Model identification method, apparatus, medium, communication device, and communication system

Abstract

A model identification method and device, a storage medium, a communication device, and a communication system, the method comprising: sending first information to a network device, the first information being used by the network device to determine AI / ML functions supported by the terminal and / or AI / ML models supported by the terminal. The present disclosure enables the network device to complete model identification of the terminal.

Description

Technical Field

[0001] This disclosure relates to the field of communication technology, and in particular to a model recognition method, apparatus, medium, communication equipment, and communication system. Background Technology

[0002] In communication systems, artificial intelligence (AI) / machine learning (ML) models can be deployed in network nodes to perform predictions and inferences, thereby improving the performance of the communication system. Summary of the Invention

[0003] This disclosure provides a model recognition method, apparatus, storage medium, communication device, and communication system.

[0004] According to a first aspect of the present disclosure, a model recognition method is provided, executed by a terminal, the method comprising:

[0005] Send first information to the network device, the first information being used by the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal.

[0006] According to a second aspect of the present disclosure, a model recognition method is provided, executed by a network device, comprising:

[0007] The first message sent by the receiving terminal;

[0008] Based on the first information, determine the AI / ML functions supported by the terminal and / or the AI / ML models supported by the terminal.

[0009] According to a third aspect of the present disclosure, a model recognition method is provided, executed by a communication system, the communication system including a terminal and a network device, the method comprising:

[0010] The terminal sends the first information to the network device;

[0011] The network device determines the AI / ML functions and / or AI / ML models supported by the terminal based on the first information.

[0012] According to a fourth aspect of the present disclosure, a model recognition apparatus is provided, the method comprising:

[0013] The first sending module is configured to send first information to a network device, the first information being used by the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal.

[0014] According to a fifth aspect of the present disclosure, a model recognition apparatus is provided, comprising:

[0015] The receiving module is configured to receive the first information sent by the terminal;

[0016] The first determining module is configured to determine, based on the first information, the AI / ML functions supported by the terminal and / or the AI / ML models supported by the terminal.

[0017] According to a sixth aspect of the present disclosure, a terminal is provided, comprising:

[0018] First processor;

[0019] A first memory for storing instructions executable by a first processor;

[0020] The first processor is configured to execute the executable instructions to implement the model recognition method provided in the first aspect of this disclosure.

[0021] According to a seventh aspect of the present disclosure, a network device is provided, comprising:

[0022] Second processor;

[0023] A second memory used to store instructions executable by a second processor;

[0024] The second processor is configured to execute the executable instructions to implement the model recognition method provided in the second aspect of this disclosure.

[0025] According to an eighth aspect of the present disclosure, a storage medium is provided that stores computer program instructions thereon, which, when executed by a processor, implement the steps of the model recognition method provided in the first or second aspect of the present disclosure.

[0026] According to a ninth aspect of the present disclosure, the present disclosure provides a program product that, when executed by a terminal, causes the terminal to perform the model recognition method as provided in the first aspect, or when executed by a network device, causes the network device to perform the method as provided in the second aspect.

[0027] According to a tenth aspect of the present disclosure, an embodiment of the present disclosure provides a computer program that, when run on a computer, causes the computer to perform the steps of the model recognition method provided in the first or second aspect.

[0028] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects: the terminal can send first information to the network device, thereby the network device can determine the AI / ML functions supported by the terminal and / or the AI / ML models supported by the terminal based on the first information, thereby completing the model recognition of the terminal.

[0029] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0030] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which:

[0031] Figure 1a This is a schematic diagram of a communication system according to some embodiments.

[0032] Figure 1b This is a schematic diagram illustrating a CSI compression feedback and recovery according to some embodiments.

[0033] Figure 2a This is a flowchart illustrating a model recognition method according to some embodiments.

[0034] Figure 2b This is a flowchart illustrating a model recognition method according to some embodiments.

[0035] Figure 2c This is a flowchart illustrating a model recognition method according to some embodiments.

[0036] Figure 3a This is a flowchart illustrating a model recognition method according to some embodiments.

[0037] Figure 3b This is a flowchart illustrating a model recognition method according to some embodiments.

[0038] Figure 3c This is a flowchart illustrating a model recognition method according to some embodiments.

[0039] Figure 3d This is a flowchart illustrating a model recognition method according to some embodiments.

[0040] Figure 3e This is a flowchart illustrating a model recognition method according to some embodiments.

[0041] Figure 3f This is a flowchart illustrating a model recognition method according to some embodiments.

[0042] Figure 4aThis is a flowchart illustrating a model recognition method according to some embodiments.

[0043] Figure 4b This is a flowchart illustrating a model recognition method according to some embodiments.

[0044] Figure 4c This is a flowchart illustrating a model recognition method according to some embodiments.

[0045] Figure 4d This is a flowchart illustrating a model recognition method according to some embodiments.

[0046] Figure 5a This is a flowchart illustrating a model recognition method according to some embodiments.

[0047] Figure 5b This is a flowchart illustrating a model recognition method according to some embodiments.

[0048] Figure 5c This is a flowchart illustrating a model recognition method according to some embodiments.

[0049] Figure 5d This is a flowchart illustrating a model recognition method according to some embodiments.

[0050] Figure 5e This is a flowchart illustrating a model recognition method according to some embodiments.

[0051] Figure 5f This is a flowchart illustrating a model recognition method according to some embodiments.

[0052] Figure 6 This is a flowchart illustrating a model recognition method according to some embodiments.

[0053] Figure 7 This is a schematic diagram illustrating the functionality and model ID of an AI-based CSI compressed feedback according to some embodiments.

[0054] Figure 8 This is a block diagram illustrating a model recognition device according to some embodiments.

[0055] Figure 9 This is a block diagram illustrating a model recognition device according to some embodiments.

[0056] Figure 10 This is a block diagram of a terminal according to some embodiments.

[0057] Figure 11 This is a block diagram of a network device according to some embodiments. Detailed Implementation

[0058] This disclosure proposes a "model recognition method". In some embodiments, the terms "model recognition method" and "information processing method" and "communication method" can be used interchangeably, as can the terms "model recognition device" and "information processing device" and "communication device", and the terms "communication system" and "information processing system" and "communication system".

[0059] This disclosure is not exhaustive, but merely illustrative of some embodiments, and is not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0060] The terminology used in the embodiments of this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. The singular expressions "a," "an," "the," "the," "the," "the," "the foregoing," "this," etc., in the embodiments of this disclosure also include the plural expressions, unless the context clearly indicates otherwise. The predefined in the embodiments of this disclosure can be understood as defined, pre-defined, stored, pre-stored, pre-negotiated, pre-configured, solidified, or pre-burned, etc.

[0061] Prefixes such as "first" and "second" in this disclosure are merely for distinguishing different descriptive objects and do not limit the position, order, priority, quantity, or content of the described objects. For example, if the described object is a "field," the ordinal numbers before "field" in "first field" and "second field" do not restrict the position or order of the "fields," nor do "first" and "second" restrict whether the "fields" they modify are in the same message, nor do they restrict the order of "first field" and "second field." Similarly, if the described object is a "level," the ordinal numbers before "level" in "first level" and "second level" do not restrict the priority between "levels." Furthermore, the quantity of described objects is not limited by ordinal numbers and can be one or more; for example, in "first device," the quantity of "device" can be one or more. Furthermore, the objects modified by different prefixes can be the same or different. For example, if the object being described is "device," then "first device" and "second device" can be devices of the same type or different types. Similarly, if the object being described is "information," then "first information" and "second information" can be information with the same content or information with different content. In summary, the use of ordinal numbers and other prefixes used to distinguish the objects described in this disclosure does not constitute a limitation on the objects being described. The description of the objects being described is based on the claims or the context of the embodiments, and should not constitute an unnecessary limitation due to the use of such prefixes.

[0062] In this embodiment of the disclosure, "multiple" refers to two or more. In this embodiment of the disclosure, "and / or" is used to describe the relationship between related objects, representing three relationships that can exist independently. For example, A and / or B can represent: A existing alone, B existing alone, or A and B existing simultaneously. Descriptions such as "at least one of A1, A2, ..., An (or at least one of them)" in this embodiment of the disclosure include the case where any one of A1, A2, ..., An exists alone, as well as the case where any combination of any multiple of A1, A2, ..., An exists, and each case can exist independently; for example, the description "at least one of A, B, C" includes the cases of A alone, B alone, C alone, a combination of A and B, a combination of A and C, a combination of B and C, and a combination of A, B, and C.

[0063] In some embodiments, the notation "in one case A, in another case B" or "in response to one case A, in response to another case B" may include the following technical solutions depending on the situation: A is executed regardless of B, i.e., A is executed in some embodiments; B is executed regardless of A, i.e., B is executed in some embodiments; A and B are selectively executed, i.e., A and B are selected for execution in some embodiments; A and B are both executed, i.e., A and B are executed in some embodiments. The same applies when there are more branches such as A, B, C, etc.

[0064] In some embodiments, the terms “in response to…”, “in response to determining…”, “in the case of…”, “when…”, “if…”, “if…”, etc., can be used interchangeably.

