Information indicating and processing method and device

CN119014034APending Publication Date: 2024-11-221FINITY INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202280094912.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2022-04-29
Filing Date
2022-11-07
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

Existing AI/ML models struggle to maintain consistent performance in the face of diverse wireless communication environments, especially in suburban areas, urban areas, indoor spaces, factories, mines, and other diverse wireless communication scenarios. Offline-trained models are unable to adapt to these changes, leading to unstable performance.

Method used

The first network device monitors the performance of the AI/ML model to determine whether to stop, update, or switch the model, and sends instruction information to the second network device, terminal device, or core network device to maintain the consistency and robustness of the model's operation.

Benefits of technology

It enables performance monitoring and management of AI/ML models, ensuring consistency in different environments and improving the robustness of model operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119014034A_ABST
    Figure CN119014034A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an information indicating and processing method and device. The method comprises the following steps: a first network device determines whether to stop, update or switch an AI / ML model or model group for a certain signal processing function; and under the condition that the AI / ML model or the model group is determined to be stopped or updated or switched, indication information is sent to second network equipment or terminal equipment or core network equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Information indication and processing method and device Technical Field

[0001] The embodiments of the present application relate to the field of communication technologies. Background Art

[0002] As low-frequency spectrum resources become scarce, millimeter-wave (mmWave) bands, offering greater bandwidth, have become a key frequency band for 5G New Radio (NR) systems. Due to their shorter wavelengths, mmWaves exhibit different propagation characteristics than traditional low-frequency bands, such as higher propagation loss and poor reflection and diffraction performance. Consequently, larger antenna arrays are typically used to form shaped beams with greater gain, overcome propagation loss, and ensure system coverage.

[0003] With the advancement of artificial intelligence (AI) and machine learning (ML) technologies, applying AI / ML to wireless communications to overcome the difficulties of traditional methods has become a current technological trend. The application of AI / ML models to wireless communication systems, particularly in air interface transmission, is a new technology in the 5G-Advanced and 6G phases.

[0004] For example, for Channel State Information (CSI) reporting, using an AI encoder to encode / compress CSI on the terminal device side and an AI decoder to decode / decompress CSI on the network device side can reduce feedback overhead. For another example, for beam management, using AI / ML models to predict the optimal spatial beam pair based on a small number of beam measurements can reduce system load and latency.

[0005] It should be noted that the above introduction to the technical background is merely intended to provide a clear and complete description of the technical solutions of this application and facilitate understanding by those skilled in the art. Simply because these solutions are described in the background technology section of this application, it should not be assumed that the above technical solutions are well known to those skilled in the art.

[0006] Summary of the Invention

[0007] However, the inventors discovered that AI / ML models trained on datasets face significant challenges in adapting to the demands of diverse wireless applications and coping with the ever-changing mobile communication environment. For a wide range of wireless communication scenarios, such as suburban areas, urban areas, indoor environments, factories, and mines, offline-trained AI / ML models struggle to maintain consistent performance across all scenarios. Therefore, it's necessary to monitor the performance of AI / ML models and, when necessary, disable them.

[0008] To address at least one of the above problems, an embodiment of the present application provides an information indication and processing method and device.

[0009] According to one aspect of an embodiment of the present application, there is provided an information indication method, including:

[0010] The first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and

[0011] When it is determined to stop, update or switch the AI / ML model or model group, an indication message is sent to the second network device, the terminal device or the core network device.

[0012] According to another aspect of an embodiment of the present application, there is provided an information indicating device, including:

[0013] a determination unit that determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and

[0014] A sending unit, which sends indication information to the second network device, the terminal device, or the core network device when it is determined to stop, update, or switch the AI / ML model or model group.

[0015] According to another aspect of an embodiment of the present application, there is provided an information processing method, including:

[0016] receiving instruction information sent by the first network device; and

[0017] Stop, update, or switch an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0018] According to another aspect of an embodiment of the present application, there is provided an information processing device, including:

[0019] a receiving unit configured to receive instruction information sent by the first network device; and

[0020] A processing unit that stops, updates, or switches an AI / ML model or model group for a certain signal processing function according to the indication information.

[0021] According to another aspect of an embodiment of the present application, a communication system is provided, including:

[0022] A network device that determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and when determining to stop, update, or switch the AI / ML model or model group, sends an indication message to a second network device, a terminal device, or a core network device.

[0023] One of the beneficial effects of the embodiments of the present application is that: a first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and when it is determined to stop, update, or switch the AI / ML model or model group, sends an instruction to a second network device, a terminal device, or a core network device. This allows monitoring the performance of the AI / ML model operation, maintaining the consistency of the AI / ML model operation, and improving the robustness of the model operation.

[0024] With reference to the following description and accompanying drawings, specific embodiments of the present application are disclosed in detail, indicating the manner in which the principles of the present application can be employed. It should be understood that the embodiments of the present application are not limited in scope. Within the spirit and scope of the appended claims, the embodiments of the present application include many variations, modifications and equivalents.

[0025] Features described and / or illustrated with respect to one embodiment may be used in the same or similar manner in one or more other embodiments, combined with features in other embodiments, or substituted for features in other embodiments.

[0026] It should be emphasized that the term "include / comprising" when used herein refers to the presence of features, integers, steps or components, but does not exclude the presence or addition of one or more other features, integers, steps or components. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The elements and features described in one figure or one embodiment of the present application can be combined with the elements and features shown in one or more other figures or embodiments. In addition, in the accompanying drawings, similar reference numerals represent corresponding parts in several figures and can be used to indicate corresponding parts used in more than one embodiment.

[0028] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application;

[0029] FIG2 is a schematic diagram of an information indication method according to an embodiment of the present application;

[0030] FIG3 is a schematic diagram of interaction between network devices according to an embodiment of the present application;

[0031] FIG4 is a schematic diagram of the interaction between the network device and the core network device according to an embodiment of the present application;

[0032] FIG5 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0033] FIG6 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0034] FIG7 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0035] FIG8 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0036] FIG9 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0037] FIG10 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0038] FIG11 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application;

[0039] FIG12 is a schematic diagram of an information processing method according to an embodiment of the present application;

[0040] FIG13 is a schematic diagram of an information processing method according to an embodiment of the present application;

[0041] FIG14 is a schematic diagram of an information processing method according to an embodiment of the present application;

[0042] FIG15 is a schematic diagram of an information indicating device according to an embodiment of the present application;

[0043] FIG16 is a schematic diagram of an information processing device according to an embodiment of the present application;

[0044] FIG17 is a schematic diagram of the structure of a network device or a core network device according to an embodiment of the present application;

[0045] FIG18 is a schematic diagram of a terminal device according to an embodiment of the present application. DETAILED DESCRIPTION

[0046] The above and other features of the present application will become apparent through the following description with reference to the accompanying drawings. In the description and the accompanying drawings, specific embodiments of the present application are disclosed in detail, which illustrate some embodiments in which the principles of the present application can be adopted. It should be understood that the present application is not limited to the described embodiments. On the contrary, the present application includes all modifications, variations and equivalents that fall within the scope of the appended claims.

[0047] In the embodiments of the present application, the terms "first", "second", etc. are used to distinguish different elements from the name, but do not indicate the spatial arrangement or temporal order of these elements, and these elements should not be limited by these terms. The term "and / or" includes any one and all combinations of one or more of the associated listed terms. The terms "comprising", "including", "having", etc. refer to the presence of the stated features, elements, components or components, but do not exclude the presence or addition of one or more other features, elements, components or components.

[0048] In the embodiments of this application, the singular forms "a," "the," etc. include plural forms and should be broadly understood to mean "a" or "a type" rather than being limited to "one." Furthermore, the term "said" should be understood to include both singular and plural forms, unless the context clearly indicates otherwise. Furthermore, the term "according to" should be understood to mean "at least in part based on...", and the term "based on" should be understood to mean "at least in part based on...", unless the context clearly indicates otherwise.

[0049] In the embodiments of the present application, the term "communication network" or "wireless communication network" may refer to a network that complies with any of the following communication standards, such as Long Term Evolution (LTE), enhanced Long Term Evolution (LTE-A, LTE-Advanced), Wideband Code Division Multiple Access (WCDMA), High-Speed ​​Packet Access (HSPA), etc.

[0050] Furthermore, communication between devices in the communication system may be carried out according to communication protocols of any stage, for example, including but not limited to the following communication protocols: 1G (generation), 2G, 2.5G, 2.75G, 3G, 4G, 4.5G and 5G, New Radio (NR), future 6G, etc., and / or other communication protocols currently known or to be developed in the future.

[0051] In the embodiments of the present application, the term "network device" refers to, for example, a device in a communication system that connects a terminal device to the communication network and provides services to the terminal device. Network devices may include, but are not limited to, the following devices: base station (BS), access point (AP), transmission reception point (TRP), broadcast transmitter, mobile management entity (MME), gateway, server, radio network controller (RNC), base station controller (BSC), etc.

[0052] Base stations may include, but are not limited to, NodeB (NB), evolved NodeB (eNodeB or eNB), and 5G base stations (gNB), among others. They may also include remote radio heads (RRHs), remote radio units (RRUs), relays, or low-power nodes (e.g., femeto, pico, etc.). The term "base station" may include some or all of their functions, and each base station may provide communication coverage for a specific geographic area. The term "cell" may refer to a base station and / or its coverage area, depending on the context in which the term is used.

[0053] In the embodiments of the present application, the term "user equipment" (UE) or "terminal equipment" (TE) refers to, for example, a device that accesses a communication network through a network device and receives network services. A terminal device can be fixed or mobile and may also be referred to as a mobile station (MS), a terminal, a subscriber station (SS), an access terminal (AT), a station, and so on.

[0054] Among them, terminal devices may include but are not limited to the following devices: cellular phones, personal digital assistants (PDAs), wireless modems, wireless communication devices, handheld devices, machine-type communication devices, laptop computers, cordless phones, smart phones, smart watches, digital cameras, etc.

[0055] For another example, in scenarios such as the Internet of Things (IoT), the terminal device can also be a machine or device for monitoring or measurement, including but not limited to: machine type communication (MTC) terminal, vehicle-mounted communication terminal, device-to-device (D2D) terminal, machine-to-machine (M2M) terminal, and so on.

[0056] In addition, the term "network side" or "network device side" refers to one side of the network, which can be a base station or one or more network devices as described above. The term "user side" or "terminal side" or "terminal device side" refers to the user or terminal side, which can be a UE or one or more terminal devices as described above. Unless otherwise specified herein, "device" can refer to either network equipment or terminal equipment.

[0057] The following describes the scenarios of the embodiments of the present application through examples, but the present application is not limited thereto.

[0058] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application, schematically illustrating a situation using a terminal device and a network device as an example. As shown in FIG1 , a communication system 100 may include a network device 101 and terminal devices 102 and 103. For simplicity, FIG1 illustrates only two terminal devices and one network device as an example, but the embodiments of the present application are not limited thereto.

[0059] In the embodiment of the present application, existing services or future services can be transmitted between the network device 101 and the terminal devices 102 and 103. For example, these services may include but are not limited to: enhanced mobile broadband (eMBB), massive machine type communication (mMTC), and ultra-reliable and low-latency communication (URLLC), etc.

[0060] It is worth noting that FIG1 shows that both terminal devices 102 and 103 are within the coverage range of network device 101, but the present application is not limited thereto. Both terminal devices 102 and 103 may not be within the coverage range of network device 101, or one terminal device 102 may be within the coverage range of network device 101 while the other terminal device 103 is outside the coverage range of network device 101.

[0061] In the embodiments of the present application, the high-layer signaling may be, for example, radio resource control (RRC) signaling; for example, an RRC message, including, for example, an MIB, system information, or a dedicated RRC message; or an RRC information element (RRC IE). The high-layer signaling may also be, for example, MAC (Medium Access Control) signaling; or a MAC control element (MAC CE). However, the present application is not limited thereto.

[0062] In embodiments of the present application, one or more AI / ML models may be configured and run in a network device and / or a terminal device. The AI / ML models may be used for various signal processing functions in wireless communications, such as CSI estimation and reporting, beam management, and beam prediction, etc.; the present application is not limited thereto.

[0063] Embodiments of the first aspect

[0064] An embodiment of the present application provides an information indication method, which is described from the perspective of a first network device.

[0065] FIG2 is a schematic diagram of an information indication method according to an embodiment of the present application. As shown in FIG2 , the method includes:

[0066] 201. The first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function.

[0067] 202. When it is determined to stop, update, or switch the AI / ML model or model group, an indication message is sent to a second network device, a terminal device, or a core network device.

[0068] It is worth noting that FIG2 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG2 above.

