AI / ML control method and apparatus

Abstract

Embodiments of the present application provide an AI / ML control method and apparatus. The method comprises: a terminal device receiving indication information sent by a network device, the indication information being used to indicate activation / deactivation / switching / backoff of an AI / ML function / model; and / or the terminal device reporting reporting information to the network device, the reporting information being used to determine activation / deactivation / switching / backoff of the AI / ML function / model.

Description

AI / ML control method and device Technical Field

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

[0002] NR Release 18 investigates artificial intelligence / machine learning (AI / ML) over the air interface. AI / ML can be used for the following use cases: channel state information (CSI) feedback enhancement, beam management, and positioning enhancement. CSI feedback enhancement can include CSI prediction and CSI compression; beam management can include spatial beam prediction and temporal beam prediction; and positioning enhancement can include direct positioning and AI / ML-assisted positioning.

[0003] In some sub-use cases, a bilateral model can be used, with the AI / ML model located on both the terminal device side and the network device side. For example, CSI compression can be a representative use case for a bilateral model. In other sub-use cases, a unilateral model can be used, with the AI / ML model located either on the terminal device side or the network device side.

[0004] 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.

[0005] Summary of the Invention

[0006] The inventors found that there is currently no specific solution for controlling the activation / deactivation / switching / rollback of AI / ML functions / models.

[0007] To address at least one of the above problems, embodiments of the present application provide an AI / ML control method and device.

[0008] According to one aspect of an embodiment of the present application, an AI / ML control method is provided, including:

[0009] The terminal device receives instruction information sent by the network device to indicate activation / deactivation / switching / rollback of AI / ML functionality / model; and / or

[0010] The terminal device reports reporting information to the network device for determining activation / deactivation / switching / rollback of AI / ML functions / models.

[0011] According to another aspect of an embodiment of the present application, an AIML control device is provided, including:

[0012] a receiving unit configured to receive indication information sent by a network device indicating activation / deactivation / switching / fallback of an AI / ML functionality / model; and / or

[0013] A sending unit reports reporting information used to determine activation / deactivation / switching / rollback of an AI / ML function / model to the network device.

[0014] According to another aspect of an embodiment of the present application, an AI / ML control method is provided, including:

[0015] The network device sends an indication message to the terminal device indicating activation / deactivation / switching / fallback of an AI / ML functionality / model; and / or

[0016] The network device receives reporting information reported by the terminal device for determining activation / deactivation / switching / rollback of AI / ML functions / models.

[0017] According to another aspect of an embodiment of the present application, an AIML control device is provided, including:

[0018] A sending unit that sends instruction information for indicating activation / deactivation / switching / fallback of AI / ML functionality / model to the terminal device; and / or

[0019] A receiving unit receives reporting information reported by the terminal device for determining activation / deactivation / switching / rollback of AI / ML functions / models.

[0020] According to another aspect of an embodiment of the present application, a communication system is provided, including:

[0021] A terminal device that receives indication information sent by a network device to indicate activation / deactivation / switching / fallback of an AI / ML function / model, and / or reports reporting information to the network device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

[0022] One of the beneficial effects of the embodiments of the present application is that the activation / deactivation / switching / fallback of AI / ML functions / models can be agreed upon between network devices and terminal devices, thereby improving the accuracy and reliability of AI / ML.

[0023] 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.

[0024] 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.

[0025] 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

[0026] 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.

[0027] FIG1 is a schematic diagram of a communication system according to an embodiment of the present application;

[0028] FIG2 is a schematic diagram of an AI / ML control method according to an embodiment of the present application;

[0029] FIG3 is another schematic diagram of the AI / ML control method according to an embodiment of the present application;

[0030] FIG4 is an example diagram of an indication via a MAC CE according to an embodiment of the present application;

[0031] FIG5 is an example diagram of an indication through DCI according to an embodiment of the present application;

[0032] FIG6 is an example diagram of RRC+MAC CE+DCI according to an embodiment of the present application;

[0033] FIG7 is an example diagram of RRC+DCI according to an embodiment of the present application;

[0034] FIG8 is another schematic diagram of the AI / ML control method according to an embodiment of the present application;

[0035] FIG9 is an example diagram of a terminal device decision and a network device confirmation according to an embodiment of the present application;

[0036] FIG10 is an example diagram of a network device decision and terminal device confirmation according to an embodiment of the present application;

[0037] FIG11 is another schematic diagram of the resource configuration method according to an embodiment of the present application;

[0038] FIG12 is a schematic diagram of an AI / ML control device according to an embodiment of the present application;

[0039] FIG13 is another schematic diagram of the AI / ML control device according to an embodiment of the present application;

[0040] FIG14 is a schematic diagram of a terminal device according to an embodiment of the present application;

[0041] FIG15 is a schematic diagram of a network device according to an embodiment of the present application. DETAILED DESCRIPTION

[0042] 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.

[0043] 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.

[0044] 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.

[0045] 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.

[0046] 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.

[0047] 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.

[0048] Among them, base stations may include but are not limited to: NodeB (NodeB or NB), evolved NodeB (eNodeB or eNB) and 5G base station (gNB), IAB host, etc., and may also include remote radio head (RRH, Remote Radio Head), remote radio unit (RRU, Remote Radio Unit), relay (relay) or low-power node (such as femeto, pico, etc.). The term "base station" can include some or all of their functions. Each base station can provide communication coverage for a specific geographical area. The term "cell" can refer to a base station and / or its coverage area, depending on the context in which the term is used.

[0049] 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.

[0050] 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.

[0051] 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.

[0052] 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.

[0053] The following describes the scenarios of the embodiments of the present application through examples, but the present application is not limited thereto.

[0054] 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.

[0055] 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.

[0056] 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.

[0057] 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.

[0058] In embodiments of the present application, one or more AI / ML models may be configured and run in network devices and / or terminal devices. These AI / ML models may be used for various signal processing functions in wireless communications, such as CSI prediction, CSI compression, beamforming, and positioning management; however, the present application is not limited thereto. Furthermore, in this application, " / " represents "and / or" or alternatively.

[0059] Embodiments of the first aspect

[0060] The embodiment of the present application provides an AI / ML control method, which is described from the perspective of a terminal device.

[0061] FIG2 is a schematic diagram of an AI / ML control method according to an embodiment of the present application. As shown in FIG2 , the method includes:

[0062] 201, the terminal device receives instruction information sent by the network device for indicating activation / deactivation / switching / fallback of AI / ML functionality / model; and / or

[0063] 202. The terminal device reports reporting information to the network device for determining activation / deactivation / switching / rollback of AI / ML functions / models.

[0064] 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.