[0065] In some embodiments, “including A,” “containing A,” “for indicating A,” and “carrying A” can be interpreted as directly carrying A or indirectly indicating A.

[0066] In some embodiments, terms such as “greater than,” “greater than or equal to,” “above,” “higher than,” and “not less than” can be used interchangeably, as can terms such as “less than,” “less than or equal to,” “below,” “lower than,” and “not greater than”.

[0067] In some embodiments, the terms “network element”, “node”, “function”, “device”, “equipment”, “system”, “chip”, “chip system”, etc., can be used interchangeably.

[0068] In some embodiments, the terms terminal, terminal device, user equipment (UE), user terminal, mobile station (MS), mobile terminal (MT), subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, and client can be used interchangeably.

[0069] Furthermore, each element, each row, or each column in the table of this disclosure can be implemented as an independent embodiment, and any combination of any element, any row, or any column can also be implemented as an independent embodiment.

[0070] It should be noted that all actions involving the acquisition of signals, information, or data in this disclosure are carried out in compliance with the relevant data protection laws and policies of the country where the location is situated, and with authorization from the owner of the relevant device.

[0071] Although operations are described in a specific order in the accompanying drawings in this disclosure, it should not be construed as requiring these operations to be performed in the specific order or serial order shown, or requiring all of the shown operations to obtain the desired result. In certain environments, multitasking and parallel processing may be advantageous. Furthermore, sending multiple messages via the same message is also advantageous.

[0072] The following describes the application environment of the model recognition method provided in this embodiment.

[0073] Figure 1a This is a schematic diagram of the structure of a communication system according to an embodiment of the present disclosure.

[0074] like Figure 1a As shown, the communication system 100 includes a terminal 101 and a network device 102.

[0075] In some embodiments, terminal 101 includes, but is not limited to, at least one of the following: mobile phone, wearable device, Internet of Things device, car with communication function, smart car, tablet computer, computer with wireless transceiver function, virtual reality (VR) terminal device, augmented reality (AR) terminal device, wireless terminal device in industrial control, wireless terminal device in self-driving, wireless terminal device in remote medical surgery, wireless terminal device in smart grid, wireless terminal device in transportation safety, wireless terminal device in smart city, and wireless terminal device in smart home.

[0076] In some embodiments, network device 102 may include at least one of access network device and core network device.

[0077] In some embodiments, the access network device is, for example, a node or device that connects a terminal to a wireless network, such as a Radio Access Network (RAN). The access network device may include, but is not limited to, at least one of the following in a 5G communication system: an evolved Node B (eNB), a next-generation eNB (ng-eNB), a next-generation Node B (gNB), a node B (NB), a home node B (HNB), a home evolved node B (HeNB), a radio backhaul device, a radio network controller (RNC), a base station controller (BSC), a base transceiver station (BTS), a base band unit (BBU), a mobile switching center, a base station in a 6G communication system, an open RAN, a cloud RAN, a base station in other communication systems, and an access node in a wireless fidelity (WiFi) system.

[0078] In some embodiments, the technical solutions of this disclosure can be applied to the Open RAN architecture. In this case, the interfaces between or within access network devices involved in the embodiments of this disclosure can be transformed into internal interfaces of Open RAN. The processes and information interactions between these internal interfaces can be implemented by software or programs.

[0079] In some embodiments, the access network device may be composed of a central unit (CU) and a distributed unit (DU). The CU may also be called a control unit. The CU-DU structure can separate the protocol layer of the access network device. Some of the protocol layer functions are centrally controlled by the CU, while the remaining part or all of the protocol layer functions are distributed in the DU and centrally controlled by the CU. However, this is not the only possibility.

[0080] In some embodiments, a core network device may be a single device comprising one or more network elements, or it may be multiple devices or a group of devices, each comprising all or part of the aforementioned one or more network elements. Network elements may be virtual or physical.

[0081] It is understood that the communication system described in this disclosure is for the purpose of more clearly illustrating the technical solutions of this disclosure, and does not constitute a limitation on the technical solutions provided in this disclosure. As those skilled in the art will know, with the evolution of system architecture and the emergence of new business scenarios, the technical solutions provided in this disclosure are also applicable to similar technical problems.

[0082] The following embodiments of this disclosure can be applied to Figure 1a The communication system 100 shown, or a part thereof, but not limited to it. Figure 1a The entities shown are illustrative; a communication system may include... Figure 1a All or part of the main body, or may include Figure 1a Other entities besides the main body, the number and form of each entity are arbitrary, the connection relationship between the entities is illustrative, the entities may not be connected or may be connected, and the connection can be in any way, it can be a direct connection or an indirect connection, it can be a wired connection or a wireless connection.

[0083] The embodiments disclosed herein can be applied to Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 5G new radio (NR), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Futuregeneration radio access (FX), Global System for Mobile communications (GSM), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), and IEEE 802.20, Ultra-Wideband (UWB), Bluetooth (a registered trademark), Public Land Mobile Network (PLMN) networks, Device-to-Device (D2D) systems, Machine-to-Machine (M2M) systems, Internet of Things (IoT) systems, Vehicle-to-Everything (V2X) systems, systems utilizing other communication methods, and next-generation systems built upon them, etc. Furthermore, multiple systems can be combined (e.g., a combination of LTE or LTE-A with 5G).

[0084] In some embodiments of this disclosure, AI / ML models can be deployed in network nodes to perform prediction and inference, thereby improving the performance of the communication system.

[0085] Alternatively, a network node can be a terminal or a network device.

[0086] Optionally, AI technology can be used to reduce terminal feedback overhead or improve the accuracy of Channel Status Information (CSI) feedback. For example, compressed feedback and recovery of CSI can be achieved using a bilateral AI / ML model based on a partial model of CSI generation on the terminal side and a partial model of CSI recovery on the network side, respectively. Another example is that time-domain CSI prediction can be achieved based on an AI / ML model on the UE side.

[0087] For example, as follows Figure 1b A schematic diagram of CSI compression feedback and recovery based on a bilateral AI / ML model is given. On the UE side, the downlink channel information H is compressed by generating a partial model through CSI and then quantized into a binary bit stream s and sent to the gNB. On the gNB side, the partial model of CSI recovery is used to recover H', which is approximately the same as the original downlink information.

[0088] For the AI / ML model on the UE side or the UE-side part of the bilateral model, consensus can be reached between the network and the UE through AI / ML function identification and / or AI / ML model identification, thereby enabling lifecycle management of the model. AI / ML function identification and / or AI / ML model identification can be referred to as model identification.

[0089] Figure 2a This is a flowchart illustrating a model recognition method according to some embodiments. For example... Figure 2a As shown, this model recognition method can be used in a terminal and includes the following steps.

[0090] In step S210a, the first information is sent to the network device.

[0091] In some implementations, the first information is used by the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal.

[0092] In some implementations, the network device can receive first information sent by the terminal and determine the AI / ML functions and / or AI / ML models supported by the terminal based on the first information, thereby completing the process of identifying the models supported by the terminal.

[0093] In some implementations, the AI / ML functions and / or AI / ML models supported by the terminal may be all available AI / ML functions and / or AI / ML models deployed in the terminal, or they may be the AI / ML functions and / or AI / ML models currently used by the terminal. This may be determined based on the first information sent by the terminal, or based on the strategy of the terminal in sending the first information.

[0094] In some implementations, the AI / ML functions and / or AI / ML models supported by the terminal may also be referred to as the AI / ML functions and / or AI / ML models corresponding to the terminal.

[0095] In some implementations, the first information may include information for the network device to determine the AI / ML functions supported by the terminal, or information for the network device to determine the AI / ML models supported by the terminal, or information for the network device to determine the AI / ML functions supported by the terminal and information for the network device to determine the AI / ML models supported by the terminal.

[0096] This disclosure specifies that a terminal can send first information to a network device through information reporting, thereby enabling the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal based on the first information, thus completing the model recognition of the terminal.

[0097] In some implementations, after determining the AI / ML functions and / or AI / ML models supported by the terminal, the network device can perform lifecycle management on the terminal's models based on the AI / ML functions and / or AI / ML models supported by the terminal.

[0098] In some implementations, network devices may manage the lifecycle of terminal models based on the AI / ML functions and / or AI / ML models supported by the terminal, including activating, deactivating, selecting, and switching the terminal's AI / ML models.

[0099] In some implementations, the network device can instruct the terminal to activate the AI / ML functions it supports, thereby explicitly activating the AI / ML functions supported by the terminal. For example, the network device can instruct the terminal to activate the AI / ML functions it supports via an instruction message.

[0100] In some implementations, network devices can configure the terminal with configuration parameters corresponding to the AI / ML functions supported by the terminal, thereby implicitly activating the AI / ML functions supported by the terminal.

[0101] In some implementations, the network device can instruct the terminal to activate the AI / ML model it supports, based on the AI / ML model supported by the terminal, thereby activating the AI / ML model supported by the terminal. For example, the network device can instruct the terminal to activate the AI / ML model it supports via an instruction message.

[0102] In some implementations, the network device can configure the configuration parameters corresponding to the AI / ML models supported by the terminal according to the AI / ML models supported by the terminal, thereby activating the AI / ML models supported by the terminal.

