to update the configuration parameters of the first AI event; the receiving module 702 is also used to receive a configuration update response sent by the second network device in response to the configuration update request, and the configuration update response is used to indicate whether the second network device has successfully or failed to update the configuration parameters of the first AI event.

[0220] Exemplarily, the update information includes at least one configuration parameter and an update operation corresponding to the at least one configuration parameter, and the update operation includes one of modification, addition, and deletion.

[0221] Exemplarily, the first AI event includes a prediction-type trigger event, and the prediction-type trigger event includes a trigger event based on a prediction result of an AI model of the second network device.

[0222] Exemplarily, the prediction trigger event includes at least one of a predicted value of the first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, and the first set threshold is greater than the second set threshold.

[0223] Exemplarily, the first AI event includes a current measurement type trigger event, and the current measurement type trigger event includes a trigger event based on a current measurement result of the second network device.

[0224] Exemplarily, the current measurement type trigger event includes at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

[0225] Exemplarily, the first measurement item includes any one of device load, device energy consumption and terminal traffic, the second measurement item includes the terminal's mobile path or the terminal's business, and the third measurement item includes any one of device load, device energy consumption and terminal traffic.

[0226] Exemplarily, when the first AI event includes a prediction-type trigger event or a current measurement-type trigger event, the configuration parameters of the first AI event include indication information of the measurement item and / or a set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the duration of the first AI event.

[0227] Exemplarily, the first AI event includes a current action type trigger event, and the current action type trigger event includes a trigger event based on a current action of the second network device.

[0228] Exemplarily, the current action-type triggering event includes at least one of cell activation, successful terminal switching, terminal being in switching, and executing an AI use case.

[0229] Exemplarily, when the first AI event includes cell activation, the configuration parameters of the first AI event include a cell identification list and the duration of the first AI event; when the first AI event includes at least one of a successful terminal handover and a terminal in handover, the configuration parameters of the first AI event include a terminal identification list and the duration; when the first AI event includes executing an AI use case, the configuration parameters of the first AI event include the AI ​​use case and the duration.

[0230] Using the data transmission device provided in this application, when a first network device sends an AI data request to a second network device, the first network device can carry event information of a first AI event in the AI ​​data request, where the first AI event is an event that triggers the second network device to send AI data. In this way, based on the event information of the first AI event, the second network device can send AI data to the first network device when it detects that the first AI event is satisfied. Since the second network device can promptly return the AI ​​data required by the first network device when triggered by a trigger event specified by the first network device, it meets the diverse data needs of the first network device and can improve the effectiveness of the first network device in training, reasoning, or optimizing AI models.

[0231] Figure 8 is a structural diagram of another data transmission device provided in an embodiment of the present application. The data transmission device can be deployed in the second network device in the aforementioned embodiment. As shown in Figure 8, the data transmission device 800 includes a receiving module 801 and a sending module 802.

[0232] The receiving module 801 is configured to receive an AI data request sent by a first network device, where the AI ​​data request includes event information of a first AI event, and the first AI event is configured to trigger the second network device to send AI data to the first network device.

[0233] The sending module 802 is used to execute step 402 in the above embodiment.

[0234] In some examples, the sending module 802 is used to execute steps 502 and 503 in the aforementioned embodiments.

[0235] In other examples, the sending module 802 is used to execute step 605 and step 606 in the aforementioned embodiment.

[0236] In this case, referring to FIG8 , the apparatus 800 may further include a storage module 803. The sending module 802 is further configured to receive an AI event information set sent by the first network device, the AI ​​event information set including configuration parameters and an event identifier of at least one AI event, the at least one AI event including a first AI event; and the storage module 803 is configured to store the AI ​​event information set.

[0237] Exemplarily, referring to FIG8 , the apparatus 800 may further include an updating module 804. The receiving module 801 is further configured to receive a configuration update request sent by the first network device, the configuration update request including an event identifier and update information of the first AI event; the updating module 804 is configured to update the configuration parameters of the first AI event based on the event identifier and update information of the first AI event; and the sending module 802 is further configured to send a configuration update response to the first network device, the configuration update response being used to indicate success or failure of the configuration parameter update for the first AI event.

[0238] Exemplarily, the update information includes at least one configuration parameter and an update operation corresponding to the at least one configuration parameter, and the update operation includes one of modification, addition, and deletion.

[0239] Exemplarily, the first AI event includes a prediction-type trigger event, and the prediction-type trigger event includes a trigger event based on a prediction result of an AI model of the second network device.

[0240] Exemplarily, the prediction trigger event includes at least one of a predicted value of the first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, and the first set threshold is greater than the second set threshold.