[0069] In some embodiments, the first network device can monitor its own AI / ML model, for example, by evaluating its performance based on the model's input and output data, and determining to stop the AI / ML model if performance is poor (e.g., performance results fall below a performance threshold). Furthermore, the first network device can obtain a model with improved performance through self-training (or online training) and determine to update its currently used AI / ML model.

[0070] In some embodiments, the first network device can monitor the AI / ML model of the second network device (within a certain area), for example, perform performance evaluation based on the signal from the second network device, and determine to stop the AI / ML model in the second network device if the performance is poor (the performance result is lower than the performance threshold). In addition, the first network device can also determine to update the AI / ML model in the area.

[0071] In some embodiments, the first network device may monitor the AI / ML model of the terminal device (within a certain cell), for example, perform performance evaluation based on a signal from the terminal device, and determine to stop the AI / ML model in the terminal device if the performance is poor (the performance result is lower than a performance threshold). In addition, the first network device may also determine to update the AI / ML model in the cell.

[0072] On the other hand, the first network device can monitor the AI / ML models of different signal processing functions separately. For example, the AI / ML model reported by CSI can be used to monitor the models of network devices and / or terminal devices in a certain cell or area; or the AI / ML model of beam management can be used to monitor the models of network devices and / or terminal devices in a certain cell or area. The AI / ML models of different signal processing functions can be monitored independently.

[0073] Thus, in response to stopping, updating, or switching the AI / ML model for a certain signal processing function, the first network device sends an instruction to the second network device, the terminal device, or the core network device. This allows monitoring the performance of the AI / ML model for a certain signal processing function within a cell or region, maintaining consistency in the AI / ML model's operation, and improving the robustness of the model's operation.

[0074] In the following description, unless otherwise specified, various embodiments are applicable to both "model" and "model group"; in addition, operations such as "stop", "update", and "switch" are used as examples for explanation. The above operations can be applied to various embodiments of this application, and can also be other AI / ML-related operations.

[0075] The following first describes the interaction between network devices.

[0076] FIG3 is a schematic diagram of interaction between network devices according to an embodiment of the present application. As shown in FIG3 , the method may include:

[0077] 301, the first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function;

[0078] 302. When it is determined to stop, update, or switch the AI / ML model or model group, the first network device sends first indication information for stopping the AI / ML model or model group and / or second indication information for updating the AI / ML model or model group to the second network device through the Xn interface.

[0079] It is worth noting that FIG3 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG3 above.

[0080] In some embodiments, the first indication information is carried by an XnAP message; for example, the XnAP message includes: a configuration update (NG-RAN node Configuration Update) message, an Xn establishment message, or a resource status update message; the present application is not limited thereto, and it may also be a new XnAP message.

[0081] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowed list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information; the present application is not limited to this, for example, it may also include other information.

[0082] For example, the first indication information includes {CSI; ID 1}, which indicates that the AI / ML model identified as ID 1 used for CSI reporting is stopped; for another example, the first indication information includes {CSI}, which indicates that all AI / ML models used for CSI reporting are stopped; for another example, the first indication information includes {CSI; ID 1; V2.0}, which indicates that the AI / ML model identified as ID 1 and version 2.0 used for CSI reporting is stopped.

[0083] The above only exemplifies the first indication information, but the present application is not limited thereto. For example, the first indication information may also be indicated by a predefined index. In addition, the first indication information may be explicitly indicated by a field or information element (IE) included in the message, or may be implicitly indicated by associating it with the message.

[0084] In some embodiments, the first indication information is used by the second network device to determine whether to stop the AI / ML model in the second network device, and / or to determine whether to stop the AI / ML model in a terminal device within a cell.

[0085] For example, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the second network device, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device determines to stop the AI / ML model in the second network device.

[0086] For example, after the second network device receives the first indication information {CSI; ID 1}, if it is running the AI / ML model identified as ID 1 for CSI reporting, it stops the AI / ML model identified as ID 1 for CSI reporting.

[0087] For another example, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device notifies the terminal device to stop the AI / ML model in the terminal device.

[0088] For example, after the second network device receives the first indication information {CSI; ID 1}, if it knows that one or more terminal devices in a certain cell are running the AI / ML model identified as ID 1 for CSI reporting, it sends a notification message to the terminal device in the cell, notifying the terminal device to stop using the AI / ML model identified as ID 1 for CSI reporting.

[0089] In some embodiments, the first network device receives a confirmation message, a rejection message, or a failure message sent by the second network device.

[0090] For example, after successfully stopping the AI / ML model identified as ID 1 for CSI reporting, the second network device feeds back a confirmation message to the first network device. For another example, if the AI / ML model identified as ID 1 for CSI reporting is not running on the second network device, a failure message is fed back to the first network device. For another example, if the AI / ML model identified as ID 1 for CSI reporting is running on the second network device but the second network device determines not to stop the model, a rejection message is fed back to the first network device.

[0091] The above-mentioned AI model identification indication is merely exemplary and may also include one or more combinations of AI function indication, AI type indication, AI model group indication, and AI model group internal indication. By using one or more corresponding indication combinations, it is possible to exchange information about stopping a specific AI model between adjacent network nodes.

[0092] The above schematically illustrates the case of stopping the model. The following describes the case of updating the model.

[0093] In some embodiments, the second indication information is carried by an XnAP message; for example, the XnAP message includes: a configuration update (NG-RAN node Configuration Update) message, an Xn establishment message, or a resource status update message; the present application is not limited thereto, and it may also be a new XnAP message.

[0094] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information; the present application is not limited to this, for example, it may also include other information.

[0095] For example, the second indication information includes {CSI; ID 1; ID2}, indicating that the AI / ML model identified as ID 1 used for CSI reporting is updated to the AI / ML model identified as ID 2; for another example, the second indication information includes {CSI}, indicating that all AI / ML models used for CSI reporting are updated; for another example, the second indication information includes {CSI; ID2; V2.0}, indicating that the AI / ML model used for CSI reporting is updated to the AI / ML model identified as ID2 and version 2.0.

[0096] The above only exemplifies the second indication information, but the present application is not limited thereto. For example, the second indication information may also be indicated by a predefined index. In addition, the second indication information may be explicitly indicated by a field or information element (IE) included in the message, or may be implicitly indicated by associating it with the message.

[0097] In some embodiments, the second indication information is used by the second network device to determine whether to update the AI / ML model in the second network device, and / or to determine whether to update the AI / ML model in a terminal device within a cell.

[0098] For example, when the identification information of the signal processing function corresponding to the AI / ML model in the second network device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device determines to update the AI / ML model in the second network device.

[0099] For example, after the second network device receives the second indication information {CSI; ID 1; ID 2}, if it is using the AI / ML model identified as ID 1 for CSI reporting, the second network device will update the AI / ML model identified as ID 1 for CSI reporting to the AI / ML model identified as ID 2.

[0100] For another example, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device notifies the terminal device to update the AI / ML model in the terminal device.

[0101] For example, after the second network device receives the second indication information {CSI; ID 1; ID2}, if it knows that one or more terminal devices in a certain cell are using the AI / ML model identified as ID 1 for CSI reporting, it sends a notification message to the terminal device in the cell, notifying the terminal device to update the AI / ML model identified as ID 1 for CSI reporting to the AI / ML model identified as ID 2.

[0102] In some embodiments, the first network device receives a confirmation message, a rejection message, or a failure message sent by the second network device.

[0103] For example, after successfully updating the AI / ML model used for CSI reporting, the second network device sends a confirmation message back to the first network device. For another example, if the second network device fails to successfully update the AI / ML model used for CSI reporting, a failure message is sent back to the first network device. For another example, if the second network device uses the AI / ML model used for CSI reporting but determines not to update the model, a rejection message is sent back to the first network device.

[0104] In some embodiments, the model itself can be updated, for example, updating a low-version model to a high-version model, or updating to a new model, or switching to a model with better performance, etc. The embodiments of the present application are not limited to this.

[0105] The following further explains the interaction between network devices and core network devices.

[0106] FIG4 is a schematic diagram of interaction between a network device and a core network device according to an embodiment of the present application. As shown in FIG4 , the method may include:

[0107] 401. The first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function.

[0108] 402. When it is determined to stop, update or switch the AI / ML model or model group, the first network device sends a first indication information for stopping the AI / ML model or model group and / or a second indication information for updating the AI / ML model or model group to the core network device through the NG interface.

[0109] It is worth noting that FIG4 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG4 above.

[0110] In some embodiments, the first indication information is carried by an NGAP message; for example, the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message, or a RAN configuration transfer message; the present application is not limited thereto, and it may also be a new NGAP message.

[0111] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowed list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information; the present application is not limited to this and may also include other information.

[0112] For example, the first indication information includes {CSI; ID 1}, which indicates that the AI / ML model identified as ID 1 used for CSI reporting is stopped; for another example, the first indication information includes {CSI}, which indicates that all AI / ML models used for CSI reporting are stopped; for another example, the first indication information includes {CSI; ID 1; V2.0}, which indicates that the AI / ML model identified as ID 1 and version 2.0 used for CSI reporting is stopped.

[0113] The above only exemplifies the first indication information, but the present application is not limited thereto. For example, the first indication information may also be indicated by a predefined index. In addition, the first indication information may be explicitly indicated by a field or information element (IE) included in the message, or may be implicitly indicated by associating it with the message.

[0114] In some embodiments, the first indication information is used by the core network device to determine whether to stop the AI / ML model in the network devices and / or terminal devices within a region.

[0115] For example, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the network device to stop the AI / ML model in the network device through an NGAP message.

[0116] For example, after the core network device receives the first indication information {CSI; ID 1}, if it knows that one or more network devices in the area are running the AI / ML model identified as ID 1 for CSI reporting, it sends an NGAP message to the network devices in the area to notify the network device to stop the AI / ML model identified as ID 1 for CSI reporting.

[0117] For another example, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the terminal device through a NAS message to stop the AI / ML model in the terminal device.

[0118] For example, after the core network device receives the first indication information {CSI; ID 1}, if it knows that one or more terminal devices in the area are running the AI / ML model identified as ID 1 for CSI reporting, it sends a NAS message to the terminal device in the area to notify the terminal device to stop the AI / ML model identified as ID 1 for CSI reporting.

[0119] In some embodiments, the first network device receives a confirmation message, a rejection message, or a failure message sent by the core network device.

[0120] For example, after the network devices or terminal devices in the area successfully stop the AI / ML model identified as ID 1 for CSI reporting, the core network device feeds back a confirmation message to the first network device. For another example, if the AI / ML model identified as ID 1 for CSI reporting is not running in the network devices and terminal devices in the area, a failure message is fed back to the first network device. For another example, if the AI / ML model identified as ID 1 for CSI reporting is running in the network devices or terminal devices in the area, but the core network device determines not to stop the model, a rejection message is fed back to the first network device.

[0121] The above schematically illustrates the case of stopping the model. The following describes the case of updating the model.

[0122] In some embodiments, the second indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message; the present application is not limited thereto, and it may also be a new NGAP message.

[0123] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowed list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information; the present application is not limited to this and may also include other information.

[0124] For example, the second indication information includes {CSI; ID 1; ID2}, indicating that the AI / ML model identified as ID 1 used for CSI reporting is updated to the AI / ML model identified as ID 2; for another example, the second indication information includes {CSI}, indicating that all AI / ML models used for CSI reporting are updated; for another example, the second indication information includes {CSI; ID 2; V2.0}, indicating that the AI / ML model used for CSI reporting is updated to the AI / ML model identified as ID 2 and version 2.0.

[0125] The above only exemplifies the second indication information, but the present application is not limited thereto. For example, the second indication information may also be indicated by a predefined index. In addition, the second indication information may be explicitly indicated by a field or information element (IE) included in the message, or may be implicitly indicated by associating it with the message.

[0126] In some embodiments, the second indication information is used by the core network device to determine whether to update the AI / ML model or model group in the network devices and / or terminal devices in a region.

[0127] For example, when the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the network device through an NGAP message to update the AI / ML model in the network device.

[0128] For example, after the core network device receives the second indication information {CSI; ID 1; ID2}, if it knows that one or more network devices in the area are using the AI / ML model identified as ID 1 for CSI reporting, the core network device sends an NGAP message to the network device, notifying the network device to update the AI / ML model identified as ID 1 for CSI reporting to the AI / ML model identified as ID 2.

[0129] For another example, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the terminal device through a NAS message to update the AI / ML model in the terminal device.

[0130] For example, after the core network device receives the second indication information {CSI; ID 1; ID2}, if it knows that one or more terminal devices in the area are using the AI / ML model identified as ID 1 for CSI reporting, it sends a NAS message to the terminal device in the area to notify the terminal device to update the AI / ML model identified as ID 1 for CSI reporting to the AI / ML model identified as ID 2.