[0103] In some implementations, the first information may include one or more of the following:

[0104] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0105] In some implementations, feature type information is used to characterize the feature types supported by the terminal.

[0106] Therefore, in this embodiment of the disclosure, the feature type information included in the first information can identify the feature type supported by the terminal. Optionally, the feature type information can be combined with other types of first information, for example, combined with at least one of the first function information, second function information, third function information, first model identifier, second model identifier, third model identifier, and terminal-associated metadata, so that the network device can determine the AI / ML function and / or AI / ML model corresponding to the feature type information.

[0107] In some implementations, the feature type may include AI-based CSI compressed feedback and AI-based temporal CSI prediction.

[0108] For example, when the feature type supported by the terminal is AI-based CSI compressed feedback, the terminal can choose to report using the first information used to determine the AI / ML functions it supports. As another example, when the feature type supported by the terminal is AI-based temporal CSI prediction, the terminal can choose to report using the first information used to determine the AI / ML models it supports.

[0109] It should be noted that, in the above example, the correspondence between the feature types supported by the terminal and the first information used to determine the AI / ML functions and / or AI / ML models supported by the terminal can be interchanged according to actual needs. For example, when the feature type supported by the terminal is AI-based temporal CSI prediction, the first information used to determine the AI / ML functions supported by the terminal can be selected for reporting.

[0110] In some implementations, the first functional information can be understood as fine-grained functional information used to describe all functional types corresponding to a complete set of AI / ML functions supported by the terminal. Therefore, the network device can determine the AI / ML functions supported by the terminal based on the received first functional information.

[0111] Optionally, the first functional information can describe all functional types corresponding to a complete set of AI / ML functions supported by the terminal in the form of functional identifiers. In this case, the correspondence between functional identifiers and fine-grained functional information can be pre-stored. For example, in a mobile scenario, suppose function 1 corresponds to the terminal being able to predict channel information on a maximum of 10 slots, with an interval of 2 between adjacent slots, and can be used for inference below 30 km / h. Function 2 is defined as the terminal being able to predict channel information on a maximum of 20 slots, with an interval of 5 between adjacent slots, and can be used for inference below 60 km / h. If the first functional information includes function 1, the network device can determine, based on the first functional information and the preset correspondence, that the AI / ML function supported by the terminal is the ability to predict channel information on a maximum of 10 slots, with an interval of 2 between adjacent slots, and can be used for inference below 30 km / h.

[0112] Optionally, the first functional information can describe all the functional types corresponding to a complete set of AI / ML functions supported by the terminal in the form of a combination of multiple functional types. In this case, the first functional information can be, for example, channel information that can predict up to 10 slots, with an interval of 2 between adjacent slots, which can be used for inference below 30 km / h. Thus, the network device can determine that the AI / ML function supported by the terminal is channel information that can predict up to 10 slots, with an interval of 2 between adjacent slots, which can be used for inference below 30 km / h based on the first functional information.

[0113] Among them, the channel information that can be predicted on a maximum of 10 slots, the interval between adjacent slots, and the inference for speeds below 30 km / h can be understood as a partial function type corresponding to a complete set of AI / ML functions.

[0114] In some implementations, a complete set of AI / ML functions can be all the function types supported by a model. Therefore, the function types corresponding to a complete set of AI / ML functions can be determined according to the actual situation of the terminal model.

[0115] Using the previous example, if a certain model in the terminal supports the function of predicting channel information on a maximum of 10 slots and the interval between adjacent slots is 2, and it can be used for inference below 30 km / h, then the various function types corresponding to a complete set of AI / ML functions can include predicting channel information on a maximum preset number of slots, the interval between adjacent slots, and the inference speed.

[0116] In some implementations, the second functional information can be understood as coarse-grained functional information used to describe partial functional types corresponding to a complete set of AI / ML functions supported by the terminal. The first model identifier is used by the network device to determine the partial functional types corresponding to a complete set of AI / ML functions supported by the terminal based on a preset correspondence.

[0117] Therefore, network devices can jointly determine the AI / ML functions supported by the terminal based on the received second functional information and the first model identifier.

[0118] For example, assuming that all function types corresponding to a complete set of AI / ML functions can include channel scenarios and model output payloads, the second function information can be, for example, Uma (Urban Macro) / UMi (Urban Micro) scenarios supporting channel scenarios, and medium-to-high-speed mobile scenarios supporting channel scenarios. In the Uma / UMi scenario, the first model identifier can include, for example, ID (Identity document) 1 and ID 2, corresponding to AI / ML functions with a maximum payload size of 120 bits after maximum compression information quantization and AI / ML functions with a maximum payload size of 60 bits, respectively. In the medium-to-high-speed mobile scenario, the first model identifier can include, for example, ID 3 and ID 4, corresponding to AI / ML models with a maximum payload size of 120 bits after maximum compression information quantization and AI / ML models with a maximum payload size of 60 bits, respectively. In this case, if the second functional information includes the Uma / UMi scenario under the support channel scenario, and the first model identifier includes ID1, the network device can determine that the AI / ML function supported by the terminal is the AI / ML function that supports the Uma / UMi scenario and supports the maximum payload size of 120 bits after the maximum compressed information quantization of inference based on the second functional information and the first model identifier.

[0119] Furthermore, considering that functional information may correspond to models, for example, some functions are only supported by a certain model, after determining the AI / ML functions supported by the terminal, the corresponding AI / ML model can be obtained. Therefore, the network device can determine the AI / ML model supported by the terminal based on the functional information and the model identifier. Thus, in some implementations, the first information may include third functional information and third model identifier, which are used by the network device to determine the AI / ML model supported by the terminal.

[0120] In some implementations, the third functional information can be understood as coarse-grained functional information used to describe partial functional types corresponding to a complete set of AI / ML functions supported by the terminal. The third model identifier is used by the network device to determine the partial functional types corresponding to a complete set of AI / ML functions supported by the terminal based on a preset correspondence.

[0121] In some implementations, the metadata associated with the terminal is used by the network device to determine the parameter configuration of the AI / ML functions supported by the terminal.

[0122] Considering that the terminal's model may be a model with strong generalization ability, which can be used for inference or prediction of various parameters, such as inference or prediction of different numbers of ports, in order to fully determine the AI / ML functions supported by the terminal, such as accurately determining the number of ports currently used by the terminal, after determining that the AI / ML function supported by the terminal can predict channel information on a maximum of 10 slots, and the interval between adjacent slots is 2, and it can be used for inference below 30 km / h, the number of ports supported by the terminal can be further determined by combining the metadata associated with the terminal.

[0123] In some implementations, the metadata associated with the terminal is used by the network device to determine the parameter configuration of the AI / ML models supported by the terminal.

[0124] Similarly, considering that the model supported by the terminal may be a model with strong generalization ability, which can be used for inference or prediction of various parameters, such as inference or prediction of different numbers of ports, in order to fully determine the AI / ML model supported by the terminal, such as accurately determining the number of ports currently used by the terminal, the metadata associated with the terminal can be further combined for determination.

[0125] In some implementations, the first information may include only the metadata associated with the terminal. That is, the network device can determine the AI / ML functions and / or AI / ML models supported by the terminal through the metadata associated with the terminal.

[0126] In some implementations, the metadata associated with the terminal may be used in combination with other information used to determine the AI / ML functions and / or AI / ML models supported by the terminal. For example, the first information may include first function information and the metadata associated with the terminal; the first information may include second function information, a first model identifier and the metadata associated with the terminal; the first information may include a second model identifier and the metadata associated with the terminal; or the first information may include third function information, a third model identifier and the metadata associated with the terminal.

[0127] In some implementations, a second identification information can correspond to a unique model, thereby allowing the network device to determine the AI / ML model supported by the terminal through the second identification information.

[0128] In some implementations, one second identification information can correspond to multiple models, thereby allowing the network device to determine the multiple AI / ML models supported by the terminal through the second identification information.

[0129] In some implementations, when the first information is used by the network device to determine the AI / ML functions supported by the terminal, the first information may include:

[0130] The first functional information describes the AI / ML functions supported by the terminal; or,

[0131] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0132] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0133] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0134] In this embodiment of the disclosure, the terminal may send first functional information to the network device so that the network can determine the AI / ML functions supported by the terminal based on the first functional information. The terminal may also send second functional information and a first model identifier to the network device so that the network can determine the AI / ML functions supported by the terminal based on the second functional information and the first model identifier. The terminal may also send a second model identifier to the network device so that the network can determine the AI / ML functions supported by the terminal based on the second model identifier. The terminal may also send terminal-associated metadata to the network device so that the network can determine the AI / ML functions supported by the terminal based on the terminal-associated metadata.

[0135] In some implementations, when the first information is used by the network device to determine the AI / ML models supported by the terminal, the first information includes:

[0136] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0137] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0138] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0139] In this embodiment of the disclosure, the terminal may send third functional information and third model identifier to the network device, so that the network can determine the AI / ML model supported by the terminal based on the third functional information and third model identifier. The terminal may also send a second model identifier to the network device, so that the network can determine the AI / ML model supported by the terminal based on the second model identifier. The terminal may also send terminal-associated metadata to the network device, so that the network can determine the AI / ML model supported by the terminal based on the terminal-associated metadata.