[0241] Exemplarily, the first AI event includes a current measurement type trigger event, and the current measurement type trigger event includes a trigger event based on a current measurement result of the second network device.

[0242] Exemplarily, the current measurement type trigger event includes at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

[0243] Exemplarily, the first measurement item includes any one of device load, device energy consumption and terminal traffic, the second measurement item includes the terminal's mobile path or the terminal's business, and the third measurement item includes any one of device load, device energy consumption and terminal traffic.

[0244] Exemplarily, when the first AI event includes a prediction-type trigger event or a current measurement-type trigger event, the configuration parameters of the first AI event include indication information of the measurement item and / or a set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the duration of the first AI event.

[0245] In this case, the sending module 802 is mainly used to: detect the first AI event based on the event identifier of the first AI event, the indication information of the measurement item and / or the set threshold corresponding to the measurement item; and send AI data to the first network device when the first AI event is continuously detected within the duration.

[0246] Exemplarily, the first AI event includes a current action type trigger event, and the current action type trigger event includes a trigger event based on a current action of the second network device.

[0247] Exemplarily, the current action-type triggering event includes at least one of cell activation, successful terminal switching, terminal being in switching, and executing an AI use case.

[0248] Exemplarily, when the first AI event includes cell on, the configuration parameters of the first AI event include a cell identifier list and the duration of the first AI event. In this case, the sending module 802 is primarily configured to: detect the on / off state of the target cell based on the event identifier of the first AI event, where the cell identifier of the target cell is any cell identifier in the cell identifier list; and, upon detecting that the target cell is on for the duration, send AI data to the first network device.

[0249] Exemplarily, when the first AI event includes a successful terminal handover, the configuration parameters of the first AI event include a terminal identifier list and the duration of the first AI event. In this case, the sending module 802 is primarily configured to: detect the resident state of the target terminal based on the event identifier of the first AI event, where the identifier of the target terminal is any identifier in the terminal identifier list, and the target terminal is a terminal that has switched to the second network device; and upon detecting that the target terminal remains resident on the second network device for the duration, send AI data to the first network device.

[0250] Exemplarily, when the first AI event includes executing an AI use case, the configuration parameters of the first AI event include the AI ​​use case and the duration of the first AI event. In this case, the sending module 802 is primarily configured to: detect, based on the event identifier of the first AI event, whether the second network device is executing the AI ​​use case; and, upon detecting that the second network device is executing the AI ​​use case within the duration, send AI data to the first network device.

[0251] Using the data transmission device provided in this application, the second network device receives an AI data request sent by the first network device, and the AI ​​data request carries event information of the first AI event. Based on the event information of the first AI event, the second network device can send AI data to the first network device when it detects that the first AI event is satisfied. Because the second network device can promptly return the AI ​​data required by the first network device when triggered by a trigger event specified by the first network device, it meets the diverse data needs of the first network device and can improve the effectiveness of the first network device in training, reasoning, or optimizing AI models.

[0252] It should be noted that the division of modules in the various data transmission devices provided in the above embodiments is illustrative and is merely a logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the present application may be integrated into a single processor, or may exist physically separately, or two or more modules may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or software functional modules.

[0253] If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for making a network device (which can be a router or switch, etc.) or a processor (processor) execute all or part of the steps of the method of each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes. For example, the storage medium can be a computer-readable storage medium, which is any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a digital versatile disc (DVD)), or a semiconductor medium (for example, a solid state disk (SSD)). For another example, the storage medium can be a read-only memory (ROM) or a random access memory (RAM). In addition, the data transmission device and the data transmission method provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0254] The present application also provides a data transmission system including a first network device and a second network device. The first network device can be used to execute the steps executed by the first network device in the method embodiment shown in FIG. 4 , FIG. 5 , or FIG. 6 , and the second network device can be used to execute the steps executed by the second network device in the method embodiment shown in FIG. 4 , FIG. 5 , or FIG. 6 , which are not described in detail here.

[0255] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0256] In the various embodiments of the present application, unless otherwise specified or logically conflicting, the terms and / or descriptions between different embodiments are consistent and can be referenced to each other. The technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships. In the present application, "at least one" means one or more, and "more" means two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. In the textual description of the present application, the character " / " generally indicates that the related objects before and after are in an "or" relationship. In the present application, "first", "second" and various numerical numbers indicate distinctions made for the convenience of description and are not used to limit the scope of the embodiments of the present application. For example, to distinguish different messages, etc., rather than to describe a specific order or sequence.

[0257] It is understood that the various numbers used in the embodiments of this application are merely for ease of description and are not intended to limit the scope of the embodiments of this application. The order of the sequence numbers of the above-mentioned processes does not necessarily imply a specific order of execution; the order of execution of the processes should be determined by their functions and inherent logic.