[0131] In some embodiments, the first network device receives a confirmation message, a rejection message, or a failure message sent by the core network device.

[0132] For example, after the network devices and terminal devices in the area successfully update the AI / ML model used for CSI reporting, the core network device feeds back a confirmation message to the first network device. For another example, if the network devices and terminal devices in the area do not successfully update the AI / ML model used for CSI reporting, a failure message is fed back to the first network device. For another example, if the network devices and terminal devices in the area use the AI / ML model used for CSI reporting, but the core network device determines not to update the model, a rejection message is fed back to the first network device.

[0133] In some embodiments, if the recipient of the message to update the AI / ML model is a terminal device, and the terminal device decides to update, and if the new AI / ML model is already stored in the terminal device, the terminal device performs a corresponding update or switching operation; or if the new AI / ML model is not stored in the terminal device, the corresponding new AI / ML model can also be downloaded from the core network.

[0134] In some embodiments, if the recipient of the message to update the AI / ML model is a network device, and the network device decides to update, and if the new AI / ML model is already stored in the network device, the network device performs a corresponding update or switching operation; or if the new AI / ML model is not stored in the network device, the corresponding new AI / ML model can also be downloaded from the core network.

[0135] The following further describes the interaction between network devices and terminal devices.

[0136] FIG5 is a schematic diagram of interaction between a network device and a terminal device according to an embodiment of the present application. As shown in FIG5 , the method may include:

[0137] 501, the first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function;

[0138] 502. When it is determined to stop, update or switch the AI / ML model or model group, the first network device sends first indication information for stopping the AI / ML model or model group and / or second indication information for updating the AI / ML model or model group to one or more terminal devices in a cell or an area through system information and / or paging messages.

[0139] It is worth noting that FIG5 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG5 above.

[0140] In some embodiments, using system information to transmit parameters or instructions related to the AI / ML model used within a cell or region can save air interface signaling overhead. For example, the AI / ML model within a cell or region can be stopped, updated, or switched through a system information update procedure.

[0141] For example, the system information may include SIB 1 or a new SIB. For another example, the indication information may be sent to the terminal devices in the cell via a short message. For another example, the indication information may be sent to the terminal devices in the cell via a short message+paging message (indicating the UE ID).

[0142] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowed list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information; the present application is not limited to this and may also include other information.

[0143] For example, the first indication information includes {CSI; ID 1}, which indicates that the AI / ML model identified as ID 1 used for CSI reporting is stopped; for another example, the first indication information includes {CSI}, which indicates that all AI / ML models used for CSI reporting are stopped; for another example, the first indication information includes {CSI; ID 1; V2.0}, which indicates that the AI / ML model identified as ID 1 and version 2.0 used for CSI reporting is stopped.

[0144] The above only exemplifies the first indication information, but the present application is not limited thereto. For example, the first indication information may also be indicated by a predefined index. In addition, the first indication information may be explicitly indicated by a field or information element (IE) included in the message, or may be implicitly indicated by associating it with the message.

[0145] In some embodiments, the first indication information is used by the terminal device to determine whether to stop the AI / ML model in the terminal device.

[0146] For example, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the terminal device determines to stop the AI / ML model in the terminal device.

[0147] For example, after the terminal device receives the first indication information {CSI; ID 1}, if it is running the AI / ML model identified as ID 1 for CSI reporting, it stops the AI / ML model identified as ID 1 for CSI reporting.

[0148] In some embodiments, the first network device receives a confirmation message, a rejection message, or a failure message sent by the terminal device.

[0149] For example, after successfully stopping the AI / ML model identified as ID 1 for CSI reporting, the terminal device feeds back a confirmation message to the first network device. For another example, if the AI / ML model identified as ID 1 for CSI reporting is not running in the terminal device, a failure message is fed back to the first network device. For another example, if the AI / ML model identified as ID 1 for CSI reporting is running in the terminal device, but the terminal device determines not to stop the model, a rejection message is fed back to the first network device.

[0150] The above schematically illustrates the case of stopping the model. The following describes the case of updating the model.

[0151] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowed list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information; the present application is not limited to this and may also include other information.

[0152] For example, the second indication information includes {CSI; ID 1; ID2}, indicating that the AI / ML model identified as ID 1 used for CSI reporting is updated to the AI / ML model identified as ID 2; for another example, the second indication information includes {CSI}, indicating that all AI / ML models used for CSI reporting are updated; for another example, the second indication information includes {CSI; ID 2; V2.0}, indicating that the AI / ML model used for CSI reporting is updated to the AI / ML model identified as ID 2 and version 2.0.

[0153] The above only exemplifies the second indication information, but the present application is not limited thereto. For example, the second indication information may also be indicated by a predefined index. In addition, the second indication information may be explicitly indicated by a field or information element (IE) included in the message, or may be implicitly indicated by associating it with the message.

[0154] In some embodiments, the second indication information is used by the terminal device to determine whether to update the AI / ML model in the terminal device.

[0155] For example, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the terminal device determines to update the AI / ML model in the terminal device.

[0156] For example, after the terminal device receives the second indication information {CSI; ID 1; ID2}, if it is using the AI / ML model identified as ID 1 for CSI reporting, the terminal device will update the AI / ML model identified as ID 1 for CSI reporting to the AI / ML model identified as ID 2.

[0157] In some embodiments, the first network device receives a confirmation message, a rejection message, or a failure message sent by the terminal device.

[0158] For example, after successfully updating the AI / ML model used for CSI reporting, the terminal device feeds back a confirmation message to the first network device. For another example, if the terminal device fails to successfully update the AI / ML model used for CSI reporting, a failure message is fed back to the first network device. For another example, if the terminal device uses the AI / ML model used for CSI reporting but determines not to update the model, a rejection message is fed back to the first network device.

[0159] In some embodiments, the terminal device downloads the updated AI / ML model from the cell, or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information; the present application is not limited to this.

[0160] In some embodiments, the network device adds the identifier of an AI / ML model with poor performance to the exclusion list of the AI / ML model for the cell corresponding to the network device and broadcasts it in the cell as system information. The exclusion list includes the AI ​​model function identifier and the corresponding AI model identifier. In addition, it may further include the AI ​​model group identifier corresponding to a certain AI function, the group identifier, etc. Similarly, for other network devices that have identified poor performance through model monitoring and that determine that the terminal device-side AI / ML model needs to be deactivated, the model identifier can also be added to the exclusion list of the cell AI / ML model as described above.

[0161] In some embodiments, network devices can also accumulate a list of AI / ML models with particularly good performance for terminal devices and place it in the cell's AI / ML model allowlist, which is broadcast as system information within the cell. The allowlist includes the AI ​​model function identifier and the corresponding AI model identifier. In addition, it can further include the AI ​​model group identifier corresponding to a specific AI function and the group identifier.

[0162] In some embodiments, the exclusion list and / or allow list of the AI / ML model can be sent to adjacent cells or adjacent network devices, or to core network devices.

[0163] The above describes the terminal equipment within the cell. The following describes the terminal equipment that moves to the cell.

[0164] In some embodiments, a terminal device in an inactive state obtains identification information of an AI / ML model and / or version information of an AI / ML model from system information. When the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0165] In some embodiments, the terminal device downloads the updated AI / ML model from the cell, or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0166] For example, if the terminal device is in an inactive state (including RRC_IDLE, RRC_INACTIVE) and has relevant functional modules, the corresponding system information can be obtained; when the index of the AI / ML model of the corresponding function included in the system information is inconsistent with the index of its own model, the updated AI / ML model can be downloaded or loaded from the memory, or the updated AI / ML model can be obtained from the system information.

[0167] In some embodiments, an activated terminal device obtains identification information of an AI / ML model and / or version information of an AI / ML model from a target cell. When the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0168] In some embodiments, the terminal device downloads the updated AI / ML model from the target cell, or loads the updated AI / ML model from memory, or obtains the updated AI / ML model from system information.

[0169] For example, if the terminal device is in an activated state (i.e., RRC_CONNECTED), the source base station includes relevant functional module information of the terminal device (such as the existing functional modules, the ongoing functional modules, and the functional modules to be performed) and AI / ML model information (such as the AI ​​index being used, parameters) in the message sent to the target base station, and / or the target base station includes relevant functional module information (such as supported functional modules) and AI / ML model information (such as supported AI index, corresponding parameters) in the message sent to the source base station.

[0170] The terminal device can be configured based on the information of the AI-related functional modules in the message sent by the target base station (such as the handover command). Alternatively, when the terminal device completes the handover to the target cell (for example, when the random access process is completed in the target cell), the terminal device obtains the system information of the target cell.

[0171] If the AI / ML model index of the corresponding function included in the target cell is inconsistent with the index of its own model; or the AI / ML model ID is consistent but the version information is inconsistent, and if the version of the target cell is higher than its own version, the terminal device downloads or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0172] In some embodiments, the first network device receives a request sent by a terminal device; and sends system information related to the AI / ML model to the terminal device.

[0173] The terminal device obtains identification information of the AI / ML model and / or version information of the AI / ML model from the system information. When the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0174] For example, after an AI / ML module with a certain signal processing function receives a stop / update message for an AI / ML model, or before the terminal device is ready to use an AI / ML model with a certain signal processing function, it may initiate a request to update the relevant system information; after receiving the request, the base station sends the corresponding system information, which includes the corresponding signal processing function ID, the corresponding stopped AI / ML model index, the updated AI / ML model index; or the corresponding AI / ML model version information. If the AI / ML model needs to be updated, the terminal device updates to the latest AI / ML model in the cell.

[0175] In some embodiments, the system information is cell-specific or area-specific.

[0176] For example, when a terminal device having a certain signal processing function moves to a cell, or when a terminal device having a corresponding signal processing function is powered on, the terminal device reads SIB1 to determine whether the corresponding system information is cell-specific or area-specific.

[0177] If the system information is cell-specific, the terminal device checks whether the cell ID of the version of the system information stored is the same as the cell ID indicated by SIB1; if they are different, the terminal device obtains the corresponding system information. If the system information is area-specific, the terminal device checks whether the area ID of the system information field of the version of the system information stored is the same as the area ID indicated by SIB1; if they are different, the terminal device obtains the corresponding system information.

[0178] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0179] As can be seen from the above embodiment, the first network device determines whether to stop, update, or switch the AI / ML model for a certain signal processing function; and if it determines to stop, update, or switch the AI / ML model, it sends an instruction to the second network device, the terminal device, or the core network device. This allows monitoring the performance of the AI / ML model, maintaining the consistency of the AI / ML model operation, and improving the robustness of the model operation.

[0180] Embodiments of the second aspect

[0181] Based on the embodiments of the first aspect, model monitoring and information interaction are further described. For example, they are related to the aforementioned exclusion list and allow list. The embodiments of the second aspect can be combined with the embodiments of the first aspect or implemented separately.

[0182] In the embodiments of this application, use cases such as CSI feedback, beam management, and positioning can be targeted, as well as other applications that use AI / ML models to implement air interface signal processing. Model monitoring can be performed on the terminal device side or on the network device side. Model monitoring can be performed on the same side as the model, or on one side while the model runs on the other side.

[0183] The following first describes the situation where the terminal device performs model monitoring, and the model runs on the network device side, for example.

[0184] Figure 6 is a schematic diagram of the interaction between the network device and the terminal device of an embodiment of the present application. As shown in Figure 6, the network device receives the model identification information sent by the terminal device and the model performance information and / or model operation information corresponding to the model identification information.

[0185] In some embodiments, the model performance information is generated by the terminal device monitoring the AI / ML model. The terminal device monitors the output performance of the AI / ML model and / or monitors the input and output data distribution of the AI / ML model.

[0186] For example, model monitoring can be performed on the output performance of AI / ML models. This can be direct performance monitoring, such as measuring accuracy or performance, such as the similarity between model input and output, the accuracy of model estimates, the accuracy of model predictions, and the accuracy of model positioning. It can also be indirect performance monitoring, such as based on the output performance of the system after using the model, such as throughput, BLER, ACK, NACK, and receiver signal-to-interference-and-noise ratio (SIN) used to determine BLER.

[0187] For another example, model monitoring can be performed on the data distribution of the model input and / or output, that is, monitoring can be performed based on the characteristics of the input and output data, such as channel characteristics, signal-to-noise ratio, channel response characteristics, distribution of channel characteristics, etc.

[0188] In the above-mentioned model monitoring, when the model monitoring results do not meet the expected performance, for example, the accuracy of the model output is not good enough, or the system performance after using the model is lower than the threshold (poor performance), or the data distribution of the model input and output does not match the applicable scope of the model, the model can be disabled, or fall back to traditional non-AI methods, or model switching and other operations can be performed.