[0140] In some implementations, the metadata associated with the terminal may include one or more of the following:

[0141] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0142] In some implementations, the terminal's location information may be the cell ID information where the terminal is located.

[0143] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0144] In this embodiment of the disclosure, the second model identifier can be pre-divided into different value ranges based on the feature types supported by the terminal, the function information supported by the terminal, the metadata associated with the terminal, the feature types and function information supported by the terminal, and the feature types, function information and metadata supported by the terminal, etc.

[0145] For example, continuing with the previous example, suppose the terminal supports feature types including AI-based CSI compressed feedback and AI-based temporal CSI prediction. The AI-based CSI compressed feedback can have 1000 models, and the AI-based temporal CSI prediction can have 1000 models. In this case, the 1000 models corresponding to the AI-based CSI compressed feedback can be numbered 1-1000, and the 1000 models corresponding to the AI-based temporal CSI prediction can be numbered 1001-2000. Thus, assuming the network device receives a second model identifier of 500, it can determine the AI ​​model numbered 500 from the models corresponding to the AI-based CSI compressed feedback, thereby accurately determining the AI / ML models supported by the terminal. Subsequently, the network device can perform lifecycle management on this model.

[0146] In some implementations, when the feature types supported by the terminal include AI-based CSI compressed feedback, the first functional information includes one or more of the following:

[0147] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0148] In some implementations, when the feature types supported by the terminal include AI-based temporal CSI prediction, the first functional information includes one or more of the following:

[0149] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0150] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0151] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0152] The third function information includes any of the following:

[0153] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0154] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0155] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0156] The third function information includes any of the following:

[0157] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0158] In some implementations, the time-domain requirements for prediction may include the number of predicted CSIs and the time interval between two adjacent predicted CSIs.

[0159] In some implementations, when the terminal supports feature types including AI-based CSI compressed feedback, as can be seen from the foregoing, the generation part model in the terminal and the recovery part model in the network device constitute a bilateral AI / ML model. In this case, the first model identifier may include the model ID and / or the model pairing ID, the second model identifier includes the model ID and / or the model pairing ID, and the third model identifier includes the model ID and / or the model pairing ID. The model pairing ID can be referred to as the Paring model ID.

[0160] In some implementations, when the feature types supported by the terminal include AI-based temporal prediction, the first model identifier includes the model ID, the second model identifier includes the model ID, and the third model identifier includes the model ID.

[0161] Figure 2b This is a flowchart illustrating a model recognition method according to some embodiments. For example... Figure 2b As shown, this model recognition method can be used in a terminal and includes the following steps.

[0162] In step S210b, a first instruction sent by the network device is received.

[0163] In step S220b, first information is sent to the network device according to the first instruction.

[0164] In some implementations, the first information is used by the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal.

[0165] In some implementations, the first instruction is used to request AI / ML functions and / or AI / ML models supported by the terminal.

[0166] In some implementations, the first instruction is used to request the terminal to return first information.

[0167] In some implementations, the first information may include a feature type, thereby allowing the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal corresponding to the feature type.

[0168] In some implementations, the first information may include one or more of the following:

[0169] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0170] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0171] The first functional information describes the AI / ML functions supported by the terminal; or,

[0172] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0173] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0174] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0175] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0176] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0177] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0178] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0179] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0180] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0181] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0182] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0183] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0184] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0185] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0186] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0187] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0188] The third function information includes any of the following:

[0189] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0190] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0191] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0192] The third function information includes any of the following:

[0193] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0194] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes the model ID and / or model pairing ID, the second model identifier includes the model ID and / or model pairing ID, and the third model identifier includes the model ID and / or model pairing ID.

[0195] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0196] This disclosure specifies that a network device can send a first instruction to a terminal, and the terminal, upon receiving the first instruction, can send first information to the network device according to the first instruction. That is, this disclosure specifies the conditions or timing for the terminal to send the first instruction to the network device.

[0197] The detailed description of the relevant steps or parameter definitions in the embodiments of this disclosure can be found in the foregoing content, and will not be repeated here.

[0198] Figure 2c This is a flowchart illustrating a model recognition method according to some embodiments. For example... Figure 2c As shown, this model recognition method can be used in a terminal and includes the following steps.

[0199] In step S210c, a second instruction sent by the network device is received.

[0200] In step S220c, according to the second instruction, the first information corresponding to the feature type is sent to the network device.

[0201] In some implementations, the second instruction includes a feature type.

[0202] In some implementations, the first information corresponding to the feature type is used to determine the AI / ML functions and / or AI / ML models supported by the terminal for the corresponding feature type.

[0203] In some implementations, the second instruction is used to request AI / ML functions and / or AI / ML models supported by the terminal for the corresponding feature type.

[0204] In some implementations, the first instruction is used to request the terminal to return first information of the corresponding feature type.

[0205] In some implementations, the first information may include one or more of the following:

[0206] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0207] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0208] The first functional information describes the AI / ML functions supported by the terminal; or,

[0209] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0210] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0211] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0212] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0213] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0214] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0215] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0216] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0217] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0218] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0219] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0220] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0221] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0222] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0223] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0224] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0225] The third function information includes any of the following:

[0226] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0227] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0228] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0229] The third function information includes any of the following:

[0230] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0231] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes the model ID and / or model pairing ID, the second model identifier includes the model ID and / or model pairing ID, and the third model identifier includes the model ID and / or model pairing ID.

[0232] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0233] This disclosure specifies that a network device can send a second instruction to a terminal, and the terminal, upon receiving the second instruction, can send first information of a corresponding feature type to the network device according to the second instruction. That is, this disclosure specifies that the network device can designate a terminal to send first information of a specified feature type, so that the network device can determine the AI / ML functions and / or AI / ML models supported by the terminal of the specified feature type.

[0234] The detailed description of the relevant steps or parameter definitions in the embodiments of this disclosure can be found in the foregoing content, and will not be repeated here.

[0235] Figure 3a This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 3a As shown, this model recognition method can be used in a terminal and includes the following steps.

[0236] In step S310a, first information is sent to the network device, the first information including first functional information.

[0237] In some implementations, the first functional information is used by the network device to determine the AI / ML functions supported by the terminal.

[0238] In some implementations, the first information may include not only the first functional information but also the terminal-associated metadata. The first functional information and the terminal-associated metadata are used by the network device to determine the AI / ML functions supported by the terminal.

[0239] For a detailed description of step S310a, please refer to the foregoing embodiments, and it will not be repeated here.

[0240] This disclosure specifies that a terminal can send first functional information to a network device, thereby allowing the network device to determine the AI / ML functions supported by the terminal based on the first functional information.

[0241] Figure 3b This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 3b As shown, this model recognition method can be used in a terminal and includes the following steps.

[0242] In step S310b, first information is sent to the network device. The first information includes second functional information and a first model identifier.

[0243] In some implementations, the first information includes second functional information and a first model identifier used by the network device to determine the AI / ML functions supported by the terminal.

[0244] In some implementations, the first information may include not only the second functional information and the first model identifier, but also the terminal-associated metadata. The second functional information, the first model identifier, and the terminal-associated metadata are used by the network device to determine the AI / ML functions supported by the terminal.

[0245] For a detailed description of step S310b, please refer to the foregoing embodiments, and it will not be repeated here.

[0246] This disclosure specifies that a terminal can send second functional information and a first model identifier to a network device, thereby enabling the network device to determine the AI / ML functions supported by the terminal based on the second functional information and the first model identifier.

[0247] Figure 3c This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 3c As shown, this model recognition method can be used in a terminal and includes the following steps.

[0248] In step S310c, first information is sent to the network device, the first information including the second model identifier.

[0249] In some implementations, the first information includes a second model identifier used by the network device to determine the AI / ML functions supported by the terminal.

[0250] In some implementations, the first information may include, in addition to the second model identifier, terminal-associated metadata. The second model identifier and the terminal-associated metadata are used by the network device to determine the AI / ML functions supported by the terminal.

[0251] For a detailed description of step S310c, please refer to the foregoing embodiments, and it will not be repeated here.

[0252] This disclosure specifies that a terminal can send a second model identifier to a network device, thereby allowing the network device to determine the AI / ML functions supported by the terminal based on the second model identifier.

[0253] Figure 3d This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 3d As shown, this model recognition method can be used in a terminal and includes the following steps.

[0254] In step S310d, first information is sent to the network device, the first information including metadata associated with the terminal.

[0255] In some implementations, the first information includes terminal-associated metadata used by the network device to determine the AI / ML functions supported by the terminal.

[0256] In some implementations, the first information includes terminal-associated metadata used by the network device to determine the AI / ML models supported by the terminal.

[0257] The detailed description of step S310d can be found in the foregoing embodiments, and will not be repeated here.

[0258] This disclosure specifies that a terminal can send terminal-associated metadata to a network device, thereby enabling the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal based on the terminal-associated metadata.

[0259] Figure 3e This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 3e As shown, this model recognition method can be used in a terminal and includes the following steps.