[0065] In some embodiments, functionality refers to an AI / ML feature / feature group enabled by a configuration, where the configuration is supported based on conditions indicated by UE capabilities.

[0066] For example, the AL / ML function may be one or more functions, or one or more logical models, or one or more sub-functions, or one or more features, or one or more feature groups.

[0067] For another example, the function can be to use AI / ML for spatial beam prediction, or to use AI / ML for time beam prediction, or to use AI / ML for CSI prediction, or to use AI / ML for direct positioning, or to use AI / ML for assisted positioning, and so on.

[0068] In some embodiments, indication information including activation / deactivation / switching / fallback of one or more AI / ML functionalities / models is sent by the network device to the terminal device.

[0069] FIG3 is another schematic diagram of the AI / ML control method according to an embodiment of the present application. As shown in FIG3 , the method includes:

[0070] 301. The terminal device receives indication information sent by the network device for indicating activation / deactivation / switching / rollback of one or more AI / ML functions / models.

[0071] 302. The terminal device sends feedback information corresponding to the indication information to the network device.

[0072] 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.

[0073] In some embodiments, the indication information is carried by MAC CE.

[0074] For example, the activation / deactivation / switching / fallback of one or more AI / ML functions / models may be performed by a MAC CE, and the network device may instruct the terminal device to activate / deactivate / switching / fallback one or more AI / ML functions / models through the MAC CE. For example, the MAC-CE may be newly defined, or an existing MAC CE may be reused.

[0075] In some embodiments, a set of AI / ML functionalities / models are configured via radio resource control (RRC), and the MAC CE indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models in the configured set of AI / ML functionalities / models.

[0076] For example, the gNB can configure a group / list of AI / ML functions / models via RRC signaling. Subsequently, the gNB can indicate the activation / deactivation / switching / fallback of one or more AI / ML functions / models in the group / list configured by RRC via MAC CE.

[0077] In some embodiments, the terminal device sends an ACK corresponding to the PDSCH to the network device in time slot n after receiving the PDSCH carrying the MAC CE; wherein the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability. In addition, ΔT may also include the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0078] Figure 4 is an example diagram of an embodiment of the present application indicated by a MAC CE. As shown in Figure 4, the UE receives a MAC CE in a certain time slot (as shown in Figure 4 A), and the UE sends HARQ-ACK (feedback information) in time slot n corresponding to the PDSCH carrying the MAC CE. The MAC CE indicates the activation / deactivation / switching / fallback of one or more AI / ML functions / models.

[0079] As shown in Figure 4 , for example, the activation / deactivation / handover / fallback indicated by the MAC CE applies starting from the first time slot after n+ΔT (as shown in B in Figure 4 ). T is greater than or equal to ΔT, where ΔT can be predefined or configurable, or depend on UE capabilities. The unit of ΔT can be time slots or other time units such as symbols.

[0080] In another example, ΔT may also include the preparation time / interruption time required for the UE to perform switching between functions / models, which may also depend on the capabilities of the UE, and the value may be zero (a zero value means that no preparation time / disconnection time is required for function / model switching).

[0081] In some embodiments, the indication information is carried by downlink control information (DCI).

[0082] For example, the activation / deactivation / switching / fallback of one or more AI / ML functions / models may be performed by DCI, and the network device may instruct the terminal device to activate / deactivate / switching / fallback one or more AI / ML functions / models through the DCI.

[0083] In some embodiments, the indication information is a field in the DCI, or an existing field in the DCI, or one or more codepoints in the DCI, or a DCI format, or a DCI scrambled by an RNTI.

[0084] In some embodiments, the DCI is used to schedule PDSCH / PUSCH, or the DCI is not used to schedule PDSCH / PUSCH; the DCI is used for one or more serving cells.

[0085] For example, a new field may be added to an existing DCI format to indicate the activation / deactivation / switching / fallback of one or more AI / MI functions / models. The DCI may be a DCI format for scheduling PDSCH or PUSCH, such as DCI format 0_1 / 0_2 / 0_3 / 1_1 / 1_2 / 1_3. The gNB may send DCI with or without scheduling PDSCH / PUSCH. The DCI indicating the activation / deactivation / switching / fallback of one or more AI / MI functions / models may apply to one serving cell or to multiple serving cells. In addition, a new RNTI may be introduced, and DCI scrambled with the new RNTI may be used to activate / deactivate / switching / fallback of one or more AI / MI functions / models.

[0086] For another example, an unused field or a reserved codepoint in an existing DCI format can be reused to indicate the activation / deactivation / switching / fallback of one or more AI / MI functions / models. In one example, the DCI may be a DCI format that does not schedule PDSCH / PUSCH, for example, DCI format 0_1 / 0_2 / 0_3 that does not schedule PUSCH or DCI format 1_1 / 1_2 / 1_3 that does not schedule PDSCH. The DCI indication of activation / deactivation / switching / fallback of one or more AI / MI functions / models may be applied to one serving cell, or may be applied to multiple serving cells. In addition, a new RNTI may be introduced, and the DCI scrambled by the new RNTI may be used for activation / deactivation / switching / fallback of one or more AI / MI functions / models.

[0087] For another example, a new DCI format may be introduced to indicate the activation / deactivation / switching / fallback of one or more AI / MI functions / models. Accordingly, a corresponding new RNTI may be introduced. The DCI indication of activation / deactivation / switching / fallback of one or more AI / MI functions / models may be applied to one serving cell or to multiple serving cells.

[0088] In some embodiments, the DCI is directed to one or more terminal devices.

[0089] For example, the DCI may indicate activation / deactivation / switching / fallback of one or more AI / ML functions / models for multiple UEs (i.e., a group of UEs). A new DCI format applicable to one or a group of UEs may be introduced, and a corresponding RNTI may be introduced. For example, a new DCI format 2_10 may be defined. The DCI indication of activation / deactivation / switching / fallback of one or more AI / ML functions / models may be applied to one serving cell, or may be applied to multiple serving cells.

[0090] In some embodiments, one or more fields in the DCI are used for validation; if the validation passes, the DCI is considered valid; if the validation fails, the DCI is discarded.

[0091] For example, in a DCI indicating activation / deactivation / switching / fallback of one or more AI / MM functions / models, a combination of specific values ​​of one or more fields can be used to verify the DCI. If the verification is successful, the UE considers the information in the DCI to be a valid command. If the verification is unsuccessful, the UE discards all information in the DCI. Specific values ​​of one or more fields used for verification can be predefined, such as specific values ​​for MCS and NDI.

[0092] In some embodiments, after receiving the DCI, the terminal device sends an ACK / NACK to the network device to indicate whether the DCI is correctly received. For example, the ACK / NACK is sent via PUCCH and / or PUSCH.