[0189] In some embodiments, the model operation information is determined by the terminal device based on a monitoring result of the AI / ML model. The monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0190] For example, a model identifier can be assigned to a monitored model. This identifier can be assigned by the network during the model registration process or by the manufacturer that developed the model. This identifier can be unique within a cell, within the coverage area of ​​the network equipment, within a specific operator's network, or within a network within a specified (or specified) range.

[0191] For another example, multiple models may be used for a specific functionality, such as CSI feedback information compression, beam space prediction, direct positioning, etc. To identify different models, different identifiers need to be obtained for different models during the model registration process. Different models may correspond to different scenarios, and / or different network configurations, and / or different locations, etc.

[0192] Based on model performance monitoring, when the model performance is poor and the model should be stopped, it can be defined as a model failure event or level. If the model performance is very good, it can be defined as a model excellence event or level.

[0193] As shown in Table 1, for example, the performance of the model can be expressed in two levels.

[0194] Table 1

[0195]

[0196] The model performance can also be further quantitatively graded, such as poor performance (model performance is unacceptable and worse than traditional methods), acceptable performance (model performance is similar to traditional methods), good performance (model performance is worse than traditional methods), excellent performance (model performance is much better than traditional methods), etc.

[0197] As shown in Table 2, for example, the performance of the model can be expressed in multiple levels,

[0198] Table 2

[0199]

[0200] The above is an exemplary description of the representation of model performance, and the embodiments of the present application are not limited to the above classification.

[0201] For monitoring based on the distribution of model input and output data, the monitored model input and output data distribution measurement results can also be classified. For example, they can be divided into two levels: special match and mismatch. Alternatively, they can be divided into multiple levels: special match, general match, mismatch, etc. The embodiments of the present application are not limited to the above classifications.

[0202] Since a model identifier is used for the model, the above monitoring results can be bound to the model identifier. In addition, the above model identifier and its corresponding model monitoring performance can be shared within a network area that maintains the uniqueness of its identifier, such as within a cell, within the coverage area of ​​a network device, within a certain operator network, or within a network within a specified (or specified) range. For example, the aforementioned exclusion list and / or allowance list can be used, but the present application is not limited thereto. This forms model monitoring historical data related to the model identifier.

[0203] In some embodiments, the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0204] For example, based on model monitoring results, network devices or terminal devices may make further model operation-related decisions, such as continuing to use the model, stopping the model, switching to another model, falling back to non-AI methods, model activation, model deactivation, or model recovery from failure.

[0205] Because the model uses a model identifier, further model operations based on the monitoring results can also be tied to the model identifier and shared within the corresponding network. For example, events such as fallback to non-AI methods, model deactivation, model switching, model activation, model deactivation, and model recovery from failure can all be shared within a certain scope.

[0206] In some embodiments, in addition to the model identifier, the model monitoring result information can be further bound to other identifiers, which can be at least one or any combination of the following: the cell identifier of the model, the model development manufacturer-related identifier, the user identifier used by the model, the scenario identifier of the model application, the configuration identifier of the model application, the location information identifier of the model application, the version information of the model, and other category identifiers associated with the monitoring model.

[0207] Table 3 shows an example of binding other identifiers.

[0208] Table 3

[0209]

[0210] In some embodiments, the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through at least one or any combination of uplink control information (UCI), MAC CE or RRC signaling.

[0211] In some embodiments, the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through an uplink control channel and / or an uplink data channel.

[0212] In some embodiments, the network device sends configuration information or instruction information to the terminal device to schedule the terminal device to send the model identification information and the model performance information and / or model operation information corresponding to the model identification information.

[0213] In some embodiments, the configuration information or indication information is dedicated information for the AI / ML model.

[0214] In some embodiments, the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through periodic reporting, semi-continuous reporting, or non-periodic reporting.

[0215] In some embodiments, the terminal device monitors the AI / ML model, and the AI / ML model is a model deployed on the first network device or a bilateral model.

[0216] In the case of model monitoring on the terminal device side, the network device can also request monitoring historical data.

[0217] FIG7 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application. As shown in FIG7 , the network device sends a request message for requesting monitoring data to the terminal device; for example, the request message includes the model identification information and the reporting configuration of the model performance information shown in FIG7 . The terminal device can send the model identification information, model performance information, and / or model operation information to the network device based on the request message.

[0218] In some embodiments, the request information includes at least one of the following: performance indicators, data distribution indicators, and model operation information after monitoring.

[0219] In some embodiments, the monitoring data is historical information, and the request information may include time range indication information.

[0220] In some embodiments, the request information includes a model identifier, but the present application is not limited thereto, and may also include other identification information corresponding to the required monitoring information.

[0221] In some embodiments, the network configures the terminal device to send model performance reports for multiple models, including model performance information and / or model operation information, or a list of related models. The model list here records multiple model identifiers owned by the terminal device and their corresponding model performance information and / or model operation information. The list can also only record a list of models with poor performance or a list of failed models; or only record a list of models with excellent performance or frequently selected models.

[0222] In some embodiments, the network side configures the terminal device to send the above-mentioned report to the network equipment of the core network, such as base station, LMF, OAM, AI / ML dedicated server, manufacturer or third-party server, etc.

[0223] The following further describes the situation where the network device performs model monitoring, and the model runs on the terminal device side, for example.

[0224] Figure 8 is a schematic diagram of the interaction between the network device and the terminal device of an embodiment of the present application. As shown in Figure 8, the network device sends model identification information and model performance information and / or model operation information corresponding to the model identification information to the terminal device.

[0225] Figure 8 illustrates the interaction between a network device and a terminal device as an example. This application is not limited to this. For example, a network device may also send model identification information and model performance information and / or model operation information corresponding to the model identification information to a core network device or other network device. Furthermore, the corresponding content already described in the preceding embodiments will be omitted below.

[0226] In some embodiments, the model performance information is generated by the network device monitoring the AI / ML model. For example, the network device monitors the output performance of the AI / ML model and / or monitors the input and output data distribution of the AI / ML model.

[0227] In some embodiments, the model operation information is determined by the network device based on a monitoring result of monitoring the AI / ML model.

[0228] In some embodiments, the monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0229] In some embodiments, the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0230] In some embodiments, the network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device through at least one or any combination of downlink control information (DCI), MAC CE or RRC signaling.

[0231] In some embodiments, the network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device via a downlink control channel and / or a downlink data channel.

[0232] In some embodiments, the network device monitors the AI / ML model, and the AI / ML model is a model deployed on the terminal device or a bilateral model.

[0233] In the case of model monitoring on the network device side, the terminal device can also request monitoring historical data.

[0234] FIG9 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application. As shown in FIG9 , the network device receives a request message sent by the terminal device for requesting monitoring data. Based on the request message, the network device may send model identification information, model performance information, and / or model operation information to the terminal device.

[0235] In some embodiments, the request information includes at least one of the following: performance indicators, data distribution indicators, and model operation information after monitoring.

[0236] In some embodiments, the monitoring data is historical information, and the request information includes time range indication information.

[0237] In some embodiments, the request information includes a model identifier, but the present application is not limited thereto, and may also include other identification information corresponding to the required monitoring information.

[0238] The above schematically illustrates model monitoring, model performance information and / or model operation information. The following schematically illustrates indicator information and / or model operation decision information related to model monitoring. The following can be combined with any of the above embodiments or implemented separately.

[0239] In some embodiments, the network device sends indicator information related to model monitoring and / or model operation decision information based on model monitoring to the terminal device.

[0240] For example, the indicator information related to model monitoring includes at least one of the following: performance indicator information related to model output, data distribution indicator information related to model output, and data distribution indicator information related to model input. The present application is not limited to this, and for example, other indicator information may also be included. The indicator information related to model monitoring may be model identification-specific information, or may be general model monitoring information.

[0241] For example, the model operation decision information based on model monitoring includes at least one of the following: information related to determining model deactivation, information related to determining model rollback, information related to determining model switching, information related to determining model stop, information related to determining model failure, information related to determining model switching from stop to enable, and information related to determining model update. The present application is not limited to this, and for example, other information may also be included. The model operation decision information based on model monitoring may be model identification-specific information, or may be general model monitoring information.

[0242] In some embodiments, the network device and / or the terminal device operates the model according to the results of model monitoring.

[0243] For example, after deciding to perform the operation, the terminal device can directly perform the corresponding operation.

[0244] For another example, the terminal device decides on an operation to be performed on the model according to the result of model monitoring, and reports the operation to the network device; after receiving a confirmation instruction from the network device, the terminal device performs the operation on the model.

[0245] For another example, the terminal device determines an operation on the model according to the result of model monitoring, and reports the operation to the network device, and the network device performs the operation on the model.

[0246] The above schematically illustrates the indicator information and / or model operation decision information related to model monitoring, and the following schematically illustrates the model query information and model feedback information. The following content can be combined with any of the above embodiments or implemented separately.

[0247] For example, the terminal device monitors the model's input information; if it finds that the input information differs significantly from the model input information used in model training, the terminal device can suggest or notify the network device to stop using the model. Alternatively, the terminal device monitors the model's input information; if it finds that the input information differs significantly from the model input information used in model training, but is similar to the input information used for training another model, the terminal device can suggest or notify the network device to switch to the other model.

[0248] For example, the model can be a bilateral model or a terminal device side model. The network device side can perform corresponding operations based on the information sent by the terminal device, such as deactivation, model rollback, model switching, or confirmation of model switching.

[0249] Figure 10 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application. As shown in Figure 10, the network device sends model query information (e.g., the model monitoring indication shown in Figure 10) to the terminal device, wherein the model query information includes a model identifier. The network device receives model feedback information (e.g., the model availability indication shown in Figure 10) sent by the terminal device and performs corresponding operations on the AI / ML model based on the model feedback information.

[0250] In some embodiments, the model feedback information includes at least one of the following: AI / ML model availability information corresponding to the model identifier, AI / ML model deactivation information corresponding to the model identifier, AI / ML model fallback information corresponding to the model identifier, and AI / ML model failure information corresponding to the model identifier.

[0251] In some embodiments, the network device continues to use the AI / ML model corresponding to the model identifier, or activates the AI / ML model corresponding to the model identifier, or deactivates the AI / ML model corresponding to the model identifier.

[0252] For example, after receiving a response from a terminal device indicating that the model can be run, the network device activates the model on the terminal device. In this case, the network device only allows the terminal device to determine whether the model can be used, assuming that the terminal device can determine whether the model is usable based on the model input.

[0253] FIG11 is a schematic diagram of the interaction between a network device and a terminal device according to an embodiment of the present application. As shown in FIG11 , the network device sends model query information (e.g., the model monitoring indication shown in FIG11 ) to the terminal device, wherein the model query information includes multiple model identifiers. The network device receives model feedback information (e.g., the selected model identifier indication information shown in FIG11 ) sent by the terminal device and performs corresponding operations on the AI / ML model based on the model feedback information.

[0254] In some embodiments, the model query information includes multiple model identifiers; the model feedback information includes a model identifier selected by the terminal device from the multiple model identifiers; and the network device activates the AI / ML model corresponding to the selected model identifier.

[0255] In some embodiments, the network only configures or instructs the terminal to select a model. The terminal selects a model based on the signal characteristics sent by the network (e.g., RS signals related to model input, such as CSI-RS and PRS), according to the network's configuration or command, and sends identification information of the selected model or models back to the network. Based on the base station's configuration, the returned information may also include corresponding monitoring performance information.

[0256] For example, a network device sends a model query to a terminal device, including multiple model identification information. The network device then receives the model identification information reported by the terminal device in response to the query. The network device then activates the model corresponding to the model identification. In this case, the network device instructs the terminal device to select a model based on the input signal characteristics. Once the model is selected, the terminal device notifies the network device, which then activates the model.

[0257] For example, the above-mentioned model query and model feedback can be mainly targeted at the bilateral model, and the model selection or model judgment can be based on the distribution characteristics of the input data.

[0258] For example, as shown in Figures 10 and 11, the network device and the terminal device can have a model generation unit and a model reconstruction unit, respectively, to form a bilateral model. Therefore, after one side determines whether the model is enabled or disabled, it needs to inform the other side.

[0259] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0260] Embodiments of the third aspect

[0261] An embodiment of the present application provides an information processing method, which is described from the perspective of the second network device, and the contents that are the same as those in the first and second embodiments will not be repeated.

[0262] FIG12 is a schematic diagram of an information processing method according to an embodiment of the present application. As shown in FIG12 , the method includes:

[0263] 1201, the second network device receives instruction information sent by the first network device;

[0264] 1202. The second network device stops, updates, or switches an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0265] It is worth noting that FIG12 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG12 above.

[0266] In some embodiments, the first network device sends first indication information for stopping the AI / ML model to the second network device through the Xn interface.