[0260] In step S310e, first information is sent to the network device, the first information including the second model identifier.

[0261] In some implementations, the first information includes a second model identifier used by the network device to determine the AI / ML models supported by the terminal.

[0262] In some implementations, the first information may include, in addition to the second model identifier, terminal-associated metadata. The second model identifier and the terminal-associated metadata are used by the network device to determine the AI / ML models supported by the terminal.

[0263] The detailed description of step S310e can be found in the foregoing embodiments, and will not be repeated here.

[0264] This disclosure specifies that a terminal can send a second model identifier to a network device, thereby enabling the network device to determine the AI / ML models supported by the terminal based on the second model identifier.

[0265] Figure 3f This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 3f As shown, this model recognition method can be used in a terminal and includes the following steps.

[0266] In step S310f, first information is sent to the network device. The first information includes third function information and third model identifier.

[0267] In some implementations, the first information includes third functional information and a third model identifier used by the network device to determine the AI / ML models supported by the terminal.

[0268] In some implementations, the first information may include not only the third function information and the third model identifier, but also the terminal-associated metadata. The third function information, the third model identifier, and the terminal-associated metadata are used by the network device to determine the AI / ML models supported by the terminal.

[0269] The detailed description of step S310f can be found in the foregoing embodiments, and will not be repeated here.

[0270] This disclosure specifies that a terminal can send third function information and a third model identifier to a network device, thereby enabling the network device to determine the AI / ML model supported by the terminal based on the third function information and the third model identifier.

[0271] Figure 4a This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 4a As shown, this model recognition method can be used in network devices and includes the following steps.

[0272] In step S410a, the receiving terminal sends the first information.

[0273] In step S420a, the AI / ML functions and / or AI / ML models supported by the terminal are determined based on the first information.

[0274] In some implementations, the terminal may send first information to the network device, thereby allowing the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal based on the first information.

[0275] In some implementations, the first information may include one or more of the following:

[0276] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0277] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0278] The first functional information describes the AI / ML functions supported by the terminal; or,

[0279] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0280] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0281] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0282] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0283] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0284] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0285] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0286] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0287] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0288] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0289] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0290] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0291] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0292] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0293] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0294] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0295] The third function information includes any of the following:

[0296] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0297] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0298] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0299] The third function information includes any of the following:

[0300] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0301] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes the model ID and / or model pairing ID, the second model identifier includes the model ID and / or model pairing ID, and the third model identifier includes the model ID and / or model pairing ID.

[0302] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0303] This disclosure specifies that a terminal can send first information to a network device through information reporting, thereby enabling the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal based on the first information, thus completing the model recognition of the terminal.

[0304] The detailed description of the relevant steps or parameter definitions in the embodiments of this disclosure can be found in the foregoing content, and will not be repeated here.

[0305] Figure 4b This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 4b As shown, this model recognition method can be used in network devices and includes the following steps.

[0306] In step S410b, the first instruction is sent to the terminal.

[0307] In some implementations, the first instruction is used by the terminal to send first information.

[0308] In step S420b, the receiving terminal sends the first information.

[0309] In step S430b, the AI / ML functions and / or AI / ML models supported by the terminal are determined based on the first information.

[0310] In some implementations, the first information may include one or more of the following:

[0311] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0312] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0313] The first functional information describes the AI / ML functions supported by the terminal; or,

[0314] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0315] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0316] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0317] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0318] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0319] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0320] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0321] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0322] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0323] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0324] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0325] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0326] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0327] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0328] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0329] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0330] The third function information includes any of the following:

[0331] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0332] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0333] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0334] The third function information includes any of the following:

[0335] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0336] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes the model ID and / or model pairing ID, the second model identifier includes the model ID and / or model pairing ID, and the third model identifier includes the model ID and / or model pairing ID.

[0337] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0338] This disclosure specifies that a network device can send a first instruction to a terminal, and the terminal, upon receiving the first instruction, can send first information to the network device according to the first instruction. That is, this disclosure specifies the conditions or timing for the terminal to send the first instruction to the network device.

[0339] The detailed description of the relevant steps or parameter definitions in the embodiments of this disclosure can be found in the foregoing content, and will not be repeated here.

[0340] Figure 4c This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 4b As shown, this model recognition method can be used in network devices and includes the following steps.

[0341] In step S410c, a second instruction is sent to the terminal.

[0342] In some implementations, the second instruction includes a feature type.

[0343] In some implementations, the second instruction is used by the terminal to send first information corresponding to the feature type.

[0344] In step S420c, the receiving terminal sends the first information corresponding to the feature type.

[0345] In step S430c, the AI / ML functions and / or AI / ML models supported by the terminal for the corresponding feature type are determined based on the first information of the corresponding feature type.

[0346] In some implementations, the first information may include one or more of the following:

[0347] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0348] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0349] The first functional information describes the AI / ML functions supported by the terminal; or,

[0350] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0351] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0352] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0353] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0354] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0355] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0356] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0357] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0358] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0359] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0360] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0361] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0362] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0363] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0364] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0365] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0366] The third function information includes any of the following:

[0367] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0368] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0369] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0370] The third function information includes any of the following:

[0371] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0372] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes the model ID and / or model pairing ID, the second model identifier includes the model ID and / or model pairing ID, and the third model identifier includes the model ID and / or model pairing ID.

[0373] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0374] This disclosure specifies that a network device can send a second instruction to a terminal, and the terminal, upon receiving the second instruction, can send first information of a corresponding feature type to the network device according to the second instruction. That is, this disclosure specifies that the network device can designate a terminal to send first information of a specified feature type, so that the network device can determine the AI / ML functions and / or AI / ML models supported by the terminal of the specified feature type.

[0375] The detailed description of the relevant steps or parameter definitions in the embodiments of this disclosure can be found in the foregoing content, and will not be repeated here.

[0376] Figure 4d This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 4d As shown, this model recognition method can be used in network devices and includes the following steps.

[0377] In step S410d, the receiving terminal sends the first information.

[0378] In step S420d, the AI / ML functions and / or AI / ML models supported by the terminal are determined based on the first information.

[0379] In step S430d, information for lifecycle management of the model on the terminal side is determined based on the AI / ML functions supported by the terminal and / or the AI / ML models supported by the terminal.

[0380] In some implementations, the first information may include one or more of the following:

[0381] First functional information, second functional information, third functional information, first model identifier, second model identifier, third model identifier, terminal associated metadata, feature type information.

[0382] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0383] The first functional information describes the AI / ML functions supported by the terminal; or,

[0384] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0385] The second model identifier is used by network devices to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0386] The metadata associated with the terminal is used by network devices to determine the parameter configurations of the AI / ML functions supported by the terminal.

[0387] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0388] The second model identifier is used by network devices to determine the AI / ML models supported by the terminal; or,

[0389] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0390] The metadata associated with the terminal is used by network devices to determine the parameter configuration of the AI / ML models supported by the terminal.

[0391] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0392] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0393] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0394] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0395] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0396] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0397] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0398] In some implementations, when the feature type supported by the terminal includes AI-based CSI compressed feedback, the second functional information includes any one of the following:

[0399] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0400] The third function information includes any of the following:

[0401] The model outputs the payload, channel scenario, network configuration parameters, and movement speed.

[0402] In some implementations, when the feature type supported by the terminal includes AI-based temporal CSI prediction, the second functional information includes any one of the following:

[0403] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0404] The third function information includes any of the following:

[0405] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0406] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes the model ID and / or model pairing ID, the second model identifier includes the model ID and / or model pairing ID, and the third model identifier includes the model ID and / or model pairing ID.

[0407] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0408] This disclosure specifies that a terminal can send first information to a network device through information reporting. The network device can then determine the AI / ML functions and / or AI / ML models supported by the terminal based on the first information, and further perform lifecycle management on the terminal-side models based on the AI / ML functions and / or AI / ML models supported by the terminal.

[0409] The detailed description of the relevant steps or parameter definitions in the embodiments of this disclosure can be found in the foregoing content, and will not be repeated here.

[0410] Figure 5a This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 5a As shown, this model recognition method can be used in network devices and includes the following steps.

[0411] In step S510a, the receiving terminal sends first information, which includes first function information.

[0412] In step S520a, the AI / ML functions supported by the terminal are determined based on the first function information.

[0413] In some implementations, the first information may include not only the first functional information but also the metadata associated with the terminal, so that the network device can determine the AI / ML functions supported by the terminal based on the first functional information and the metadata associated with the terminal.

[0414] For a detailed description of steps S510a and S520a, please refer to the foregoing embodiments, and they will not be repeated here.

[0415] This disclosure specifies that a terminal can send first functional information to a network device, thereby allowing the network device to determine the AI / ML functions supported by the terminal based on the first functional information.

[0416] Figure 5b This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 5b As shown, this model recognition method can be used in network devices and includes the following steps.

[0417] In step S510b, the receiving terminal sends first information, which includes second function information and a first model identifier.

[0418] In step S520b, the AI / ML functions supported by the terminal are determined based on the second function information and the first model identifier.

[0419] In some implementations, the first information may include not only the second functional information and the first model identifier, but also the metadata associated with the terminal. Thus, the network device can determine the AI / ML functions supported by the terminal based on the second functional information, the first model identifier, and the metadata associated with the terminal.