Claims

1. A data transmission method, characterized in that: The method is used for a first network device, and the method comprises: Sending an AI data request to a second network device, where the AI ​​data request includes event information of a first AI event, where the first AI event is used to trigger the second network device to send AI data to the first network device; The AI ​​data is received.

2. The method according to claim 1, characterized in that: The event information of the first AI event includes an event identifier and configuration parameters of the first AI event.

3. The method according to claim 1, characterized in that The event information of the first AI event includes an event identifier of the first AI event.

4. The method according to claim 3, characterized in that Before sending the AI ​​data request to the second network device, the method further includes: An AI event information set is sent to the second network device to instruct the second network device to store the AI ​​event information set, wherein the AI ​​event information set includes configuration parameters and an event identifier of at least one AI event, and the at least one AI event includes the first AI event.

5. The method according to any one of claims 1 to 4, characterized in that: The method further comprises: Sending a configuration update request to the second network device, where the configuration update request includes an event identifier and update information of the first AI event, and the configuration update request is used to request the second network device to update configuration parameters of the first AI event; A configuration update response is received from the second network device in response to the configuration update request, where the configuration update response is used to indicate whether the second network device succeeds or fails in updating the configuration parameters of the first AI event.

6. The method according to claim 5, characterized in that The update information includes at least one configuration parameter and an update operation corresponding to the at least one configuration parameter, and the update operation includes one of modification, addition and deletion.

7. The method according to any one of claims 2, 4 to 6, characterized in that: The first AI event includes a prediction type trigger event, and the prediction type trigger event includes a trigger event based on a prediction result of an AI model of the second network device.

8. The method according to claim 7, characterized in that The prediction trigger event includes at least one of a predicted value of a first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, and the first set threshold is greater than the second set threshold.

9. The method according to any one of claims 2, 4 to 6, characterized in that: The first AI event includes a current measurement type trigger event, and the current measurement type trigger event includes a trigger event based on a current measurement result of the second network device.

10. The method according to claim 9, characterized in that The current measurement type trigger event includes at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

11. The method according to claim 10, characterized in that The first measurement item includes any one of device load, device energy consumption and terminal traffic, the second measurement item includes the terminal's moving path or the terminal's service, and the third measurement item includes any one of device load, device energy consumption and terminal traffic.

12. The method according to any one of claims 8, 10 and 11, characterized in that: When the first AI event includes the prediction type trigger event or the current measurement type trigger event, the configuration parameters of the first AI event include indication information of the measurement item and / or the set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the duration of the first AI event.

13. The method according to any one of claims 2, 4 to 6, characterized in that: The first AI event includes a current action type trigger event, and the current action type trigger event includes a trigger event based on a current action of the second network device.

14. The method according to claim 13, characterized in that The current action trigger event includes at least one of cell activation, terminal switching success, terminal switching in progress, and execution of an AI use case.

15. The method according to claim 14, characterized in that When the first AI event includes cell activation, the configuration parameters of the first AI event include a cell identifier list and a duration of the first AI event; When the first AI event includes at least one of a successful terminal handover and a terminal being in handover, the configuration parameters of the first AI event include a terminal identification list and the duration; When the first AI event includes executing an AI use case, the configuration parameters of the first AI event include the AI ​​use case and the duration.

16. A data transmission method, characterized in that: The method is used for a second network device, and the method comprises: receiving an AI data request sent by a first network device, where the AI ​​data request includes event information of a first AI event, where the first AI event is used to trigger the second network device to send AI data to the first network device; Based on the event information of the first AI event, the AI ​​data is sent to the first network device.

17. The method according to claim 16, characterized in that The event information of the first AI event includes an event identifier and configuration parameters of the first AI event; The second network device sending the AI ​​data to the first network device based on the event information of the first AI event includes: The second network device sends the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event.

18. The method according to claim 16, characterized in that The event information of the first AI event includes an event identifier of the first AI event; The sending the AI ​​data to the first network device based on the event information of the first AI event includes: Based on the event identifier of the first AI event, acquiring configuration parameters of the first AI event; Based on the event identifier of the first AI event and the configuration parameters, the AI ​​data is sent to the first network device.

19. The method according to claim 18, characterized in that Before receiving the AI ​​data request sent by the first network device, the method further includes: receiving an AI event information set sent by the first network device, the AI ​​event information set including a configuration parameter and an event identifier of at least one AI event, the at least one AI event including the first AI event; The AI ​​event information set is stored.