[0093] In some embodiments, the terminal device sends an ACK corresponding to the DCI in time slot n after correctly receiving the DCI; the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capabilities. In addition, ΔT may also include the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0094] Figure 5 is an example diagram of an embodiment of the present application using DCI. As shown in Figure 5, the UE receives DCI in a certain time slot (as shown in Figure 5 A), and the UE sends HARQ-ACK in the time slot n corresponding to the DCI. The DCI indicates the activation / deactivation / switching / fallback of one or more AI / ML functions / models.

[0095] As shown in Figure 5 , for example, the activation / deactivation / handover / fallback indicated by the DCI applies starting from the first time slot after n+ΔT (as shown in B in Figure 5 ). T is greater than or equal to ΔT, where ΔT can be predefined or configurable, or depend on UE capabilities. The unit of ΔT can be time slots or other time units such as symbols.

[0096] In another example, ΔT may also include the preparation time / interruption time required for the UE to perform switching between functions / models, which may also depend on the capabilities of the UE, and the value may be zero (a zero value means that no preparation time / disconnection time is required for function / model switching).

[0097] In some embodiments, a set of AI / ML functionalities / models is configured by radio resource control (RRC), a subset of the configured set of AI / ML functionalities / models is selected by a MAC CE, and the DCI indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models in the selected subset.

[0098] Figure 6 is an example diagram of RRC+MAC CE+DCI according to an embodiment of the present application. As shown in Figure 6 , for example, the gNB can configure a group / list of AI / ML functions / models via RRC signaling, as shown in Figure 6 as AI / ML functions / models #1 to #M.

[0099] As shown in Figure 6 , a subset of AI / ML functions / models may be selected by the MAC CE, as shown in Figure 6 as AI / ML functions / models #K to AI / ML functions / models #N. The MAC CE may be newly defined. In one example, the MAC CE may select one or more AI / ML functions / models as candidate functions / models, and the determination of the candidate functions / models may be based on performance monitoring of the AI / ML functions / models.

[0100] As shown in Figure 6, the DCI can indicate the activation / deactivation / switching / fallback of one or more AI / ML functions / models within the subset selected by the MAC CE (such as AI / ML function / model #X and AI / ML function / model #Y in Figure 6), for example, AI / ML function / model #X and AI / ML function / model #Y are activated by the DCI.

[0101] In some embodiments, the subset is selected based on performance monitoring of the AI / ML functionality / model; the length of the field in the DCI indicating activation / deactivation / switching / fallback of the AI / ML functionality / model is determined by the number of AI / ML functionality / models in the subset selected by the MAC CE.

[0102] For example, in the DCI, the length of the field indicating activation / deactivation / handover / fallback may be determined by the number of AI / ML functions / models selected by the MAC CE. For example, if the number of AI / ML functions / models selected by the MAC CE is N, the length of the field is determined to be In another example, the DCI field may be a bitmap; if the number of AI / ML functions / models selected by the MAC CE is N, the length of the field is N.

[0103] In some embodiments, a set of AI / ML functionalities / models are configured by radio resource control (RRC), and the DCI indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models among the configured AI / ML functionalities / models.

[0104] FIG7 is an example diagram of RRC+DCI according to an embodiment of the present application. As shown in FIG7 , for example, the gNB may configure a group / list of AI / ML functions / models via RRC signaling, as shown in FIG7 as AI / ML functions / models #1 to #M.

[0105] As shown in Figure 7, DCI can indicate the activation / deactivation / switching / fallback of one or more AI / MI functions / models within the group / list configured by RRC (such as AI / ML function / model #X and AI / ML function / model #Y in Figure 7), for example, AI / ML function / model #X and AI / ML function / model #Y are activated by the DCI.

[0106] In some embodiments, the length of the field in the DCI indicating activation / deactivation / switching / fallback of the AI / ML functionality / model is determined by the number of AI / ML functionality / models configured by the RRC.

[0107] For example, in the DCI, the length of the field indicating activation / deactivation / handover / fallback may be determined by the number of AI / ML functions / models configured by RRC. For example, if the number of AI / ML functions / models configured by RRC is N, the length of the field is determined to be In another example, the DCI field may be a bitmap; if the number of AI / ML functions / models configured by RRC is N, the length of the field is N.

[0108] In some embodiments, the indication information is carried by radio resource control (RRC).

[0109] For example, activation / deactivation / switching / fallback of one or more AI / ML functions / models may be performed via RRC signaling. Upon receiving the corresponding RRC message, the activation / deactivation / switching / fallback of the one or more AI / ML functions / models is applied or takes effect.

[0110] The above schematically illustrates the case where the activation / deactivation / switching / fallback of an AI / ML function / model is initiated by a network device. The following schematically illustrates the case where the activation / deactivation / switching / fallback of an AI / ML function / model is initiated by a terminal device.

[0111] In some embodiments, reporting information including activation / deactivation / switching / fallback of one or more AI / ML functionalities / models is sent by the terminal device to the network device.

[0112] FIG8 is another schematic diagram of the AI / ML control method according to an embodiment of the present application. As shown in FIG8 , the method includes:

[0113] 801, the terminal device sends reporting information to the network device for determining activation / deactivation / switching / fallback of one or more AI / ML functions / models;

[0114] 802. The terminal device receives feedback information corresponding to the reported information sent by the network device.

[0115] It is worth noting that FIG8 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 FIG8 above.

[0116] The following first schematically illustrates the case where the decision is made by the terminal device.

[0117] In some embodiments, the activation / deactivation / switching / fallback of one or more AI / ML functions / models is determined by the terminal device, and the reporting information includes the activation / deactivation / switching / fallback of one or more AI / ML functions / models determined by the terminal device.

[0118] For example, the UE can decide the activation / deactivation / switching / fallback of one or more AI / MI functions / models, and can report the activation / deactivation / switching / fallback of the one or more AI / MI functions / models to the network device, that is, the control of the function / model is on the UE side.

[0119] For example, the network (e.g., gNB) may provide performance metrics, thresholds, performance criteria, etc. as guidance, parameters, or basis for the UE's decision / judgment / decision. For example, one or more of an event trigger, performance metric, timer, or counter may be configured by the network device for the terminal device.

[0120] For another example, the UE's decision / judgment / decision is made autonomously by the UE. The UE can independently decide to activate / deactivate / switch / fallback one or more AI / ML functionalities / models. The UE can also report to the network device for final decision or confirmation.