[0420] For a detailed description of steps S510b and S520b, please refer to the foregoing embodiments, and they will not be repeated here.

[0421] This disclosure specifies that a terminal can send second functional information and a first model identifier to a network device, thereby enabling the network device to determine the AI / ML functions supported by the terminal based on the second functional information and the first model identifier.

[0422] Figure 5c This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 5c As shown, this model recognition method can be used in network devices and includes the following steps.

[0423] In step S510c, the receiving terminal sends first information, which includes a second model identifier.

[0424] In step S520c, the AI / ML functions supported by the terminal are determined based on the second model identifier.

[0425] In some implementations, the first information may include not only the second model identifier but also metadata associated with the terminal, so that the network device can determine the AI / ML functions supported by the terminal based on the second model identifier and the metadata associated with the terminal.

[0426] For a detailed description of steps S510c and S520c, please refer to the foregoing embodiments, and they will not be repeated here.

[0427] This disclosure specifies that a terminal can send a second model identifier to a network device, thereby allowing the network device to determine the AI / ML functions supported by the terminal based on the second model identifier.

[0428] Figure 5d This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 5d As shown, this model recognition method can be used in network devices and includes the following steps.

[0429] In step S510d, the receiving terminal sends first information, which includes metadata associated with the terminal.

[0430] In step S520d, the AI / ML functions and / or AI / ML models supported by the terminal are determined based on the metadata associated with the terminal.

[0431] For a detailed description of steps S510d and S520d, please refer to the foregoing embodiments, and they will not be repeated here.

[0432] This disclosure specifies that a terminal can send terminal-associated metadata to a network device, thereby enabling the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal based on the terminal-associated metadata.

[0433] Figure 5e This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 5e As shown, this model recognition method can be used in network devices and includes the following steps.

[0434] In step S510e, the receiving terminal sends first information, which includes a second model identifier.

[0435] In step S520e, the AI / ML models supported by the terminal are determined based on the second model identifier.

[0436] In some implementations, the first information may include not only the second model identifier but also metadata associated with the terminal, so that the network device can determine the AI / ML model supported by the terminal based on the second model identifier and the metadata associated with the terminal.

[0437] For a detailed description of steps S510e and S520e, please refer to the foregoing embodiments, and they will not be repeated here.

[0438] This disclosure specifies that a terminal can send a second model identifier to a network device, thereby enabling the network device to determine the AI / ML models supported by the terminal based on the second model identifier.

[0439] Figure 5f This is a flowchart illustrating a model recognition method according to some other embodiments. For example... Figure 5f As shown, this model recognition method can be used in network devices and includes the following steps.

[0440] In step S510f, the receiving terminal sends first information, which includes third function information and third model identifier.

[0441] In step S520f, the AI / ML models supported by the terminal are determined based on the third function information and the third model identifier.

[0442] In some implementations, the first information may include not only the third function information and the third model identifier, but also the terminal-associated metadata. Thus, the network device can determine the AI / ML model supported by the terminal based on the third function information, the third model identifier, and the terminal-associated metadata.

[0443] For a detailed description of steps S510f and S520f, please refer to the foregoing embodiments, and they will not be repeated here.

[0444] This disclosure specifies that a terminal can send third function information and a third model identifier to a network device, thereby enabling the network device to determine the AI / ML model supported by the terminal based on the third function information and the third model identifier.

[0445] Figure 6 This is a flowchart illustrating a model recognition method according to some embodiments. For example... Figure 6 As shown, this model recognition method can be used in communication systems, which include terminals and network devices. The method includes the following steps:

[0446] In step S610, the terminal sends the first information to the network device.

[0447] In step S620, the network device determines the AI / ML functions and / or AI / ML models supported by the terminal based on the first information.

[0448] For a detailed description of steps S610 and S620, please refer to the foregoing embodiments, and they will not be repeated here.

[0449] The following is yet another embodiment provided in this disclosure, in which:

[0450] The base station requires the UE to report the AI / ML functions supported by its AI-based CSI compressed feedback. Therefore, the base station sends a UE capability query command to the UE, such as the UECapabilityEnquiry command, to query which functions and / or AI / ML functions the UE supports. Upon receiving this query command, the UE reports UECapabilityInformation to the base station, indicating which AI / ML functions the UE supports, and also indicating the model ID included under that function. For example, AI-based CSI compressed feedback for the UE is considered a feature type for that UE. The UE indicates in its capability report that it supports AI-based CSI compressed feedback in UMa or UMi scenarios, and also supports AI-based CSI compressed feedback in medium- and high-speed mobile scenarios. These two functions are defined as AI Function ID1 and AI Function ID2, respectively. Furthermore, the CSI compressed feedback function supporting UMA or UMi scenarios includes an AI / ML model with a maximum payload size of 120 bits after quantization of the maximum compressed information for inference, and an AI / ML model with a maximum payload size of 60 bits. The IDs of these two models are defined as model ID1 and model ID2. Correspondingly, the AI-based CSI compressed feedback function supporting high-speed mobile scenarios also includes two AI / ML models: an AI / ML model with a maximum payload size of 120 bits after quantization of the maximum compressed information for inference, and two AI / ML models with a maximum payload size of 60 bits. The IDs of these two models are assumed and defined as model ID3 and model ID4. For example... Figure 7 As shown, the base station can determine the AI / ML functions supported by the UE based on the AI / ML functions and AI / ML model ID reported by the UE.

[0451] In some implementations, after determining the AI / ML functions supported by the UE, the base station activates a specific model ID under one of the functions via RRC signaling, based on the current environment or application conditions of the UE. Alternatively, it may first activate one function via RRC signaling and then activate a specific AI / ML model ID under that function via other signaling such as MAC-CE or DCI.

[0452] The following is yet another embodiment provided in this disclosure, in which:

[0453] The base station sends a UE capability query command to the UE, such as the UECapabilityEnquiry command, to directly query which AI / ML models the UE supports. Each model is associated with a mode ID. This mode ID is then associated with a specific function. Assuming that each mode ID is predefined to be associated with a specific function, in this case, the UE does not need to report the entire function, but only reports the supported mode IDs. For example, the UE reports only mode ID1 and mode ID4 through capability reporting, and based on... Figure 7 As shown, the predefined configurations indicate that model ID1 and model ID4 respectively support AI inference with a maximum payload of 120 bits in UMa / UMi scenarios and AI inference with a maximum payload of 60 bits in medium-to-high-speed mobile scenarios. By using the model ID reported by the UE and combining it with the predefined specifications, the base station can determine which specific AI inference the UE supports, and thus configure appropriate parameters for AI inference on the UE.

[0454] The following is yet another embodiment provided in this disclosure, in which:

[0455] The base station requires the UE to report the AI / ML functions supported by its AI-based temporal CSI prediction. The base station sends a UE capability query command, such as the UECapabilityEnquiry command, to the UE to query which AI / ML functions the UE supports. Upon receiving this query command, the UE reports UEcapabilityInformation to the base station, indicating which AI / ML functions it supports. For example, the UE reports AI / ML function 1 and AI / ML function 2. Assume AI / ML function 1 is defined as the UE being able to predict channel information on a maximum of 10 slots with an interval of 2 between adjacent slots, and can be used for inference below 30 km / h; AI / ML function 2 is defined as the UE being able to infer channel information on a maximum of 20 slots below 60 km / h with an interval of 5 between adjacent slots. Therefore, the base station can determine the AI / ML functions supported by the UE based on the information reported by the UE.

[0456] In some implementations, the base station indicates which AI / ML function to activate based on the UE's moving speed and the requirement for time-domain CSI prediction. Assuming the current UE's speed is 15 km / h and it needs to predict time slots on slot 5, with 2 adjacent slots, the base station can directly instruct the UE to activate AI / ML function 1 via RRC signaling. The RRC information can implicitly instruct the UE to activate AI / ML function 1 by configuring the number of CSI-RS resources or the interval between adjacent time slots to 2 slots.

[0457] It should be noted that the use of "or" to separate multiple embodiments or features, even if some of the features are not feasible, will not affect other solutions. Furthermore, unless contradictory, the embodiments of this disclosure can be combined with other embodiments or implementation methods and their various optional solutions involved in model recognition methods, which will not be elaborated further here.

[0458] Figure 8 This is a block diagram illustrating a model recognition device 800 according to some embodiments. (Refer to...) Figure 8 The model recognition device 800 is applied to a terminal, and the model recognition device 800 may include:

[0459] The first sending module 810 is configured to send first information to a network device, the first information being used by the network device to determine the AI / ML functions and / or AI / ML models supported by the terminal.

[0460] In some implementations, the first transmitting module 810 includes:

[0461] The first receiving submodule is configured to receive a first instruction sent by the network device;

[0462] The first sending submodule is configured to send first information to the network device according to the first instruction.

[0463] In some implementations, the first transmitting module 810 includes:

[0464] The second receiving submodule is configured to receive a second instruction sent by the network device, the second instruction including a feature type;

[0465] The second sending submodule is configured to send first information corresponding to the feature type to the network device according to the second instruction. The first information corresponding to the feature type is used to determine the AI / ML functions and / or AI / ML models supported by the terminal corresponding to the feature type.