[0267] In some embodiments, the first indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message, or a resource status update message.

[0268] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0269] In some embodiments, the first indication information is used by the second network device to determine whether to stop the AI / ML model in the second network device, and / or to determine whether to stop the AI / ML model in a terminal device within a cell.

[0270] In some embodiments, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the second network device, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device determines to stop the AI / ML model in the second network device; and / or

[0271] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device notifies the terminal device to stop the AI / ML model in the terminal device.

[0272] In some embodiments, the second network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0273] In some embodiments, the first network device sends second indication information for updating the AI / ML model to the second network device through the Xn interface.

[0274] In some embodiments, the second indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message, or a resource status update message.

[0275] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0276] In some embodiments, the second indication information is used by the second network device to determine whether to update the AI / ML model in the second network device, and / or to determine whether to update the AI / ML model in a terminal device within a cell.

[0277] In some embodiments, when the identification information of the signal processing function corresponding to the AI / ML model in the second network device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device determines to update the AI / ML model in the second network device; and / or

[0278] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device notifies the terminal device to update the AI / ML model in the terminal device.

[0279] In some embodiments, the second network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0280] In some embodiments, the second network device further receives an exclusion list and / or an allow list sent by the first network device.

[0281] In some embodiments, the second network device receives the model identification information sent by the first network device and the model performance information and / or model operation information corresponding to the model identification information.

[0282] In some embodiments, the model performance information is generated by the first network device monitoring the AI / ML model. The first network device monitors the output performance of the AI / ML model and / or the first network device monitors the input and output data distribution of the AI / ML model.

[0283] In some embodiments, the model operation information is determined by the first network device based on a monitoring result of the AI / ML model. The monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0284] In some embodiments, the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0285] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0286] As can be seen from the above embodiment, the second network device receives the instruction information sent by the first network device and, based on the instruction information, stops, updates, or switches the AI / ML model for a certain signal processing function. This allows monitoring the performance of the AI / ML model, maintaining the consistency of the AI / ML model operation, and improving the robustness of the model operation.

[0287] Embodiments of the fourth aspect

[0288] An embodiment of the present application provides an information processing method, which is explained from the core network device side, and the contents that are the same as those in the first and second aspects of the embodiments will not be repeated.

[0289] FIG13 is a schematic diagram of an information processing method according to an embodiment of the present application. As shown in FIG13 , the method includes:

[0290] 1301. The core network device receives instruction information sent by the first network device.

[0291] 1302. The core network device stops, updates, or switches an AI / ML model or model group for a certain signal processing function according to the indication information.

[0292] It is worth noting that FIG13 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG13 above.

[0293] In some embodiments, the first network device sends first indication information for stopping the AI / ML model to the core network device through the NG interface.

[0294] In some embodiments, the first indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message.

[0295] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0296] In some embodiments, the first indication information is used by the core network device to determine whether to stop the AI / ML model in the network devices and / or terminal devices within a region.

[0297] In some embodiments, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the network device through an NGAP message to stop the AI / ML model in the network device; and / or

[0298] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the terminal device through a NAS message to stop the AI / ML model in the terminal device.

[0299] In some embodiments, the core network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0300] In some embodiments, the first network device sends second indication information for updating the AI / ML model to the core network device through the NG interface.

[0301] In some embodiments, the second indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message.

[0302] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0303] In some embodiments, the second indication information is used by the core network device to determine whether to update the AI / ML model in the network devices and / or terminal devices in a region.

[0304] In some embodiments, when the identification information of the signal processing function corresponding to the AI / ML model in the network device in the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the network device through an NGAP message to update the AI / ML model in the network device; and / or

[0305] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the terminal device through a NAS message to update the AI / ML model in the terminal device.

[0306] In some embodiments, the core network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0307] In some embodiments, the core network device further receives an exclusion list and / or an allow list sent by the first network device.

[0308] In some embodiments, the core network device receives the model identification information sent by the first network device and the model performance information and / or model operation information corresponding to the model identification information.

[0309] In some embodiments, the model performance information is generated by the first network device monitoring the AI / ML model. The first network device monitors the output performance of the AI / ML model and / or the first network device monitors the input and output data distribution of the AI / ML model.

[0310] In some embodiments, the model operation information is determined by the first network device based on a monitoring result of the AI / ML model. The monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0311] In some embodiments, the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0312] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0313] As can be seen from the above embodiment, the core network device receives the instruction information sent by the first network device and, based on the instruction information, stops, updates, or switches the AI / ML model for a certain signal processing function. This allows monitoring the performance of the AI / ML model, maintaining the consistency of the AI / ML model operation, and improving the robustness of the model operation.

[0314] Embodiments of the fifth aspect

[0315] An embodiment of the present application provides an information processing method, which is explained from the perspective of a terminal device, and the contents that are the same as those in the first and second aspects of the embodiments will not be repeated.

[0316] FIG14 is a schematic diagram of an information processing method according to an embodiment of the present application. As shown in FIG14 , the method includes:

[0317] 1401, a terminal device receives instruction information sent by a first network device;

[0318] 1402. The terminal device stops, updates, or switches an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0319] It is worth noting that FIG14 above is merely a schematic illustration of an embodiment of the present application, and the present application is not limited thereto. For example, the execution order of the various operations may be appropriately adjusted, and other operations may be added or some operations may be reduced. Those skilled in the art may make appropriate modifications based on the above description, and are not limited to the description of FIG14 above.

[0320] In some embodiments, the first network device sends first indication information for stopping the AI / ML model to one or more terminal devices in a cell or an area through system information and / or paging message.

[0321] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0322] In some embodiments, the first indication information is used by the terminal device to determine whether to stop the AI / ML model in the terminal device.

[0323] In some embodiments, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the terminal device determines to stop the AI / ML model in the terminal device.

[0324] In some embodiments, the first network device sends second indication information for updating the AI / ML model to one or more terminal devices in a cell or an area through system information and / or paging messages.

[0325] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0326] In some embodiments, the second indication information is used by the terminal device to determine whether to update the AI / ML model in the terminal device.

[0327] In some embodiments, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the terminal device determines to update the AI / ML model in the terminal device.

[0328] In some embodiments, a terminal device in an inactive state obtains identification information of an AI / ML model and / or version information of an AI / ML model from system information. When the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0329] In some embodiments, the terminal device downloads the updated AI / ML model from the cell, or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0330] In some embodiments, an activated terminal device obtains identification information of an AI / ML model and / or version information of an AI / ML model from a target cell. When the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0331] In some embodiments, the terminal device downloads the updated AI / ML model from the target cell, or loads the updated AI / ML model from memory, or obtains the updated AI / ML model from system information.

[0332] In some embodiments, the terminal device sends a request to a first network device; and the terminal device receives system information related to the AI / ML model.

[0333] In some embodiments, the terminal device obtains identification information of the AI / ML model and / or version information of the AI / ML model from the system information. When the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0334] In some embodiments, the system information is cell-specific or area-specific.

[0335] In some embodiments, the terminal device also receives an exclusion list and / or an allow list sent by the first network device.

[0336] In some embodiments, the terminal device sends model identification information and model performance information and / or model operation information corresponding to the model identification information to the network device.

[0337] In some embodiments, the model performance information is generated by the terminal device monitoring the AI / ML model. The terminal device monitors the output performance of the AI / ML model and / or monitors the input and output data distribution of the AI / ML model.

[0338] In some embodiments, the model operation information is determined by the terminal device based on a monitoring result of the AI / ML model. The monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0339] In some embodiments, the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0340] In some embodiments, the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through at least one or any combination of uplink control information (UCI), MAC CE or RRC signaling.

[0341] In some embodiments, the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through an uplink control channel and / or an uplink data channel.

[0342] In some embodiments, the terminal device receives configuration information or indication information sent by the network device, and the terminal device sends the model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0343] In some embodiments, the configuration information or indication information is dedicated information for the AI / ML model.

[0344] In some embodiments, the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through periodic reporting, semi-continuous reporting, or non-periodic reporting.

[0345] In some embodiments, the terminal device monitors the AI / ML model, and the AI / ML model is a model deployed on the first network device or a bilateral model.

[0346] In some embodiments, the terminal device receives request information sent by the network device for requesting monitoring data.

[0347] In some embodiments, the request information includes at least one of the following: performance indicators, data distribution indicators, and model operation information after monitoring.

[0348] In some embodiments, the monitoring data is historical information, and the request information includes time range indication information.

[0349] In some embodiments, the request information includes a model identifier.

[0350] In some embodiments, the terminal device receives the model identification information sent by the network device and the model performance information and / or model operation information corresponding to the model identification information.

[0351] In some embodiments, the model performance information is generated by the network device monitoring the AI / ML model. The network device monitors the output performance of the AI / ML model and / or monitors the input and output data distribution of the AI / ML model.

[0352] In some embodiments, the model operation information is determined by the network device based on a monitoring result of the AI / ML model. The monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0353] In some embodiments, the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0354] In some embodiments, the network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device through at least one or any combination of downlink control information (DCI), MAC CE or RRC signaling.

[0355] In some embodiments, the network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device via a downlink control channel and / or a downlink data channel.

[0356] In some embodiments, the network device monitors the AI / ML model, and the AI / ML model is a model deployed on the terminal device or a bilateral model.

[0357] In some embodiments, the terminal device sends a request message for requesting monitoring data to the network device, wherein the request message includes at least one of the following: performance indicators, data distribution indicators, and model operation information after monitoring.

[0358] In some embodiments, the monitoring data is historical information, and the request information includes time range indication information.

[0359] In some embodiments, the request information includes a model identifier.

[0360] In some embodiments, the terminal device receives indicator information related to model monitoring and / or model operation decision information based on model monitoring sent by the network device.

[0361] In some embodiments, the indicator information related to model monitoring includes at least one of the following: performance indicator information related to model output, data distribution indicator information related to model output, and data distribution indicator information related to model input.

[0362] In some embodiments, the indicator information related to model monitoring is model identification-specific information, or is model monitoring general information.

[0363] In some embodiments, the model operation decision information based on model monitoring includes at least one of the following: information related to determining model deactivation, information related to determining model rollback, information related to determining model switching, information related to determining model stoppage, information related to determining model failure, information related to determining model switching from stop to enable, and information related to determining model update.

[0364] In some embodiments, the model operation decision information based on model monitoring is model identification specific information, or is model monitoring general information.

[0365] In some embodiments, the network device and / or the terminal device operates the model according to the results of model monitoring.

[0366] In some embodiments, the terminal device decides on an operation to be performed on the model based on the results of model monitoring, and reports the operation to the network device; after receiving a confirmation instruction from the network device, the terminal device performs the operation on the model.

[0367] In some embodiments, the terminal device determines an operation to be performed on the model based on the result of model monitoring, and reports the operation to the network device, and the network device performs the operation on the model.

[0368] In some embodiments, the terminal device receives model query information sent by the network device; and sends model feedback information to the network device, and the network device performs corresponding operations on the AI / ML model based on the model feedback information.

[0369] In some embodiments, the model query information includes a model identifier; the model feedback information includes at least one of the following: AI / ML model availability information corresponding to the model identifier, AI / ML model deactivation information corresponding to the model identifier, AI / ML model fallback information corresponding to the model identifier, and AI / ML model failure information corresponding to the model identifier.

[0370] In some embodiments, the network device continues to use the AI / ML model corresponding to the model identifier, or activates the AI / ML model corresponding to the model identifier, or deactivates the AI / ML model corresponding to the model identifier.

[0371] In some embodiments, the model query information includes multiple model identifiers; and the model feedback information includes a model identifier selected by the terminal device from the multiple model identifiers.

[0372] In some embodiments, the network device activates the AI / ML model corresponding to the selected model identifier.

[0373] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used individually, or one or more of the above embodiments may be combined.

[0374] As can be seen from the above embodiment, the terminal device receives the instruction information sent by the first network device and stops, updates, or switches the AI / ML model for a certain signal processing function according to the instruction information. This allows monitoring the performance of the AI / ML model, maintaining the consistency of the AI / ML model operation, and improving the robustness of the model operation.

[0375] Embodiments of the sixth aspect

[0376] The embodiment of the present application provides an information indication device. The device may be, for example, a network device (such as the aforementioned first network device), or one or more components or assemblies configured on the network device. The contents that are the same as those in the first and second aspects of the embodiment are not repeated here.

[0377] FIG15 is a schematic diagram of an information indication device according to an embodiment of the present application. As shown in FIG15 , the information indication device 1500 includes:

[0378] a determining unit 1501 , which determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and

[0379] The sending unit 1502 sends indication information to the second network device, the terminal device or the core network device when it is determined to stop, update or switch the AI / ML model or model group.