[0121] In some examples, the UE's decision to activate / deactivate / switch / fallback one or more AI / ML functions / models can be based on event triggering, function / model performance monitoring / evaluation, or condition changes. Some examples are as follows:

[0122] - One or more AI / MI functions / models have failed based on function / model performance monitoring;

[0123] - Based on function / model performance monitoring, obtain applicable / candidate / recommended / preferred AI / ML functions / models or select AI / ML functions / models;

[0124] - Obtain applicable / candidate / recommended / preferred AI / ML functions / models or select AI / ML functions / models based on changing conditions, such as scene changes, cell changes, UE speed changes, UE mobility, etc.;

[0125] - Make UE decisions based on hardware conditions, such as insufficient power, insufficient computing power, overheating, insufficient memory, etc.

[0126] - Model updated, and request to switch to the new model.

[0127] In some embodiments, the reported information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling, for example, by any one of RRC, MAC CE, and DCI, or by any combination of the three.

[0128] For example, the UE's decision to activate / deactivate / switching / fallback one or more AI / MM functions / models may be reported via an RRC message. For example, the RRC message may be one of the following:

[0129] -UAI(UEAssistanceInformation);

[0130] -RRCReconfigurationComplete message;

[0131] -RRC Reestablishment Request message (RRCReestablishmentRequest message); for example, a new cause value of RestablishmentCause may be introduced, such as AI / ML function / model failure, or activation / deactivation / switching / fallback of AI / ML function / model;

[0132] -Added RRC message.

[0133] For another example, the UE's decision to activate / deactivate / switching / fallback one or more AI / MM functions / models may be reported via a MAC CE. In one example, the MAC CE may be newly defined; in another example, an existing MAC CE may be reused.

[0134] For another example, the UE's decision to activate / deactivate / switching / fallback one or more AI / MM functions / models can be reported via physical layer signaling. The physical layer signaling can be transmitted via PRACH, PUCCH, or PUSCH. For example, it can be existing UCI or a newly defined UCI.

[0135] For another example, the reporting information can be a combination of L1 (layer 1) and L2 (layer 2) signaling, such as physical layer signaling and MAC CE. For example, if there are no uplink resources for transmission, the UE may transmit an SR-like PUCCH (which may be a dedicated SR-like PUCCH resource) after one or more AI / ML functions / models fail. After receiving the SR-like PUCCH, the gNB may transmit a DCI containing an uplink grant for PUSCH transmission. The UE then transmits a MAC CE, which may include information about which AI / ML function / model failed.

[0136] In some embodiments, the gNB may optionally provide feedback regarding activation / deactivation / switching / fallback of one or more AI / MI functions / models. The terminal device receives the confirmation information provided by the network device; the confirmation information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0137] For example, if the UE reports a decision to activate / deactivate / handover / fallback one or more AI / MM functions / modes, the gNB may send back a confirmation (or acknowledgment) via an RRC message, MAC CE, or DCI. In one example, a HARQ-ACK from the gNB (corresponding to the PUSCH carrying the reporting information) may be considered as the gNB's acknowledgment. In another example, an uplink grant scheduling a new transmission for the same HARQ process as the PUSCH carrying the reporting information may be considered as the gNB's acknowledgment.

[0138] In some embodiments, the network device sends a confirmation message at time slot n; wherein the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect at the first time slot after n+ΔT, where ΔT is a predefined value, a configured value, or a value dependent on the terminal capabilities. ΔT may also include the preparation time or interruption time required by the terminal device to perform the functionality / model switching.

[0139] Figure 9 illustrates an example of a terminal device decision and network device confirmation according to an embodiment of the present application. As shown in Figure 9 , the UE sends reporting information in a certain time slot (as shown in Figure 9 A), and the gNB sends confirmation information in the corresponding time slot n. This reporting information indicates the activation / deactivation / switching / fallback of one or more AI / ML functions / models determined by the UE.

[0140] As shown in Figure 9 , for example, the activation / deactivation / handover / fallback indicated by the reported information applies starting from the first time slot after n+ΔT (as shown in B in Figure 9 ). T is greater than or equal to ΔT, where ΔT can be predefined or configurable, or depend on UE capabilities. The unit of ΔT can be time slots or other time units such as symbols.

[0141] In another example, ΔT may also include the preparation time / interruption time required for the UE to perform switching between functions / models, which may also depend on the capabilities of the UE, and the value may be zero (a zero value means that no preparation time / disconnection time is required for function / model switching).

[0142] The following schematically illustrates the situation where the decision is made by the network device.

[0143] In some embodiments, the network device determines the activation / deactivation / switching / fallback of the one or more AI / ML functions / models, and the reported information includes the activation / deactivation / switching / fallback of one or more AI / ML functions / models selected, candidated, recommended, or preferred by the terminal device.

[0144] For example, the gNB can decide the activation / deactivation / switching / fallback of one or more AI / MI functions / models, and can indicate the activation / deactivation / switching / fallback of the one or more AI / MI functions / models to the terminal device, that is, the control of the function / model is on the gNB side.

[0145] In some examples, the UE's activation / deactivation / switching / fallback of one or more selected, applicable, candidate, recommended, or preferred AI / ML functions / models may be based on event triggering, function / model performance monitoring / evaluation, or conditional changes. Some examples are as follows:

[0146] - One or more AI / MI functions / models have failed based on function / model performance monitoring;

[0147] - Based on function / model performance monitoring, obtain applicable / candidate / recommended / preferred AI / ML functions / models or select AI / ML functions / models;

[0148] - Obtain applicable / candidate / recommended / preferred AI / ML functions / models or select AI / ML functions / models based on changing conditions, such as scene changes, cell changes, UE speed changes, UE mobility, etc.;

[0149] - Make UE decisions based on hardware conditions, such as insufficient power, insufficient computing power, overheating, insufficient memory, etc.

[0150] - Model updated, and request to switch to the new model.

[0151] In some embodiments, the reported information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling, for example, by any one of RRC, MAC CE, and DCI, or by any combination of the three.

[0152] For example, the UE may report the activation / deactivation / switching / fallback of one or more selected, applicable, candidate, recommended, or preferred AI / ML functions / models via an RRC message. For example, the RRC message may be one of the following:

[0153] -UAI(UEAssistanceInformation);

[0154] -RRCReconfigurationComplete message;

[0155] -RRC Reestablishment Request message (RRCReestablishmentRequest message); for example, a new cause value of RestablishmentCause may be introduced, such as AI / ML function / model failure, or activation / deactivation / switching / fallback of AI / ML function / model;

[0156] -Added RRC message.

[0157] For another example, the UE may report activation / deactivation / switching / fallback of one or more selected, applicable, candidate, recommended, or preferred AI / ML functions / models via a MAC CE. In one example, the MAC CE may be newly defined; in another example, an existing MAC CE may be reused.