[0466] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0467] The first functional information describes the AI / ML functions supported by the terminal; or,

[0468] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0469] The second model identifier is used by the network device to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0470] The metadata associated with the terminal is used by the network device to determine the parameter configuration of the AI / ML functions supported by the terminal.

[0471] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0472] A second model identifier is used by the network device to determine the AI / ML models supported by the terminal; or,

[0473] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0474] The metadata associated with the terminal is used by the network device to determine the parameter configuration of the AI / ML models supported by the terminal.

[0475] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0476] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0477] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0478] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0479] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0480] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0481] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0482] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the second functional information includes any of the following:

[0483] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0484] The third function information includes any of the following:

[0485] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0486] The terminal supports feature types including AI-based temporal CSI prediction, and the second functional information includes any one of the following:

[0487] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0488] The third function information includes any of the following:

[0489] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0490] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes a model ID and / or a model pairing ID, the second model identifier includes a model ID and / or a model pairing ID, and the third model identifier includes a model ID and / or a model pairing ID.

[0491] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0492] Figure 9 This is a block diagram illustrating a model recognition device 900 according to some embodiments. (Refer to...) Figure 9 The model recognition device 900 is applied to network equipment, and the model recognition device 900 may include:

[0493] The receiving module 910 is configured to receive the first information sent by the terminal;

[0494] The first determining module 920 is configured to determine, based on the first information, the AI / ML functions supported by the terminal and / or the AI / ML models supported by the terminal.

[0495] In some embodiments, the device 900 further includes:

[0496] The second sending module is configured to send a first instruction to the terminal, the first instruction being used by the terminal to send the first information.

[0497] In some embodiments, the device 900 further includes:

[0498] The third sending module is configured to send a second instruction to the terminal. The second instruction includes a feature type, and the second instruction is used by the terminal to send first information corresponding to the feature type.

[0499] Accordingly, the receiving module 910 is also configured to receive first information corresponding to the feature type sent by the terminal.

[0500] In some embodiments, the device 900 further includes:

[0501] The second determining module is configured to determine information for lifecycle management of the model on the terminal side based on the AI / ML functions supported by the terminal and / or the AI / ML models supported by the terminal.

[0502] In some implementations, the first information is used by the network device to determine the AI / ML functions supported by the terminal, and the first information includes:

[0503] The first functional information describes the AI / ML functions supported by the terminal; or,

[0504] The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the terminal; or,

[0505] The second model identifier is used by the network device to determine the AI / ML functions supported by the terminal based on a preset correspondence; or,

[0506] The metadata associated with the terminal is used by the network device to determine the parameter configuration of the AI / ML functions supported by the terminal.

[0507] In some implementations, the first information is used by the network device to determine the AI / ML models supported by the terminal, and the first information includes:

[0508] A second model identifier is used by the network device to determine the AI / ML models supported by the terminal; or,

[0509] The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the terminal; or

[0510] The metadata associated with the terminal is used by the network device to determine the parameter configuration of the AI / ML models supported by the terminal.

[0511] In some implementations, the metadata associated with the terminal includes one or more of the following:

[0512] The terminal's corresponding transmit antenna port, the terminal's corresponding receive antenna port, the terminal's carrier information, the number of subbands supported by the terminal, and the terminal's location information.

[0513] In some implementations, the range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the terminal, the functional information supported by the terminal, and the metadata associated with the terminal.

[0514] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the first functional information includes one or more of the following:

[0515] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0516] The terminal supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following:

[0517] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0518] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, and the second functional information includes any of the following:

[0519] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0520] The third function information includes any of the following:

[0521] The model outputs the payload, channel scenario, network configuration parameters, and moving speed.

[0522] The terminal supports feature types including AI-based temporal CSI prediction, and the second functional information includes any one of the following:

[0523] Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed;

[0524] The third function information includes any of the following:

[0525] The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

[0526] In some implementations, the terminal supports feature types including AI-based CSI compressed feedback, where the first model identifier includes a model ID and / or a model pairing ID, the second model identifier includes a model ID and / or a model pairing ID, and the third model identifier includes a model ID and / or a model pairing ID.

[0527] The terminal supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

[0528] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0529] This disclosure also provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the steps of the model recognition method provided in this disclosure.

[0530] Figure 10 This is a block diagram illustrating a terminal according to some embodiments. For example, terminal 1000 may be a mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, medical device, fitness equipment, personal digital assistant, smart car, etc.

[0531] Reference Figure 10 The terminal 1000 may include one or more of the following components: a first processing component 1002, a first memory 1004, a first power supply component 1006, a multimedia component 1008, an audio component 1010, a first input / output interface 1012, a sensor component 1014, and a communication component 1016.

[0532] The first processing component 1002 typically controls the overall operation of the terminal 1000, such as operations associated with display, telephone calls, data communication, camera operation, and recording. The first processing component 1002 may include one or more first processors 1020 to execute instructions to complete all or part of the steps of the model recognition method described above. Furthermore, the first processing component 1002 may include one or more modules to facilitate interaction between the first processing component 1002 and other components. For example, the first processing component 1002 may include a multimedia module to facilitate interaction between the multimedia component 1008 and the first processing component 1002.

[0533] The first memory 1004 is configured to store various types of data to support operation on the terminal 1000. Examples of this data include instructions for any application or method operating on the terminal 1000, contact data, phonebook data, messages, pictures, videos, etc. The first memory 1004 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0534] The first power supply component 1006 provides power to various components of the terminal 1000. The first power supply component 1006 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the terminal 1000.

[0535] The multimedia component 1008 includes a screen that provides an output interface between the terminal 1000 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touchscreen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensors may sense not only the boundaries of the touch or swipe action but also the duration and pressure associated with the touch or swipe operation. In some embodiments, the multimedia component 1008 includes a front-facing camera and / or a rear-facing camera. When the terminal 1000 is in an operating mode, such as a shooting mode or a video mode, the front-facing camera and / or the rear-facing camera may receive external multimedia data. Each front-facing camera and rear-facing camera may be a fixed optical lens system or have focal length and optical zoom capabilities.

[0536] Audio component 1010 is configured to output and / or input audio signals. For example, audio component 1010 includes a microphone (MIC) configured to receive external audio signals when terminal 1000 is in an operating mode, such as call mode, recording mode, and voice recognition mode. The received audio signals may be further stored in first memory 1004 or transmitted via communication component 1016. In some embodiments, audio component 1010 also includes a speaker for outputting audio signals.

[0537] The first input / output interface 1012 provides an interface between the first processing component 1002 and the peripheral interface module. The peripheral interface module may be a keyboard, click wheel, buttons, etc. These buttons may include, but are not limited to, a home button, volume buttons, a start button, and a lock button.

[0538] Sensor assembly 1014 includes one or more sensors for providing state assessments of various aspects of terminal 1000. For example, sensor assembly 1014 may detect the on / off state of terminal 1000, the relative positioning of components such as the display and keypad of terminal 1000, changes in position of terminal 1000 or a component of terminal 1000, the presence or absence of user contact with terminal 1000, the orientation or acceleration / deceleration of terminal 1000, and temperature changes of terminal 1000. Sensor assembly 1014 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. Sensor assembly 1014 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, sensor assembly 1014 may also include an accelerometer, a gyroscope, a magnetometer, a pressure sensor, or a temperature sensor.

[0539] Communication component 1016 is configured to facilitate wired or wireless communication between terminal 1000 and other devices. Terminal 1000 can access wireless networks based on communication standards, such as WiFi, 2G, or 3G, or combinations thereof. In one exemplary embodiment, communication component 1016 receives broadcast signals or broadcast-related information from an external broadcast management system via a broadcast channel. In one exemplary embodiment, communication component 1016 further includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0540] In an exemplary embodiment, the terminal 1000 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the model recognition method described above.

[0541] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a first memory 1004 including instructions, which can be executed by a first processor 1020 of a terminal 1000 to complete the aforementioned model recognition method. For example, the non-transitory computer-readable storage medium may be a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc.

[0542] The aforementioned device can be a standalone electronic device or a part of a standalone electronic device. For example, in one embodiment, the device can be an integrated circuit (IC) or a chip, wherein the integrated circuit can be a single IC or a collection of multiple ICs. The chip can include, but is not limited to, the following types: GPU (Graphics Processing Unit), CPU (Central Processing Unit), FPGA (Field Programmable Gate Array), DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), and SoC (System on Chip). The aforementioned integrated circuit or chip can be used to execute executable instructions (or code) to implement the aforementioned model recognition method. The executable instructions can be stored in the integrated circuit or chip or obtained from other devices or equipment. For example, the integrated circuit or chip includes a processor, memory, and an interface for communicating with other devices. The executable instructions can be stored in the memory, and when the executable instructions are executed by the processor, the above-mentioned model recognition method can be implemented; or, the integrated circuit or chip can receive the executable instructions through the interface and transmit them to the processor for execution to implement the above-mentioned model recognition method.

[0543] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the model recognition method described above when executed by the programmable device.