[0380] In some embodiments, first indication information for stopping the AI / ML model or model group and / or second indication information for updating the AI / ML model or model group are sent to the second network device via the Xn interface.

[0381] In some embodiments, the first indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message, or a resource status update message;

[0382] The first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0383] In some embodiments, the first indication information is used by the second network device to determine whether to stop the AI / ML model in the second network device, and / or to determine whether to stop the AI / ML model in a terminal device within a cell;

[0384] Wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the second network device, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device determines to stop the AI / ML model in the second network device; and / or

[0385] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device notifies the terminal device to stop the AI / ML model in the terminal device.

[0386] In some embodiments, the second indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message, or a resource status update message;

[0387] The second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, internal identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, internal identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0388] In some embodiments, the second indication information is used by the second network device to determine whether to update the AI / ML model in the second network device, and / or to determine whether to update the AI / ML model in a terminal device within a cell;

[0389] In which, when the identification information of the signal processing function corresponding to the AI / ML model in the second network device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device determines to update the AI / ML model in the second network device; and / or

[0390] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device notifies the terminal device to update the AI / ML model in the terminal device.

[0391] In some embodiments, first indication information for stopping the AI / ML model and / or second indication information for updating the AI / ML model are sent to the core network device via the NG interface.

[0392] In some embodiments, the first indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message, or a RAN configuration transfer message;

[0393] The first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0394] In some embodiments, the first indication information is used by the core network device to determine whether to stop the AI / ML model in the network devices and / or terminal devices in a region;

[0395] Wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the network device through an NGAP message to stop the AI / ML model in the network device; and / or

[0396] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the terminal device through a NAS message to stop the AI / ML model in the terminal device.

[0397] In some embodiments, the second indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message, or a RAN configuration transfer message;

[0398] The second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0399] In some embodiments, the second indication information is used by the core network device to determine whether to update the AI / ML model in the network devices and / or terminal devices in a region;

[0400] In which, when the identification information of the signal processing function corresponding to the AI / ML model in the network device in the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the network device through an NGAP message to update the AI / ML model in the network device; and / or

[0401] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the terminal device through a NAS message to update the AI / ML model in the terminal device.

[0402] In some embodiments, the first network device sends first indication information for stopping the AI / ML model and / or second indication information for updating the AI / ML model to one or more terminal devices in a cell or an area through system information and / or paging messages.

[0403] In some embodiments, the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0404] In some embodiments, the first indication information is used by the terminal device to determine whether to stop the AI / ML model in the terminal device;

[0405] In which, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the terminal device determines to stop the AI / ML model in the terminal device.

[0406] In some embodiments, the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0407] In some embodiments, the second indication information is used by the terminal device to determine whether to update the AI / ML model in the terminal device;

[0408] In which, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the terminal device determines to update the AI / ML model in the terminal device.

[0409] In some embodiments, a terminal device in an inactive state obtains identification information of an AI / ML model and / or version information of an AI / ML model from system information. If the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0410] The terminal device downloads the updated AI / ML model from the cell, loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0411] In some embodiments, the terminal device in the activated state obtains identification information of the AI / ML model and / or version information of the AI / ML model from the target cell. If the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device;

[0412] The terminal device downloads the updated AI / ML model from the target cell, or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0413] In some embodiments, the first network sends a model exclusion list and / or an allow list, or model identification information and model performance information and / or model operation information corresponding to the model identification information, to other network elements of the core network, such as LMF, OAM, etc. Alternatively, the above-mentioned network element, such as LMF, sends a model exclusion list and / or an allow list, or model identification information and model performance information and / or model operation information corresponding to the model identification information, to other network elements, including the first network device.

[0414] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0415] It is worth noting that the above description only describes the components or modules related to the present application, but the present application is not limited thereto. The information indicating device 1500 may also include other components or modules. For details of these components or modules, reference may be made to related technologies.

[0416] In addition, for the sake of simplicity, FIG15 only illustrates the connection relationship or signal direction between various components or modules. However, it should be clear to those skilled in the art that various related technologies such as bus connection can be used. The above-mentioned components or modules can be implemented by hardware facilities such as processors, memories, transmitters, and receivers; the implementation of this application is not limited to this.

[0417] As can be seen from the above embodiment, the first network device determines whether to stop, update, or switch the AI / ML model for a certain signal processing function; and if it determines to stop, update, or switch the AI / ML model, it sends an instruction to the second network device, the terminal device, or the core network device. This allows monitoring the performance of the AI / ML model, maintaining the consistency of the AI / ML model operation, and improving the robustness of the model operation.

[0418] Embodiments of the seventh aspect

[0419] The embodiments of the present application provide an information processing device. The device may be, for example, a network device (such as the aforementioned second network device), a core network device, or a terminal device, or may be one or more components or assemblies configured on the network device, the core network device, or the terminal device. The contents that are the same as those in the third to fifth aspects of the embodiments are not repeated here.

[0420] FIG16 is a schematic diagram of an information processing device according to an embodiment of the present application. As shown in FIG16 , the information processing device 1600 includes:

[0421] A receiving unit 1601 receives instruction information sent by a first network device; and

[0422] The processing unit 1602 stops, updates, or switches an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0423] The above embodiments are merely exemplary of the present invention, but the present invention is not limited thereto. Appropriate modifications may be made based on the above embodiments. For example, the above embodiments may be used alone, or one or more of the above embodiments may be combined.

[0424] It is worth noting that the above description only describes the components or modules related to the present application, but the present application is not limited thereto. The information processing device 1600 may also include other components or modules. For details of these components or modules, reference may be made to related technologies.

[0425] In addition, for the sake of simplicity, FIG16 only illustrates the connection relationship or signal direction between various components or modules. However, it should be clear to those skilled in the art that various related technologies such as bus connection can be used. The above-mentioned components or modules can be implemented by hardware facilities such as processors, memories, transmitters, and receivers; the implementation of this application is not limited to this.

[0426] As can be seen from the above embodiment, by receiving the instruction information sent by the first network device and stopping, updating, or switching the AI / ML model for a certain signal processing function according to the instruction information, it is possible to monitor the performance of the AI / ML model operation, maintain the consistency of the AI / ML model operation, and improve the robustness of the model operation.

[0427] Embodiments of the eighth aspect

[0428] An embodiment of the present application also provides a communication system, and reference may be made to FIG1 . The contents identical to those in the first to seventh aspects of the embodiments will not be repeated.

[0429] In some embodiments, the communication system 100 may include at least:

[0430] The network device 101 determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and when it is determined to stop, update, or switch the AI / ML model or model group, sends an indication message to a second network device, a terminal device, or a core network device.

[0431] An embodiment of the present application also provides a network device or core network device, which may be, for example, a base station, but the present application is not limited thereto and may also be other network devices or core network devices.

[0432] Figure 17 is a schematic diagram illustrating the configuration of a network device or core network device according to an embodiment of the present application. As shown in Figure 17 , network device or core network device 1700 may include a processor 1710 (e.g., a central processing unit (CPU)) and a memory 1720 ; the memory 1720 is coupled to the processor 1710 . The memory 1720 may store various data and may also store an information processing program 1730 , which is executed under the control of the processor 1710 .

[0433] For example, the processor 1710 may be configured to execute a program to implement the information indication method as described in the embodiment of the first aspect. For example, the processor 1710 may be configured to perform the following control: determine whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and when it is determined to stop, update, or switch the AI / ML model or model group, send indication information to the second network device, the terminal device, or the core network device.

[0434] For example, the processor 1710 may be configured to execute a program to implement the information processing method described in the second to fourth aspects of the embodiments. For example, the processor 1710 may be configured to perform the following control: receiving instruction information sent by the first network device; and stopping, updating, or switching an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0435] In addition, as shown in Figure 17, the network device or core network device 1700 may further include: a transceiver 1740 and an antenna 1750; wherein, the functions of the above components are similar to those in the prior art and are not further described here. It is worth noting that the network device or core network device 1700 does not necessarily include all the components shown in Figure 17; in addition, the network device or core network device 1700 may also include components not shown in Figure 17, and reference may be made to the prior art.

[0436] The embodiment of the present application also provides a terminal device, but the present application is not limited thereto and may also be other devices.

[0437] Figure 18 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in Figure 18 , terminal device 1800 may include a processor 1810 and a memory 1820. Memory 1820 stores data and programs and is coupled to processor 1810. It should be noted that this diagram is exemplary; other types of structures may be used to supplement or replace this structure to implement telecommunication or other functions.

[0438] For example, the processor 1810 may be configured to execute a program to implement the information processing method as described in the embodiment of the fifth aspect. For example, the processor 1810 may be configured to perform the following control: receiving instruction information sent by the first network device; and stopping, updating, or switching an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0439] As shown in Figure 18 , the terminal device 1800 may further include: a communication module 1830, an input unit 1840, a display 1850, and a power supply 1860. The functions of these components are similar to those in the prior art and are not described in detail here. It is worth noting that the terminal device 1800 does not necessarily include all of the components shown in Figure 18 , and these components are not essential. Furthermore, the terminal device 1800 may also include components not shown in Figure 18 , for which reference may be made to the prior art.

[0440] An embodiment of the present application also provides a computer program, wherein when the program is executed in a terminal device, the program causes the terminal device to execute the information processing method described in the embodiment of the fifth aspect.

[0441] An embodiment of the present application also provides a storage medium storing a computer program, wherein the computer program enables a terminal device to execute the information processing method described in the embodiment of the fifth aspect.

[0442] An embodiment of the present application also provides a computer program, wherein when the program is executed in a network device or a core network device, the program enables the network device or the core network device to execute the information indication method described in the embodiment of the first aspect or the information processing method described in the embodiments of the second to fourth aspects.

[0443] An embodiment of the present application also provides a storage medium storing a computer program, wherein the computer program enables a network device or a core network device to execute the information indication method described in the embodiment of the first aspect or the information processing method described in the embodiments of the second to fourth aspects.

[0444] The above devices and methods of the present application can be implemented by hardware or by a combination of hardware and software. The present application relates to such a computer-readable program that, when executed by a logic component, enables the logic component to implement the devices or components described above, or enables the logic component to implement the various methods or steps described above. The present application also relates to a storage medium for storing the above program, such as a hard disk, a magnetic disk, an optical disk, a DVD, a flash memory, etc.

[0445] The method / device described in conjunction with the embodiments of the present application can be directly embodied as hardware, a software module executed by a processor, or a combination of the two. For example, one or more of the functional block diagrams shown in the figure and / or one or more combinations of functional block diagrams can correspond to various software modules of the computer program flow or to various hardware modules. These software modules can respectively correspond to the various steps shown in the figure. These hardware modules can be implemented by solidifying these software modules, for example, using a field programmable gate array (FPGA).

[0446] The software module may be located in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. A storage medium may be coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium; or the storage medium may be an integral part of the processor. The processor and the storage medium may be located in an ASIC. The software module may be stored in the memory of the mobile terminal or in a memory card that can be inserted into the mobile terminal. For example, if the device (such as a mobile terminal) uses a large-capacity MEGA-SIM card or a large-capacity flash memory device, the software module may be stored in the MEGA-SIM card or the large-capacity flash memory device.

[0447] One or more of the functional blocks and / or one or more combinations of functional blocks described in the accompanying drawings may be implemented as a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, or any appropriate combination thereof for performing the functions described in this application. One or more of the functional blocks and / or one or more combinations of functional blocks described in the accompanying drawings may also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors in communication with a DSP, or any other such configuration.

[0448] The present application has been described above in conjunction with specific embodiments. However, those skilled in the art should understand that these descriptions are merely illustrative and are not intended to limit the scope of protection of the present application. Those skilled in the art may make various modifications and variations to the present application based on the spirit and principles of the present application, and such modifications and variations are also within the scope of the present application.

[0449] Regarding the implementation methods including the above embodiments, the following additional notes are also disclosed:

[0450] 1. An information indication method, comprising:

[0451] The first network device determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; and

[0452] When it is determined to stop, update or switch the AI / ML model or model group, an indication message is sent to the second network device, the terminal device or the core network device.

[0453] 2. The method according to Note 1, wherein the first network device sends a first indication message for stopping the AI / ML model or model group to the second network device through the Xn interface.

[0454] 3. The method according to Note 2, wherein the first indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message or a resource status update message.

[0455] 4. The method according to Note 2, wherein the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowed list of models, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0456] 5. The method according to Note 2, wherein the first indication information is used by the second network device to determine whether to stop the AI / ML model or model group in the second network device, and / or to determine whether to stop the AI / ML model or model group in a terminal device within a cell.

[0457] 6. The method according to Note 5, wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the second network device, the identification information of the AI / ML model, the identification information of the AI / ML model group, the group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the second network device determines to stop the AI / ML model in the second network device; and / or

[0458] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device notifies the terminal device to stop the AI / ML model in the terminal device.