[0158] For another example, the UE can report the activation / deactivation / switching / fallback of one or more selected, applicable, candidate, recommended, or preferred AI / ML functions / models via physical layer signaling. The physical layer signaling can be transmitted via PRACH, PUCCH, or PUSCH. For example, it can be existing UCI or a newly defined UCI.

[0159] For another example, the reporting information can be a combination of L1 (layer 1) and L2 (layer 2) signaling, such as physical layer signaling and MAC CE. For example, if there are no uplink resources for transmission, the UE may transmit an SR-like PUCCH (which may be a dedicated SR-like PUCCH resource) after one or more AI / ML functions / models fail. After receiving the SR-like PUCCH, the gNB may transmit a DCI containing an uplink grant for PUSCH transmission. The UE then transmits a MAC CE, which may include information about which AI / ML function / model failed.

[0160] In some embodiments, the terminal device receives confirmation information or command information fed back by the network device; wherein the confirmation information or command information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0161] For example, if the UE reports the activation / deactivation / handover / fallback of one or more AI / ML functions / models as selected, applicable, candidate, recommended, or preferred, the gNB shall make a decision and send a command on activation / deactivation / handover / fallback for the one or more AI / ML functions / models. The command may be an RRC or MAC CE or DCI containing information on activation / deactivation / handover / fallback of the one or more AI / ML functions / models.

[0162] In some embodiments, the terminal device sends an ACK to the network device at time slot n; wherein the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect at the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capabilities. In addition, ΔT may also include the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0163] Figure 10 illustrates an example of a network device decision and terminal device confirmation according to an embodiment of the present application. As shown in Figure 10 , the UE sends reporting information in a certain time slot (as shown in A1 in Figure 10 ), and the gNB sends command information in another time slot (as shown in A2 in Figure 10 ). The reporting information indicates the activation / deactivation / switching / fallback of one or more AI / ML functions / models selected, applicable, candidate, recommended, or preferred, and the command information indicates the activation / deactivation / switching / fallback of one or more AI / ML functions / models determined by the gNB.

[0164] As shown in Figure 10, for example, if the UE sends an ACK in the corresponding time slot n, the activation / deactivation / handover / fallback indicated by the command information will be applied starting from the first time slot after n+ΔT (as shown in B in Figure 10). T is greater than or equal to ΔT, where ΔT can be predefined or configurable, or depends on the UE's capabilities. The unit of ΔT can be time slots or other time units such as symbols.

[0165] In another example, ΔT may also include the preparation time / interruption time required for the UE to perform switching between functions / models, which may also depend on the capabilities of the UE, and the value may be zero (a zero value means that no preparation time / disconnection time is required for function / model switching).

[0166] The above schematically illustrates the activation / deactivation / switching / fallback of AI / ML functionality / model initiated by a network device and a terminal device, respectively, but the present application is not limited thereto.

[0167] In some embodiments, the indication information or the reporting information includes activation / deactivation / switching / fallback for AI / ML functionality and AI / ML model; or, the indication information or the reporting information includes activation / deactivation / switching / fallback for AI / ML functionality; or, the indication information or the reporting information includes activation / deactivation / switching / fallback for AI / ML model.

[0168] For example, the signaling (RRC message, MAC CE, or DCI) regarding activation / deactivation / switching / fallback of one or more AI / ML functions / models sent from the gNB side or the UE side may include activation / deactivation / switching / fallback of both the AI / ML functions and the AI / ML models. Alternatively, it may include activation / deactivation / switching / fallback of only the AI / ML functions, or it may include activation / deactivation / switching / fallback of only the AI / ML models.

[0169] In one example, DCI can introduce two domains, one for activation / deactivation / switching / fallback of AI / ML functions and the other for activation / deactivation / switching / fallback of AI / ML models. Alternatively, DCI can introduce one domain for activation / deactivation / switching / fallback of AI / ML models, and activation / deactivation / switching / fallback of AI / ML functions can be implicitly performed through the association between AI / ML functions and AI / ML models.

[0170] For example, an AI / ML function can be associated / mapped to one or more AI / ML models. For example, the gNB can configure one or more AI / ML functions to the UE, and each AI / ML function can be configured with one or more AI / ML models.

[0171] For another example, an AI / ML model can be associated / mapped to one or more AI / ML functions; for example, the function is an AI / ML feature / feature group (FG) enabled through configuration. For example, the gNB can configure one or more AI / ML models to the UE, and each AI / ML model can be configured with one or more AI / ML functions.

[0172] In the embodiments of the present application, it can be applied to one or more or all of the following use cases (including bilateral and unilateral models): beam management, CSI compression and CSI prediction, and positioning. The activation / deactivation / switching / fallback options of the AI / ML function / model are the same for different use cases, or the activation / deactivation / switching / fallback options of the AI / ML function / model are different for different use cases.

[0173] 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.

[0174] As can be seen from the above embodiments, the activation / deactivation / switching / fallback of AI / ML functions / models can be agreed upon between network devices and terminal devices, thereby improving the accuracy and reliability of AI / ML.

[0175] Embodiments of the second aspect

[0176] The embodiment of the present application provides an AI / ML control method, which is described from the perspective of a network device. The embodiment of the second aspect can be combined with the embodiment of the first aspect, and the content that is the same as the embodiment of the first aspect will not be repeated.

[0177] FIG11 is another schematic diagram of a resource configuration method according to an embodiment of the present application. As shown in FIG11 , the method includes:

[0178] 1101, the network device sends an instruction message indicating activation / deactivation / switching / rollback of an AI / ML functionality / model to the terminal device; and / or

[0179] 1102. The network device receives reporting information reported by the terminal device for determining activation / deactivation / switching / rollback of AI / ML functions / models.

[0180] It is worth noting that FIG11 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 FIG11 above.

[0181] In some embodiments, indication information including activation / deactivation / switching / fallback of one or more AI / ML functionalities / models is sent by the network device to the terminal device.

[0182] In some embodiments, the indication information is carried by MAC CE.

[0183] In some embodiments, a set of AI / ML functionalities / models are configured via radio resource control (RRC), and the MAC CE indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models in the configured set of AI / ML functionalities / models.

[0184] In some embodiments, the terminal device sends an ACK corresponding to the PDSCH to the network device in time slot n after receiving the PDSCH carrying the MAC CE;

[0185] The activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability.

[0186] In some embodiments, ΔT also includes the preparation time or interruption time required for the terminal device to perform functionality / model switching.

[0187] In some embodiments, the indication information is carried by downlink control information (DCI).

[0188] In some embodiments, the indication information is a field in the DCI, or an existing field in the DCI, or one or more codepoints in the DCI, or a DCI format, or a DCI scrambled by an RNTI.