[0544] Figure 11 This is a block diagram illustrating a network device according to an exemplary embodiment. For example, network device 1100 may be provided as another logical network entity in a core network or access network. (See also...) Figure 11 The network device 1100 includes a processing component 1122, which further includes one or more processors, and memory resources represented by memory 1132 for storing instructions, such as application programs, that can be executed by the processing component 1122. The application programs stored in memory 1132 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1122 is configured to execute instructions to perform the steps of the communication method provided in the above-described method embodiments.

[0545] Network device 1100 may also include a power supply component 1126 configured to perform power management of device 1100, a wired or wireless network interface 1150 configured to connect network device 1100 to a network, and an input / output interface 1158. Network device 1100 can operate on an operating system stored in memory 1132, such as Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, or similar.

[0546] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0547] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A model recognition method, characterized in that, Performed by a user equipment (UE), the method includes: Receive a second instruction sent by a network device, the second instruction including a feature type; According to the second instruction, first information is sent to the network device. The first information corresponds to the feature type. The first information is used to determine the artificial intelligence (AI) / machine learning (ML) functions supported by the UE corresponding to the feature type and / or the AI / ML models supported by the UE. The first information includes at least one of the following: The first functional information describes the AI / ML functions supported by the UE; The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the UE. The second model identifier is used by the network device to determine the AI / ML functions supported by the UE based on a preset correspondence, and / or by the network device to determine the AI / ML models supported by the UE; The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the UE; The metadata associated with the UE is used by the network device to determine the parameter configuration of the AI / ML model supported by the UE and / or the parameter configuration of the AI / ML function. The metadata associated with the UE includes one or more of the following: The UE's corresponding transmit antenna port, the UE's corresponding receive antenna port, the UE's carrier information, the number of subbands supported by the UE, and the UE's location information.

2. The method according to claim 1, characterized in that, The range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the UE, the functional information supported by the UE, and the metadata associated with the UE.

3. The method according to claim 1, characterized in that, The UE supports feature types including AI-based Channel State Information (CSI) compressed feedback, where the first functional information includes one or more of the following: The model outputs the payload, channel scenario, network configuration parameters, and moving speed. The UE supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following: The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

4. The method according to claim 1, characterized in that, The UE supports feature types including AI-based CSI compressed feedback, and the second functional information includes any of the following: The model outputs the payload, channel scenario, network configuration parameters, and moving speed. The third functional information includes any of the following: The model outputs the payload, channel scenario, network configuration parameters, and moving speed. The UE supports feature types including AI-based temporal CSI prediction, and the second functional information includes any one of the following: Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed; The third functional information includes any of the following: The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

5. The method according to claim 1, characterized in that, The UE supports the following feature types: AI-based CSI compressed feedback, the first model identifier includes model identifier ID and / or model pair ID, the second model identifier includes model ID and / or model pair ID, and the third model identifier includes model ID and / or model pair ID. The UE supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

6. A model recognition method, characterized in that, Performed by a network device, the method includes: A second instruction sent to the user equipment (UE), the second instruction including a feature type; Receive first information corresponding to the feature type sent by the UE; Based on the first information, determine the AI / ML functions supported by the UE corresponding to the feature type and / or the AI / ML models supported by the UE; The first information includes at least one of the following: The first functional information describes the AI / ML functions supported by the UE; The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the UE. The second model identifier is used by the network device to determine the AI / ML functions supported by the UE based on a preset correspondence, and / or by the network device to determine the AI / ML models supported by the UE; The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the UE; The metadata associated with the UE is used by the network device to determine the parameter configuration of the AI / ML model supported by the UE and / or the parameter configuration of the AI / ML function. The metadata associated with the UE includes one or more of the following: The UE's corresponding transmit antenna port, the UE's corresponding receive antenna port, the UE's carrier information, the number of subbands supported by the UE, and the UE's location information.

7. The method according to claim 6, characterized in that, The method further includes: Based on the AI / ML functions supported by the UE and / or the AI / ML models supported by the UE, determine the information used for lifecycle management of the models on the UE side.

8. The method according to claim 6, characterized in that, The range of values ​​for the second model identifier is divided by one or more of the following information: the feature types supported by the UE, the functional information supported by the UE, and the metadata associated with the UE.

9. The method according to claim 6, characterized in that, The UE supports feature types including AI-based Channel State Information (CSI) compressed feedback, where the first functional information includes one or more of the following: The model outputs the payload, channel scenario, network configuration parameters, and moving speed. The UE supports feature types including AI-based temporal CSI prediction, and the first functional information includes one or more of the following: The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

10. The method according to claim 6, characterized in that, The UE supports the following feature types: AI-based CSI compressed feedback. The second functional information includes any of the following: The model outputs the payload, channel scenario, network configuration parameters, and moving speed. The third function information includes any of the following: The model outputs the payload, channel scenario, network configuration parameters, and moving speed. The UE supports the following feature types: AI-based temporal CSI prediction. The second functional information includes any one of the following: Predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed; The third function information includes any of the following: The predicted time-domain conditions, channel scenarios, network configuration parameters, and mobile speed are all considered.

11. The method according to claim 6, characterized in that, The UE supports the following feature types: AI-based CSI compressed feedback, the first model identifier includes model identifier ID and / or model pair ID, the second model identifier includes model ID and / or model pair ID, and the third model identifier includes model ID and / or model pair ID. The UE supports feature types including AI-based temporal prediction, with the first model identifier including the model ID, the second model identifier including the model ID, and the third model identifier including the model ID.

12. A model management method, characterized in that, Performed by a communication system, the communication system including a user equipment (UE) and network equipment, the method includes: The network device sends a second instruction to the UE, the second instruction including a feature type; The UE sends first information corresponding to the feature type to the network device; The network device determines, based on the first information, the AI / ML functions and / or AI / ML models supported by the UE corresponding to the feature type; The first information includes at least one of the following: The first functional information describes the AI / ML functions supported by the UE; The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the UE. The second model identifier is used by the network device to determine the AI / ML functions supported by the UE based on a preset correspondence, and / or by the network device to determine the AI / ML models supported by the UE; The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the UE; The metadata associated with the UE is used by the network device to determine the parameter configuration of the AI / ML model supported by the UE and / or the parameter configuration of the AI / ML function. The metadata associated with the UE includes one or more of the following: The UE's corresponding transmit antenna port, the UE's corresponding receive antenna port, the UE's carrier information, the number of subbands supported by the UE, and the UE's location information.

13. A model recognition device, characterized in that, include: The first receiving module is configured to receive a second instruction sent by a network device, the second instruction including a feature type; The first sending module is configured to send first information to the network device according to the second instruction. The first information corresponds to the feature type. The first information is used to determine the artificial intelligence (AI) / machine learning (ML) functions supported by the user equipment (UE) corresponding to the feature type and / or the AI / ML models supported by the UE. The first information includes at least one of the following: The first functional information describes the AI / ML functions supported by the UE; The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the UE. The second model identifier is used by the network device to determine the AI / ML functions supported by the UE based on a preset correspondence, and / or by the network device to determine the AI / ML models supported by the UE; The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the UE; The metadata associated with the UE is used by the network device to determine the parameter configuration of the AI / ML model supported by the UE and / or the parameter configuration of the AI / ML function. The metadata associated with the UE includes one or more of the following: The UE's corresponding transmit antenna port, the UE's corresponding receive antenna port, the UE's carrier information, the number of subbands supported by the UE, and the UE's location information.

14. A model recognition device, characterized in that, include: The second transmitting module is configured to transmit a second instruction to a user equipment (UE), the second instruction including a feature type; The second receiving module is configured to receive first information corresponding to the feature type sent by the UE; The first determining module is configured to determine, based on the first information, the artificial intelligence (AI) / machine learning (ML) functions supported by the UE corresponding to the feature type and / or the AI / ML models supported by the UE; The first information includes at least one of the following: The first functional information describes the AI / ML functions supported by the UE; The second functional information and the first model identifier are used by the network device to determine the AI / ML functions supported by the UE. The second model identifier is used by the network device to determine the AI / ML functions supported by the UE based on a preset correspondence, and / or by the network device to determine the AI / ML models supported by the UE; The third functional information and the third model identifier are used by the network device to determine the AI / ML models supported by the UE; The metadata associated with the UE is used by the network device to determine the parameter configuration of the AI / ML model supported by the UE and / or the parameter configuration of the AI / ML function. The metadata associated with the UE includes one or more of the following: The UE's corresponding transmit antenna port, the UE's corresponding receive antenna port, the UE's carrier information, the number of subbands supported by the UE, and the UE's location information.

15. A communication device, characterized in that, include: One or more processors; The processor is used to invoke computer instructions to cause the communication device to execute the model recognition method according to any one of claims 1-11.

16. A communication system, characterized in that, The device includes a user equipment (UE) and a network device, wherein the UE is configured to implement the model recognition method according to any one of claims 1-5, and the network device is configured to implement the model recognition method according to any one of claims 6-11.

17. A computer-readable storage medium storing computer instructions, characterized in that, When the computer instructions are executed on the communication device, the communication device performs the model recognition method as described in any one of claims 1-11.

Images