20. The method according to any one of claims 16 to 19, characterized in that: The method further comprises: Receive a configuration update request sent by the first network device, wherein the configuration update request includes an event of the first AI event Identify and update information; updating configuration parameters of the first AI event based on the event identifier and update information of the first AI event; A configuration update response is sent to the first network device, where the configuration update response is used to indicate success or failure of updating the configuration parameters of the first AI event.

21. The method according to claim 20, characterized in that The update information includes at least one configuration parameter and an update operation corresponding to the at least one configuration parameter, and the update operation includes one of modification, addition and deletion.

22. The method according to any one of claims 17 to 19, characterized in that: The first AI event includes a prediction type trigger event, and the prediction type trigger event includes a trigger event based on a prediction result of an AI model of the second network device.

23. The method according to claim 22, characterized in that The prediction trigger event includes at least one of a predicted value of a first measurement item being higher than a corresponding first set threshold and a predicted value of the first measurement item being lower than a corresponding second set threshold, and the first set threshold is greater than the second set threshold.

24. The method according to any one of claims 17 to 19, characterized in that: The first AI event includes a current measurement type trigger event, and the current measurement type trigger event includes a trigger event based on a current measurement result of the second network device.

25. The method according to claim 24, characterized in that The current measurement type trigger event includes at least one of the following: the current measurement value of the first measurement item is higher than the corresponding third set threshold, the current measurement value of the first measurement item is lower than the corresponding fourth set threshold, the current measurement value of the second measurement item is different from the predicted value before the current moment, the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is higher than the corresponding fifth set threshold, and the error ratio between the current measurement value of the third measurement item and the predicted value before the current moment is lower than the corresponding sixth set threshold, the third set threshold is greater than the fourth set threshold, and the fifth set threshold is greater than the sixth set threshold.

26. The method according to claim 25, characterized in that The first measurement item includes any one of device load, device energy consumption and terminal traffic, the second measurement item includes the terminal's moving path or the terminal's service, and the third measurement item includes any one of device load, device energy consumption and terminal traffic.

27. The method according to any one of claims 23, 25 and 26, characterized in that When the first AI event includes the prediction type trigger event or the current measurement type trigger event, the configuration parameters of the first AI event include indication information of the measurement item and / or the set threshold corresponding to the measurement item, and the configuration parameters of the first AI event also include the duration of the first AI event.

28. The method according to claim 27, characterized in that The sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event includes: detecting the first AI event based on an event identifier of the first AI event, the indication information of the measurement item, and / or a set threshold corresponding to the measurement item; In a case where the first AI event is continuously detected within the duration, the AI ​​data is sent to the first network device.

29. The method according to any one of claims 17 to 19, characterized in that: The first AI event includes a current action type trigger event, and the current action type trigger event includes a trigger event based on a current action of the second network device.

30. The method according to claim 29, characterized in that The current action trigger event includes at least one of cell activation, terminal switching success, terminal switching in progress, and execution of an AI use case.

31. The method according to claim 30, characterized in that When the first AI event includes cell activation, the configuration parameters of the first AI event include a cell identifier list and a duration of the first AI event; The sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event includes: Detecting a switch state of a target cell based on an event identifier of the first AI event, where the cell identifier of the target cell is any cell identifier in the cell identifier list; When it is detected that the target cell is in the turned-on state during the duration, the AI ​​data is sent to the first network device.

32. The method according to claim 30, characterized in that When the first AI event includes a successful terminal switching, the configuration parameters of the first AI event include a terminal identification list and a duration of the first AI event; The sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event includes: Detecting a resident state of a target terminal based on an event identifier of the first AI event, where the identifier of the target terminal is any identifier in the terminal identifier list, and the target terminal is a terminal switched to the second network device; When it is detected that the target terminal continuously resides on the second network device within the duration, the AI ​​data is sent to the first network device.

33. The method according to claim 30, characterized in that When the first AI event includes executing an AI use case, the configuration parameters of the first AI event include the AI ​​use case and a duration of the first AI event; The sending the AI ​​data to the first network device based on the event identifier and configuration parameters of the first AI event includes: Detecting, based on the event identifier of the first AI event, that the second network device executes the AI ​​use case; In case of detecting that the second network device executes the AI ​​use case within the duration, the AI ​​data is sent to the first network device.

34. A data transmission device, characterized in that: The data transmission device comprises at least one module, and the at least one module is used to execute the data transmission method according to any one of claims 1 to 16 or claims 17 to 33.

35. A data transmission device, characterized in that: The data transmission device includes a processor, which is used to execute at least one program instruction or code stored in a memory to implement the data transmission method described in any one of claims 1 to 16 or claims 17 to 33.

36. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the instructions are executed on a network device, the network device executes the data transmission method according to any one of claims 1 to 16 or claims 17 to 33.