[0189] In some embodiments, the DCI is used to schedule PDSCH / PUSCH, or the DCI is not used to schedule PDSCH / PUSCH;

[0190] The DCI is used for one or more serving cells.

[0191] In some embodiments, the DCI is directed to one or more terminal devices.

[0192] In some embodiments, one or more fields in the DCI are used for validation; if the validation passes, the DCI is considered valid; if the validation fails, the DCI is discarded.

[0193] In some embodiments, after receiving the DCI, the terminal device sends an ACK / NACK to the network device to indicate whether the DCI is correctly received.

[0194] In some embodiments, the terminal device sends an ACK corresponding to the DCI in time slot n after correctly receiving the DCI;

[0195] The activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability.

[0196] In some embodiments, ΔT also includes the preparation time or interruption time required for the terminal device to perform functionality / model switching.

[0197] In some embodiments, a set of AI / ML functionalities / models is configured by radio resource control (RRC), a subset of the configured set of AI / ML functionalities / models is selected by a MAC CE, and the DCI indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models in the selected subset.

[0198] In some embodiments, the subset is selected based on performance monitoring of AI / ML functionality / model;

[0199] The length of the field in the DCI indicating activation / deactivation / switching / fallback of the AI / ML functionality / model is determined by the number of AI / ML functionality / models in the subset selected by the MAC CE.

[0200] In some embodiments, a set of AI / ML functionalities / models are configured by radio resource control (RRC), and the DCI indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models among the configured AI / ML functionalities / models.

[0201] In some embodiments, the length of the field in the DCI indicating activation / deactivation / switching / fallback of the AI / ML functionality / model is determined by the number of AI / ML functionality / models configured by the RRC.

[0202] In some embodiments, the indication information is carried by radio resource control (RRC).

[0203] In some embodiments, reporting information including activation / deactivation / switching / fallback of one or more AI / ML functionalities / models is sent by the terminal device to the network device.

[0204] In some embodiments, the reporting information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0205] In some embodiments, the activation / deactivation / switching / fallback of the one or more AI / ML functions / models is determined by the terminal device, and the reporting information includes the activation / deactivation / switching / fallback of the one or more AI / ML functions / models determined by the terminal device.

[0206] In some embodiments, the terminal device receives confirmation information fed back by the network device;

[0207] The confirmation information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0208] In some embodiments, the network device sends the confirmation information in time slot n;

[0209] The activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability.

[0210] In some embodiments, ΔT also includes the preparation time or interruption time required for the terminal device to perform functionality / model switching.

[0211] In some embodiments, the network device determines the activation / deactivation / switching / fallback of the one or more AI / ML functions / models, and the reported information includes the activation / deactivation / switching / fallback of one or more AI / ML functions / models selected, candidated, recommended, or preferred by the terminal device.

[0212] In some embodiments, the terminal device receives confirmation information or command information fed back by the network device;

[0213] The confirmation information or command information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0214] In some embodiments, the terminal device sends an ACK to the network device in time slot n;

[0215] The activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability.

[0216] In some embodiments, ΔT also includes the preparation time or interruption time required for the terminal device to perform functionality / model switching.

[0217] In some embodiments, the indication information or reporting information includes activation / deactivation / switching / fallback for AI / ML functionality and AI / ML model;

[0218] Alternatively, the indication information or the reporting information includes activation / deactivation / switching / fallback for AI / ML functionality;

[0219] Alternatively, the indication information or the reporting information includes activation / deactivation / switching / fallback for an AI / ML model.

[0220] 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.

[0221] As can be seen from the above embodiments, the activation / deactivation / switching / fallback of AI / ML functions / models can be agreed upon between network devices and terminal devices, thereby improving the accuracy and reliability of AI / ML.

[0222] Embodiments of the third aspect

[0223] The present application provides an AI / ML control device. The device may be, for example, a terminal device, or one or more components or assemblies configured in the terminal device. The contents that are the same as those in the first and second aspects of the embodiments are not repeated here.

[0224] FIG12 is a schematic diagram of an AI / ML control device according to an embodiment of the present application. As shown in FIG12 , the AI / ML control device 1200 according to an embodiment of the present application includes:

[0225] A receiving unit 1201 receives instruction information sent by a network device to indicate activation / deactivation / switching / rollback of an AI / ML function / model; and / or

[0226] The sending unit 1202 reports reporting information used to determine activation / deactivation / switching / rollback of AI / ML functions / models to the network device.

[0227] In some embodiments, indication information including activation / deactivation / switching / fallback of one or more AI / ML functionalities / models is sent by the network device to the terminal device.

[0228] In some embodiments, the indication information is carried by MAC CE.

[0229] In some embodiments, a set of AI / ML functionalities / models are configured via radio resource control (RRC), and the MAC CE indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models in the configured set of AI / ML functionalities / models.

[0230] In some embodiments, the terminal device sends an ACK corresponding to the PDSCH to the network device in time slot n after receiving the PDSCH carrying the MAC CE.

[0231] In some embodiments, the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0232] In some embodiments, the indication information is carried by downlink control information (DCI).

[0233] In some embodiments, the indication information is a field in the DCI, or an existing field in the DCI, or one or more codepoints in the DCI, or a DCI format, or a DCI scrambled by an RNTI.

[0234] In some embodiments, the DCI is used to schedule PDSCH / PUSCH.

[0235] In some embodiments, the DCI is not used to schedule PDSCH / PUSCH.

[0236] In some embodiments, the DCI is used for one or more serving cells.

[0237] In some embodiments, the DCI is directed to one or more terminal devices.

[0238] In some embodiments, one or more fields in the DCI are used for validation; if the validation passes, the DCI is considered valid; if the validation fails, the DCI is discarded.

[0239] In some embodiments, after receiving the DCI, the terminal device sends an ACK / NACK to the network device to indicate whether the DCI is correctly received.

[0240] In some embodiments, the terminal device sends an ACK corresponding to the DCI in time slot n after correctly receiving the DCI.

[0241] In some embodiments, the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0242] In some embodiments, a set of AI / ML functionalities / models is configured by radio resource control (RRC), a subset of the configured set of AI / ML functionalities / models is selected by a MAC CE, and the DCI indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models in the selected subset.

[0243] In some embodiments, the subset is selected based on performance monitoring of AI / ML functionality / model.

[0244] In some embodiments, the length of the field in the DCI indicating activation / deactivation / switching / fallback of the AI / ML functionality / model is determined by the number of AI / ML functionality / models in the subset selected by the MAC CE.

[0245] In some embodiments, a set of AI / ML functionalities / models are configured by radio resource control (RRC), and the DCI indicates activation / deactivation / switching / fallback of one or more AI / ML functionalities / models among the configured AI / ML functionalities / models.