[0459] 7. The method according to any one of Notes 2 to 6, wherein the method comprises:

[0460] The first network device receives a confirmation message, a rejection message, or a failure message sent by the second network device.

[0461] 8. The method according to Note 1, wherein the first network device sends second indication information for updating the AI / ML model to the second network device through the Xn interface.

[0462] 9. The method according to Note 8, wherein the second indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message or a resource status update message.

[0463] 10. The method according to Note 8, wherein the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0464] 11. The method according to Note 8, wherein the second indication information is used by the second network device to determine whether to update the AI / ML model in the second network device, and / or to determine whether to update the AI / ML model in a terminal device within a cell.

[0465] 12. The method according to Note 11, wherein, when the identification information of the signal processing function corresponding to the AI / ML model in the second network device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device determines to update the AI / ML model in the second network device; and / or

[0466] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device notifies the terminal device to update the AI / ML model in the terminal device.

[0467] 13. The method according to any one of Notes 8 to 12, wherein the method comprises:

[0468] The first network device receives a confirmation message, a rejection message, or a failure message sent by the second network device.

[0469] 14. The method according to Note 1, wherein the first network device sends a first indication message for stopping the AI / ML model to the core network device through the NG interface.

[0470] 15. The method according to Note 14, wherein the first indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message.

[0471] 16. The method according to Note 14, wherein the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0472] 17. The method according to Note 14, wherein the first indication information is used by the core network device to determine whether to stop the AI / ML model in the network devices and / or terminal devices within a region.

[0473] 18. The method according to Note 17, wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the network device through an NGAP message to stop the AI / ML model in the network device; and / or

[0474] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the terminal device through a NAS message to stop the AI / ML model in the terminal device.

[0475] 19. The method according to any one of Notes 14 to 18, wherein the method comprises:

[0476] The first network device receives a confirmation message, a rejection message, or a failure message sent by the core network device.

[0477] 20. The method according to Note 1, wherein the first network device sends second indication information for updating the AI / ML model to the core network device through the NG interface.

[0478] 21. The method according to Note 20, wherein the second indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message.

[0479] 22. The method according to Note 20, wherein the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0480] 23. The method according to Note 20, wherein the second indication information is used by the core network device to determine whether to update the AI / ML model in the network devices and / or terminal devices within a region.

[0481] 24. The method according to Note 23, wherein, when the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the network device through an NGAP message to update the AI / ML model in the network device; and / or

[0482] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the terminal device through a NAS message to update the AI / ML model in the terminal device.

[0483] 25. The method according to any one of Notes 20 to 24, wherein the method comprises:

[0484] The first network device receives a confirmation message, a rejection message, or a failure message sent by the core network device.

[0485] 26. The method according to Note 1, wherein the first network device sends a first indication information for stopping the AI / ML model to one or more terminal devices in a cell or an area through system information and / or paging messages.

[0486] 27. The method according to Note 26, wherein the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0487] 28. The method according to Note 27, wherein the first indication information is used by the terminal device to determine whether to stop the AI / ML model in the terminal device.

[0488] 29. The method according to Note 28, wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the terminal device determines to stop the AI / ML model in the terminal device.

[0489] 30. The method according to Note 1, wherein the first network device sends second indication information for updating the AI / ML model to one or more terminal devices within a cell or an area through system information and / or paging messages.

[0490] 31. The method according to Note 30, wherein the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0491] 32. The method according to Note 30, wherein the second indication information is used by the terminal device to determine whether to update the AI / ML model in the terminal device.

[0492] 33. The method according to Note 32, wherein, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the terminal device determines to update the AI / ML model in the terminal device.

[0493] 34. The method according to Note 1, wherein a terminal device in an inactive state obtains identification information of an AI / ML model and / or version information of an AI / ML model from system information, and when the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0494] 35. The method according to Note 34, wherein the terminal device downloads the updated AI / ML model or model group from the cell, or loads the updated AI / ML model or model group from the memory, or obtains the updated AI / ML model or model group from the system information.

[0495] 36. The method according to Note 1, wherein the terminal device in an activated state obtains identification information of the AI / ML model and / or version information of the AI / ML model from the target cell, and when the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0496] 37. The method according to Note 36, wherein the terminal device downloads the updated AI / ML model or model group from the target cell, or loads the updated AI / ML model or model group from the memory, or obtains the updated AI / ML model or model group from the system information.

[0497] 38. The method according to Supplementary Note 1, wherein the method further comprises:

[0498] The first network device receives the request sent by the terminal device; and

[0499] Send system information related to the AI / ML model to the terminal device.

[0500] 39. The method according to Note 38, wherein the terminal device obtains identification information of the AI / ML model and / or version information of the AI / ML model from the system information, and when the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0501] 40. The method according to Note 38, wherein the system information is cell-specific or area-specific.

[0502] 41. The method according to any one of Notes 1 to 40, wherein the method further comprises:

[0503] The first network device further generates an exclusion list and / or an allow list; and

[0504] Send the exclusion list and / or the allow list to the second network device or the terminal device or the core network device.

[0505] 42. The method according to Supplement 41, wherein the method further comprises:

[0506] The first network device adds the AI / ML model identifier whose performance is lower than the threshold (poor performance) to the exclusion list of the AI / ML model of the cell corresponding to the network device.

[0507] 43. The method according to Note 42, wherein the exclusion list includes an AI model function identifier and a corresponding AI model identifier.

[0508] 44. The method according to Note 43, wherein the exclusion list further includes the AI ​​model group identifier and the intra-group identifier corresponding to a certain AI function.

[0509] 45. The method according to Supplement 41, wherein the method further comprises:

[0510] The first network device adds the AI / ML model identifier whose performance is higher than or equal to the threshold (better performance) to the allowed list of AI / ML models of the cell corresponding to the network device.

[0511] 46. ​​The method according to Note 45, wherein the allowed list includes an AI model function identifier and a corresponding AI model identifier.

[0512] 47. The method according to Note 46, wherein the allowed list further includes the AI ​​model group identifier and the intra-group identifier corresponding to a certain AI function.

[0513] 48. An information processing method comprising:

[0514] The second network device receives the instruction information sent by the first network device; and

[0515] Stop, update, or switch an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0516] 49. The method according to Note 48, wherein the first network device sends a first indication message for stopping the AI / ML model to the second network device through an Xn interface.

[0517] 50. The method according to Note 49, wherein the first indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message or a resource status update message.

[0518] 51. The method according to Note 49, wherein the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0519] 52. The method according to Note 49, wherein the first indication information is used by the second network device to determine whether to stop the AI / ML model or model group in the second network device, and / or to determine whether to stop the AI / ML model or model group in the terminal device within a cell.

[0520] 53. The method according to Note 52, wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the second network device, the identification information of the AI / ML model, the identification information of the AI / ML model group, the group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the second network device determines to stop the AI / ML model in the second network device; and / or

[0521] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the second network device notifies the terminal device to stop the AI / ML model in the terminal device.

[0522] 54. The method according to any one of Notes 49 to 53, wherein the method comprises:

[0523] The second network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0524] 55. The method according to Note 48, wherein the first network device sends second indication information for updating the AI / ML model to the second network device through the Xn interface.

[0525] 56. The method according to Note 55, wherein the second indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message or a resource status update message.

[0526] 57. The method according to Note 55, wherein the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0527] 58. The method according to Note 55, wherein the second indication information is used by the second network device to determine whether to update the AI / ML model in the second network device, and / or to determine whether to update the AI / ML model in the terminal device within a cell.

[0528] 59. The method according to Note 58, wherein, when the identification information of the signal processing function corresponding to the AI / ML model in the second network device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device determines to update the AI / ML model in the second network device; and / or

[0529] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device notifies the terminal device to update the AI / ML model in the terminal device.

[0530] 60. The method according to any one of Notes 55 to 59, wherein the method comprises:

[0531] The second network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0532] 61. The method according to any one of Notes 48 to 60, further comprising:

[0533] The second network device further receives the exclusion list and / or the allow list sent by the first network device.

[0534] 62. An information processing method comprising:

[0535] The core network device receives the instruction information sent by the first network device; and

[0536] Stop, update, or switch an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0537] 63. The method according to Note 62, wherein the first network device sends a first indication message for stopping the AI / ML model or model group to the core network device through the NG interface.

[0538] 64. The method according to Note 63, wherein the first indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message.

[0539] 65. The method according to Note 63, wherein the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0540] 66. A method according to Note 63, wherein the first indication information is used by the core network device to determine whether to stop the AI / ML model or model group in the network equipment and / or terminal equipment within a region.

[0541] 67. The method according to Note 66, wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the network device through an NGAP message to stop the AI / ML model in the network device; and / or

[0542] When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area, the identification information of the AI / ML model, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the terminal device through a NAS message to stop the AI / ML model in the terminal device.

[0543] 68. The method according to any one of Notes 63 to 67, wherein the method comprises:

[0544] The core network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0545] 69. The method according to Note 62, wherein the first network device sends second indication information for updating the AI / ML model to the core network device through the NG interface.

[0546] 70. The method according to Note 69, wherein the second indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message.

[0547] 71. The method according to Note 69, wherein the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0548] 72. The method according to Note 69, wherein the second indication information is used by the core network device to determine whether to update the AI / ML model or model group in the network devices and / or terminal devices within a region.

[0549] 73. The method according to Note 72, wherein, when the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the network device through an NGAP message to update the AI / ML model in the network device; and / or

[0550] When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the terminal device through a NAS message to update the AI / ML model in the terminal device.

[0551] 74. The method according to any one of Notes 69 to 73, wherein the method comprises:

[0552] The core network device sends a confirmation message, a rejection message, or a failure message to the first network device.

[0553] 75. The method according to any one of Notes 62 to 74, wherein the method further comprises:

[0554] The core network device also receives the exclusion list and / or the allow list sent by the first network device.

[0555] 76. An information processing method comprising:

[0556] The terminal device receives the instruction information sent by the first network device; and

[0557] Stop, update, or switch an AI / ML model or model group for a certain signal processing function according to the instruction information.

[0558] 77. The method according to Note 76, wherein the first network device sends a first indication information for stopping the AI / ML model or model group to one or more terminal devices within a cell or an area through system information and / or paging messages.

[0559] 78. The method according to Note 77, wherein the first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0560] 79. The method according to Note 77, wherein the first indication information is used by the terminal device to determine whether to stop the AI / ML model or model group in the terminal device.

[0561] 80. The method according to Note 79, wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the terminal device determines to stop the AI / ML model in the terminal device.

[0562] 81. The method according to Note 76, wherein the first network device sends second indication information for updating the AI / ML model to one or more terminal devices within a cell or an area through system information and / or paging messages.

[0563] 82. The method according to Note 81, wherein the second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

[0564] 83. The method according to Note 81, wherein the second indication information is used by the terminal device to determine whether to update the AI / ML model or model group in the terminal device.

[0565] 84. The method according to Note 83, wherein, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the terminal device determines to update the AI / ML model in the terminal device.

[0566] 85. The method according to Note 76, wherein a terminal device in an inactive state obtains identification information of an AI / ML model and / or version information of an AI / ML model from system information, and when the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0567] 86. The method according to Note 85, wherein the terminal device downloads the updated AI / ML model from the cell, or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0568] 87. The method according to Note 76, wherein the terminal device in an activated state obtains identification information of the AI / ML model and / or version information of the AI / ML model from the target cell, and when the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0569] 88. The method according to Note 87, wherein the terminal device downloads the updated AI / ML model from the target cell, or loads the updated AI / ML model from the memory, or obtains the updated AI / ML model from the system information.

[0570] 89. The method according to Supplement 76, wherein the method further comprises:

[0571] The terminal device sends a request to the first network device; and

[0572] The terminal device receives system information related to the AI / ML model.

[0573] 90. The method according to Note 89, wherein the terminal device obtains identification information of the AI / ML model and / or version information of the AI / ML model from the system information, and when the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model in the terminal device.

[0574] 91. A method according to Note 89, wherein the system information is cell-specific or area-specific.

[0575] 92. The method according to any one of Notes 76 to 91, further comprising:

[0576] The terminal device also receives the exclusion list and / or the allow list sent by the first network device.

[0577] 93. An information processing method comprising:

[0578] The network device receives the model identification information sent by the terminal device and the model performance information and / or model operation information corresponding to the model identification information.

[0579] 94. The method according to Note 93, wherein the model performance information is generated by the terminal device monitoring the AI / ML model.

[0580] 95. The method according to Note 94, wherein the terminal device monitors the output performance of the AI / ML model, and / or the terminal device monitors the input and output data distribution of the AI / ML model.