[0246] In some embodiments, the length of the field in the DCI indicating activation / deactivation / switching / fallback of the AI / ML functionality / model is determined by the number of AI / ML functionality / models configured by the RRC.

[0247] In some embodiments, the indication information is carried by radio resource control (RRC).

[0248] In some embodiments, reporting information including activation / deactivation / switching / fallback of one or more AI / ML functionalities / models is sent by the terminal device to the network device.

[0249] In some embodiments, the reporting information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0250] In some embodiments, the activation / deactivation / switching / fallback of the one or more AI / ML functions / models is determined by the terminal device, and the reporting information includes the activation / deactivation / switching / fallback of the one or more AI / ML functions / models determined by the terminal device.

[0251] In some embodiments, the terminal device receives confirmation information fed back by the network device; wherein, the confirmation information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0252] In some embodiments, the network device sends the confirmation information in time slot n.

[0253] In some embodiments, the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0254] In some embodiments, the network device determines the activation / deactivation / switching / fallback of the one or more AI / ML functions / models, and the reported information includes the activation / deactivation / switching / fallback of one or more AI / ML functions / models selected, candidated, recommended, or preferred by the terminal device.

[0255] In some embodiments, the terminal device receives confirmation information or command information fed back by the network device; wherein, the confirmation information or command information is carried by radio resource control (RRC) and / or MAC CE and / or physical layer signaling.

[0256] In some embodiments, the terminal device sends an ACK to the network device in time slot n.

[0257] In some embodiments, the activation / deactivation / switching / fallback of the AI / ML functionality / model is applied or takes effect in the first time slot after n+ΔT, where ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform the functionality / model switching.

[0258] In some embodiments, the indication information or the reporting information includes activation / deactivation / switching / fallback for AI / ML functionality and AI / ML model.

[0259] In some embodiments, the indication information or the reporting information includes activation / deactivation / switching / fallback for AI / ML functionality.

[0260] In some embodiments, the indication information or the reporting information includes activation / deactivation / switching / fallback for an AI / ML model.

[0261] 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.

[0262] 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 AI / ML control device 1200 may also include other components or modules. For the specific contents of these components or modules, reference may be made to the relevant art.

[0263] In addition, for the sake of simplicity, FIG12 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.

[0264] As can be seen from the above embodiments, the activation / deactivation / switching / fallback of AI / ML functions / models can be agreed upon between network devices and terminal devices, thereby improving the accuracy and reliability of AI / ML.

[0265] Embodiments of the fourth aspect

[0266] The present application embodiment provides an AI / ML control configuration. The device may be, for example, a 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 to third aspects of the embodiment are not repeated here.

[0267] FIG13 is another schematic diagram of an AI / ML control device according to an embodiment of the present application. As shown in FIG13 , the AI / ML control device 1300 includes:

[0268] A sending unit 1301, which sends instruction information for indicating activation / deactivation / switching / rollback of AI / ML functionality / model to the terminal device; and / or

[0269] The receiving unit 1302 receives reporting information reported by the terminal device for determining activation / deactivation / switching / rollback of AI / ML functions / models.

[0270] 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.

[0271] 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 AI / ML control device 1300 may also include other components or modules. For the specific contents of these components or modules, reference may be made to the relevant art.

[0272] In addition, for the sake of simplicity, FIG13 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.

[0273] As can be seen from the above embodiments, the activation / deactivation / switching / fallback of AI / ML functions / models can be agreed upon between network devices and terminal devices, thereby improving the accuracy and reliability of AI / ML.

[0274] Embodiments of the fifth aspect

[0275] An embodiment of the present application also provides a communication system, and reference may be made to FIG1 . The contents that are the same as those in the first to fourth aspects of the embodiments will not be repeated.

[0276] In some embodiments, the communication system 100 may include at least:

[0277] A terminal device that receives indication information sent by a network device to indicate activation / deactivation / switching / fallback of an AI / ML function / model, and / or reports reporting information to the network device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

[0278] In some embodiments, the communication system 100 may include at least:

[0279] A network device that sends indication information for indicating activation / deactivation / switching / fallback of an AI / ML function / model to a terminal device, and / or receives reporting information reported by the terminal device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

[0280] 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.

[0281] Figure 14 is a schematic diagram of a terminal device according to an embodiment of the present application. As shown in Figure 14 , terminal device 1400 may include a processor 1410 and a memory 1420. Memory 1420 stores data and programs and is coupled to processor 1410. 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.

[0282] For example, the processor 1410 may be configured to execute a program to implement the AI / ML control method as described in the embodiment of the first aspect. For example, the processor 1410 may be configured to perform the following control: receiving indication information sent by a network device to indicate activation / deactivation / switching / fallback of an AI / ML function / model; and / or reporting reporting information to the network device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

[0283] As shown in Figure 14 , the terminal device 1400 may further include: a communication module 1430, an input unit 1440, a display 1450, and a power supply 1460. 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 1400 does not necessarily include all of the components shown in Figure 14 , and these components are not essential. Furthermore, the terminal device 1400 may also include components not shown in Figure 14 , for which reference may be made to the prior art.

[0284] An embodiment of the present application further provides a 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.

[0285] Figure 15 is a schematic diagram illustrating the structure of a network device according to an embodiment of the present application. As shown in Figure 15 , network device 1500 may include a processor 1510 (e.g., a central processing unit (CPU)) and a memory 1520; memory 1520 is coupled to processor 1510. Memory 1520 may store various data and may also store an information processing program 1530, which is executed under the control of processor 1510.

[0286] For example, the processor 1510 may be configured to execute a program to implement the AI / ML control method as described in the embodiment of the second aspect. For example, the processor 1510 may be configured to perform the following control: sending indication information indicating activation / deactivation / switching / fallback of an AI / ML function / model to a terminal device; and / or receiving reporting information reported by the terminal device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

[0287] In addition, as shown in FIG15 , network device 1500 may further include: a transceiver 1540 and an antenna 1550, etc.; wherein, the functions of the above components are similar to those in the prior art and are not described in detail here. It is worth noting that network device 1500 does not necessarily include all the components shown in FIG15 ; in addition, network device 1500 may also include components not shown in FIG15 , and reference may be made to the prior art for details.

[0288] 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 AI / ML control method described in the embodiment of the first aspect.

[0289] 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 AI / ML control method described in the embodiment of the first aspect.

[0290] An embodiment of the present application also provides a computer program, wherein when the program is executed in a network device, the program causes the network device to execute the AI / ML control method described in the embodiment of the second aspect.

[0291] An embodiment of the present application also provides a storage medium storing a computer program, wherein the computer program enables the network device to execute the AI / ML control method described in the embodiment of the second aspect.

[0292] 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.

[0293] 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).

[0294] 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.

[0295] 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.

[0296] 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.

[0297] Regarding the implementation methods including the above embodiments, the following additional notes are also disclosed:

[0298] 1. An AI / ML control method, comprising:

[0299] The terminal device receives instruction information sent by the network device to indicate activation / deactivation / switching / rollback of AI / ML functionality / model; and / or

[0300] The terminal device reports reporting information to the network device for determining the activation / deactivation / switching / fallback of AI / ML functions / models.

[0301] 2. An AI / ML control method, comprising:

[0302] The network device sends an indication message to the terminal device indicating activation / deactivation / switching / fallback of an AI / ML functionality / model; and / or

[0303] The network device receives reporting information reported by the terminal device for determining activation / deactivation / switching / fallback of AI / ML functions / models.

[0304] 3. 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 AI / ML control method as described in Note 1.

[0305] 4. 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 AI / ML control method as described in Note 2.

[0306] 5. A communication system comprising:

[0307] A network device that sends indication information for indicating activation / deactivation / switching / fallback of an AI / ML function / model to a terminal device, and / or receives reporting information reported by the terminal device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

Claims

1. An AIML control device, include: A receiving unit, which receives indication information sent by a network device for indicating activation / deactivation / switching / rollback of an AI / ML function / model; and / or A sending unit that reports reporting information to the network device for determining activation / deactivation / switching / fallback of AI / ML functions / models.

2. The device according to claim 1, in, Indication information including activation / deactivation / switching / rollback of one or more AI / ML functions / models is sent by the network device to the terminal device.

3. The device according to claim 2, in, The indication information is carried by the MAC CE; A set of AI / ML functions / models are configured through radio resource control, and the MAC CE indicates activation / deactivation / switching / fallback of one or more AI / ML functions / models in the configured set of AI / ML functions / models.

4. The device according to claim 3, in, The terminal device sends an ACK corresponding to the PDSCH to the network device in time slot n after receiving the PDSCH carrying the MAC CE; Among them, the activation / deactivation / switching / fallback of the AI / ML function / model is applied or takes effect in the first time slot after n+ΔT, ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform function / model switching.

5. The device according to claim 2, in, The indication information is carried by downlink control information; The indication information is a field in the downlink control information, or an existing field in the downlink control information, or one or more code points in the downlink control information, or a downlink control information format, or downlink control information scrambled by an RNTI.

6. The device according to claim 5, in, The downlink control information is used for scheduling PDSCH / PUSCH, or the downlink control information is not used for scheduling PDSCH / PUSCH; The downlink control information is used for one or more serving cells; The downlink control information is directed to one or more terminal devices.

7. The device according to claim 5, in, One or more fields in the downlink control information are used for verification; if the verification passes, the downlink control information is considered valid, and if the verification fails, the downlink control information is discarded.

8. The device according to claim 5, in, After receiving the downlink control information, the terminal device sends an ACK / NACK to the network device to indicate whether the downlink control information is correctly received.

9. The device according to claim 5, in, The terminal device sends an ACK corresponding to the downlink control information in time slot n after correctly receiving the downlink control information; Among them, the activation / deactivation / switching / fallback of the AI / ML function / model is applied or takes effect in the first time slot after n+ΔT, ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform function / model switching.

10. The device according to claim 5, in, A set of AI / ML functions / models is configured by radio resource control, a subset of the configured set of AI / ML functions / models is selected by MAC CE, and the downlink control information indicates activation / deactivation / switching / fallback of one or more AI / ML functions / models in the selected subset; The subset is selected based on performance monitoring of the AI / ML functions / models; The length of the field indicating activation / deactivation / switching / fallback of the AI / ML function / model in the downlink control information is determined by the number of AI / ML functions / models in the subset selected by the MAC CE.

11. The device according to claim 5, in, A set of AI / ML functions / models are configured by radio resource control, and the downlink control information indicates activation / deactivation / switching / fallback of one or more AI / ML functions / models among the configured AI / ML functions / models; The length of the field indicating activation / deactivation / switching / fallback of the AI / ML function / model in the downlink control information is determined by the number of AI / ML functions / models configured by the radio resource control.

12. The device according to claim 2, in, The indication information is carried by the radio resource control.

13. The device according to claim 1, in, Reporting information including activation / deactivation / switching / fallback of one or more AI / ML functions / models is sent by the terminal device to the network device; The reporting information is carried by radio resource control and / or MAC CE and / or physical layer signaling.

14. The device according to claim 13, in, The activation / deactivation / switching / fallback of the one or more AI / ML functions / models is determined by the terminal device, and the reporting information includes the activation / deactivation / switching / fallback of one or more AI / ML functions / models determined by the terminal device.

15. The device according to claim 14, in, The terminal device receives confirmation information fed back by the network device; wherein the confirmation information is carried by radio resource control and / or MAC CE and / or physical layer signaling; The network device sends the confirmation information in time slot n; Among them, the activation / deactivation / switching / fallback of the AI / ML function / model is applied or takes effect in the first time slot after n+ΔT, ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform function / model switching.

16. The device according to claim 13, in, The network device determines the activation / deactivation / switching / fallback of the one or more AI / ML functions / models, and the reported information includes the activation / deactivation / switching / fallback of one or more AI / ML functions / models selected or candidate or recommended or preferred by the terminal device.

17. The device according to claim 16, in, The terminal device receives confirmation information or command information fed back by the network device; wherein the confirmation information or command information is carried by radio resource control and / or MAC CE and / or physical layer signaling; The terminal device sends an ACK to the network device in time slot n; Among them, the activation / deactivation / switching / fallback of the AI / ML function / model is applied or takes effect in the first time slot after n+ΔT, ΔT is a predefined value or a configured value or depends on the terminal capability; ΔT also includes the preparation time or interruption time required for the terminal device to perform function / model switching.

18. The device according to claim 1, in, The indication information or the reporting information includes activation / deactivation / switching / rollback for the AI / ML function and the AI / ML model; Alternatively, the indication information or the reporting information includes activation / deactivation / switching / fallback for the AI / ML function; Alternatively, the indication information or the reporting information includes activation / deactivation / switching / rollback for the AI / ML model.

19. An AIML control device, include: A sending unit, which sends indication information for indicating activation / deactivation / switching / rollback of an AI / ML function / model to a terminal device; and / or A receiving unit receives reporting information reported by the terminal device for determining activation / deactivation / switching / rollback of the AI / ML function / model.

20. A communication system, include: A terminal device that receives indication information sent by a network device to indicate activation / deactivation / switching / fallback of an AI / ML function / model, and / or reports reporting information to the network device for determining activation / deactivation / switching / fallback of an AI / ML function / model.

Images