[0581] 96. The method according to Note 93, wherein the model operation information is determined by the terminal device based on the monitoring results of the AI / ML model.

[0582] 97. A method according to any one of Notes 94 to 96, wherein the monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0583] 98. A method according to any one of Notes 93 to 97, wherein the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model rollback information, and model failure to recovery information.

[0584] 99. A method according to any one of Notes 93 to 98, wherein the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through at least one or any combination of uplink control information (UCI), MAC CE or RRC signaling.

[0585] 100. A method according to any one of Notes 93 to 98, wherein the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through an uplink control channel and / or an uplink data channel.

[0586] 101. The method according to any one of Notes 93 to 100, wherein the method further comprises:

[0587] The network device sends configuration information or instruction information to the terminal device to schedule the terminal device to send the model identification information and the model performance information and / or model operation information corresponding to the model identification information.

[0588] 102. The method according to Note 101, wherein the configuration information or indication information is dedicated information for the AI / ML model.

[0589] 103. A method according to any one of Notes 93 to 102, wherein the terminal device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information through periodic reporting, semi-continuous reporting or non-periodic reporting.

[0590] 104. The method according to any one of Notes 93 to 103, wherein the AI / ML model is monitored by the terminal device, and the AI / ML model is a model deployed on the first network device or a bilateral model.

[0591] 105. The method according to any one of Notes 93 to 104, wherein the method further comprises:

[0592] The network device sends request information for requesting monitoring data to the terminal device.

[0593] 106. The method according to Note 105, wherein the request information includes at least one of the following: performance indicators, data distribution indicators, and model operation information after monitoring.

[0594] 107. The method according to Note 105, wherein the monitoring data is historical information, and the request information includes time range indication information.

[0595] 108. The method according to Note 105, wherein the request information includes a model identifier.

[0596] 109. An information processing method comprising:

[0597] The network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device or the core network device or other network devices.

[0598] 110. The method according to Note 109, wherein the model performance information is generated by the network device monitoring the AI / ML model.

[0599] 111. The method according to Note 110, wherein the network device performs the monitoring on the output performance of the AI / ML model, and / or the network device performs the monitoring on the input and output data distribution of the AI / ML model.

[0600] 112. The method according to Note 109, wherein the model operation information is determined by the network device based on the monitoring results of the AI / ML model.

[0601] 113. The method according to any one of Notes 110 to 112, wherein the monitoring result is related to the output performance of the AI / ML model or the input and output data distribution of the AI / ML model, and is represented by two or more levels.

[0602] 114. A method according to any one of Notes 109 to 113, wherein the model operation information includes at least one of the following: model failure information, model activation information, model deactivation information, model switching information, model continued use information, model discontinued use information, model fallback information, and model failure to recovery information.

[0603] 115. A method according to any one of Notes 109 to 114, wherein the network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device through at least one or any combination of downlink control information (DCI), MAC CE or RRC signaling.

[0604] 116. A method according to any one of Notes 109 to 115, wherein the network device sends the model identification information and the model performance information and / or model operation information corresponding to the model identification information to the terminal device through a downlink control channel and / or a downlink data channel.

[0605] 117. The method according to any one of Notes 109 to 116, wherein the AI / ML model is monitored by the network device, and the AI / ML model is a model deployed on the terminal device or a bilateral model.

[0606] 118. The method according to any one of Notes 109 to 117, wherein the method further comprises:

[0607] The network device receives the request information sent by the terminal device for requesting monitoring data.

[0608] 119. The method according to Note 118, wherein the request information includes at least one of the following: performance indicators, data distribution indicators, and model operation information after monitoring.

[0609] 120. The method according to Note 118, wherein the monitoring data is historical information, and the request information includes time range indication information.

[0610] 121. The method according to Note 118, wherein the request information includes a model identifier.

[0611] 122. An information processing method comprising:

[0612] The network device sends indicator information related to model monitoring and / or model operation decision information based on model monitoring to the terminal device.

[0613] 123. The method according to Note 122, wherein the indicator information related to model monitoring includes at least one of the following: performance indicator information related to model output, data distribution indicator information related to model output, and data distribution indicator information related to model input.

[0614] 124. The method according to Note 122 or 123, wherein the indicator information related to model monitoring is model identification-specific information, or is model monitoring general information.

[0615] 125. The method according to Note 122, wherein the model operation decision information based on model monitoring includes at least one of the following: information related to determining model deactivation, information related to determining model rollback, information related to determining model switching, information related to determining model stoppage, information related to determining model failure, information related to determining model switching from stop to enable, and information related to determining model update.

[0616] 126. The method according to Note 122 or 125, wherein the model operation decision information based on model monitoring is model identification-specific information, or is model monitoring general information.

[0617] 127. A method according to any one of Notes 122 to 126, wherein the network device and / or the terminal device operates the model based on the results of model monitoring.

[0618] 128. A method according to any one of Notes 122 to 126, wherein the terminal device decides on the operation of the model based on the results of model monitoring and reports the operation to the network device; after receiving the confirmation instruction from the network device, the terminal device performs the operation on the model.

[0619] 129. A method according to any one of Notes 122 to 126, wherein the terminal device decides on an operation to be performed on the model based on a result of model monitoring, and reports the operation to the network device, and the network device performs the operation on the model.

[0620] 130. An information processing method comprising:

[0621] The network device sends a model query message to the terminal device; and

[0622] The network device receives the model feedback information sent by the terminal device and performs corresponding operations on the AI / ML model according to the model feedback information.

[0623] 131. The method according to Note 130, wherein the model query information includes a model identifier; the model feedback information includes at least one of the following: AI / ML model availability information corresponding to the model identifier, AI / ML model deactivation information corresponding to the model identifier, AI / ML model fallback information corresponding to the model identifier, and AI / ML model failure information corresponding to the model identifier.

[0624] 132. The method according to Note 131, wherein the network device continues to use the AI / ML model corresponding to the model identifier, or activates the AI / ML model corresponding to the model identifier, or deactivates the AI / ML model corresponding to the model identifier.

[0625] 133. The method according to Note 130, wherein the model query information includes multiple model identifiers; and the model feedback information includes a model identifier selected by the terminal device from the multiple model identifiers.

[0626] 134. The method according to Note 133, wherein the network device activates the AI / ML model corresponding to the selected model identifier.

[0627] 135. A network device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the information indication method as described in any one of Notes 1 to 47 or the information processing method as described in any one of Notes 48 to 61 or the information processing method as described in any one of Notes 93 to 134.

[0628] 136. A core network device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the information processing method as described in any one of Notes 62 to 75.

[0629] 137. A terminal device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to implement the information processing method as described in any one of Notes 76 to 92.

Claims

1. An information indicating device, comprising: a determination unit that determines whether to stop, update, or switch an AI / ML model or model group for a certain signal processing function; as well as A sending unit, which sends indication information to the second network device, the terminal device, or the core network device when it is determined to stop, update, or switch the AI / ML model or model group.

2. The device according to claim 1, wherein Sending first indication information for stopping the AI / ML model or model group, and / or second indication information for updating the AI / ML model or model group to the second network device through the Xn interface.

3. The device according to claim 2, wherein The first indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message or a resource status update message; The first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

4. The device according to claim 2, wherein The first indication information is used by the second network device to determine whether to stop the AI / ML model or model group in the second network device, and / or to determine whether to stop the AI / ML model or model group in a terminal device within a cell; In which, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the second network device, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the second network device determines to stop the AI / ML model in the second network device; and / or when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the second network device notifies the terminal device to stop the AI / ML model in the terminal device.

5. The device according to claim 2, wherein The second indication information is carried by an XnAP message; the XnAP message includes: a configuration update message, an Xn establishment message or a resource status update message; The second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, and version information of the original AI / ML model.

6. The device according to claim 2, wherein The second indication information is used by the second network device to determine whether to update the AI / ML model or model group in the second network device, and / or to determine whether to update the AI / ML model or model group in a terminal device within a cell; In which, when the identification information of the signal processing function corresponding to the AI / ML model in the second network device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device determines to update the AI / ML model in the second network device; and / or When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the cell is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the second network device notifies the terminal device to update the AI / ML model in the terminal device.

7. The device according to claim 1, wherein Sending first indication information for stopping the AI / ML model or model group, and / or second indication information for updating the AI / ML model or model group to the core network device through the NG interface.

8. The device according to claim 7, wherein The first indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message; The first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

9. The device according to claim 7, wherein The first indication information is used by the core network device to determine whether to stop the AI / ML model or model group in the network devices and / or terminal devices in a region; Wherein, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the network device within the area, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the network device through an NGAP message to stop the AI / ML model or model group in the network device; and / or When at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area, the identification information of the AI / ML model, the identification information of the AI / ML model group, the intra-group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the core network device notifies the terminal device through a NAS message to stop the AI / ML model or model group in the terminal device.

10. The device according to claim 7, wherein The second indication information is carried by an NGAP message; the NGAP message includes: a RAN configuration update message, an NG establishment message, an Xn establishment message or a RAN configuration transfer message; The second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, and version information of the original AI / ML model.

11. The device according to claim 7, wherein The second indication information is used by the core network device to determine whether to update the AI / ML model or model group in the network devices and / or terminal devices in a region; In which, when the identification information of the signal processing function corresponding to the AI / ML model in the network device in the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the network device through an NGAP message to update the AI / ML model in the network device; and / or When the identification information of the signal processing function corresponding to the AI / ML model in the terminal device within the area is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the core network device notifies the terminal device through a NAS message to update the AI / ML model in the terminal device.

12. The device according to claim 1, wherein First indication information for stopping the AI / ML model or model group, and / or second indication information for updating the AI / ML model or model group are sent to one or more terminal devices in a cell or an area through system information and / or paging messages.

13. The device according to claim 12, wherein The first indication information includes at least one of the following: identification information of the signal processing function, identification information of the AI / ML model, version information of the AI / ML model, identification information of the AI / ML model group, intra-group identification information of the AI / ML model group, type information of the AI / ML model, storage size information of the AI / ML model or model group, exclusion list and / or allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

14. The device according to claim 12, wherein The first indication information is used by the terminal device to determine whether to stop the AI / ML model or model group in the terminal device; In which, when at least one of the identification information of the signal processing function corresponding to the AI / ML model in the terminal device, the identification information of the AI / ML model, the identification information of the AI / ML model group, the group identification information of the AI / ML model group, and the version information of the AI / ML model is consistent with the first indication information, the terminal device determines to stop the AI / ML model in the terminal device.

15. The device according to claim 12, wherein The second indication information includes at least one of the following: identification information of the signal processing function, identification information of the new AI / ML model, identification information of the new AI / ML model group, intra-group identification information of the new AI / ML model group, version information of the new AI / ML model, identification information of the original AI / ML model, identification information of the original AI / ML model group, intra-group identification information of the original AI / ML model group, version information of the original AI / ML model, an exclusion list and / or an allowance list of the model, or model identification information and model performance information and / or model operation information corresponding to the model identification information.

16. The device according to claim 12, wherein The second indication information is used by the terminal device to determine whether to update the AI / ML model or model group in the terminal device; In which, when the identification information of the signal processing function corresponding to the AI / ML model in the terminal device is consistent with the second indication information, and the identification information of the AI / ML model is inconsistent with the second indication information and / or the version information of the AI / ML model is lower than the version information of the AI / ML model in the second indication information, the terminal device determines to update the AI / ML model in the terminal device.

17. The device according to claim 1, wherein The terminal device in an inactive state obtains identification information of the AI / ML model and / or version information of the AI / ML model from the system information. If the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model or model group in the terminal device; The terminal device downloads the updated AI / ML model or model group from the cell, or loads the updated AI / ML model or model group from the memory, or obtains the updated AI / ML model or model group from the system information.

18. The device according to claim 1, wherein The terminal device in the activated state obtains identification information of the AI / ML model and / or version information of the AI / ML model from the target cell. If the identification information of the AI / ML model in the terminal device is inconsistent with the obtained identification information and / or the version information of the AI / ML model is lower than the obtained version information, the terminal device determines to update the AI / ML model or model group in the terminal device; The terminal device downloads the updated AI / ML model or model group from the target cell, or loads the updated AI / ML model or model group from the memory, or obtains the updated AI / ML model or model group from the system information.

19. An information processing device comprising: a receiving unit configured to receive instruction information sent by the first network device; as well as A processing unit that stops, updates, or switches an AI / ML model or model group for a certain signal processing function according to the indication information.

20. A communication system comprising: Network equipment that determines whether to stop, update, or switch an AI / ML model or model group for a signal processing function; And when it is determined to stop, update or switch the AI / ML model or model group, an indication message is sent to the second network device or the terminal device or the core network device.