Method and apparatus for artificial intelligence (AI) / machine learning (ML) operation in wireless networks

Abstract

A method performed by a User Equipment (UE) for performing an Artificial Intelligence (AI) / Machine Learning (ML) operation is provided. The method receives, from a Base Station (BS), a first message granting permission for the UE to report AI / ML functionalities for autonomous activation. The method transmits, to the BS, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.　

Description

METHOD AND APPARATUS FOR ARTIFICIAL INTELLIGENCE (AI) / MACHINE LEARNING (ML) OPERATION IN WIRELESS NETWORKS

[0001] The present disclosure is related to wireless communication and, more specifically, to a User Equipment (UE), Base Station (BS), and method for performing an Artificial Intelligence (AI) / Machine Learning (ML) operation in the wireless communication networks.

[0002] Various efforts have been made to improve different aspects of wireless communication for the cellular wireless communication systems, such as the 5thGeneration (5G) New Radio (NR), by improving data rate, latency, reliability, and mobility. The 5G NR system is designed to provide flexibility and configurability to optimize network services and types, accommodating various use cases, such as enhanced Mobile Broadband (eMBB), massive Machine-Type Communication (mMTC), and Ultra-Reliable and Low-Latency Communication (URLLC). As the demand for radio access continues to grow, however, there exists a need for further improvements in the next-generation wireless communication systems, such as improvements in AI / ML operations.

[0003] The present disclosure is related to a UE, a BS, and a method for performing an AI / ML operation in the wireless communication networks.

[0004] In a first aspect of the present disclosure, a UE for performing an AI / ML operation is provided. The UE includes at least one processor and at least one non-transitory computer-readable medium that is coupled to the at least one processor and that stores one or more computer-executable instructions. The computer-executable instructions, when executed by the at least one processor, cause the UE to: receive, from a BS, a first message granting permission for the UE to report AI / ML functionalities for autonomous activation; and transmit, to the BS, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.

[0005] In some implementations of the first aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to activate the at least one AI / ML functionality indicated in the second message upon transmitting the second message.

[0006] In some implementations of the first aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: receive, from the BS, a third message including a configuration associated with the at least one AI / ML functionality indicated in the second message; and activate the at least one AI / ML functionality indicated in the second message upon receiving the third message.

[0007] In some implementations of the first aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: receive, from the BS, a third message including a configuration associated with the at least one AI / ML functionality indicated in the second message, the configuration indicating an activation status associated with a functionality identifier (ID); and activate an AI / ML functionality corresponding to the functionality ID based on the activation status upon receiving the third message.

[0008] In some implementations of the first aspect, the first message, the second message, and the third message are transmitted via Radio Resource Control (RRC) signaling.

[0009] In some implementations of the first aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: receive, from the BS, a fourth message including an identifier (ID) field and an activation / deactivation field; and activate or deactivate an AI / ML functionality corresponding to the ID field based on the activation / deactivation field.

[0010] In some implementations of the first aspect, the fourth message includes a Medium Access Control (MAC) Control Element (CE).

[0011] In some implementations of the first aspect, each AI / ML functionality corresponds to at least one AI / ML model, each AI / ML model corresponds to at least one additional condition (AC), and each AC corresponds to a condition under which data is collected for the corresponding AI / ML model.

[0012] In a second aspect of the present application, a BS for configuring a beam management operation is provided. The BS includes at least one processor and at least one non-transitory computer-readable medium that is coupled to the at least one processor and that stores one or more computer-executable instructions. The computer-executable instructions, when executed by the at least one processor, cause the BS to: transmit, to a UE, a first message granting permission for the UE to report AI / ML functionalities for autonomous activation; and receive, from the UE, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.

[0013] In some implementations of the second aspect, the UE activates the at least one AI / ML functionality indicated in the second message upon transmitting the second message.

[0014] In some implementations of the second aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to transmit, to the UE, a third message including a configuration associated with the at least one AI / ML functionality indicated in the second message. The UE activates the at least one AI / ML functionality indicated in the second message upon receiving the third message.

[0015] In some implementations of the second aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to transmit, to the UE, a third message including a configuration associated with the at least one AI / ML functionality indicated in the second message, the configuration indicating an activation status associated with a functionality ID. The UE activates an AI / ML functionality corresponding to the functionality ID based on the activation status upon receiving the third message.

[0016] In some implementations of the second aspect, the first message, the second message, and the third message are transmitted via RRC signaling.

[0017] In some implementations of the second aspect, the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to transmit, to the UE, a fourth message including an ID field and an activation / deactivation field. The UE activates or deactivates an AI / ML functionality corresponding to the ID field based on the activation / deactivation field.

[0018] In a third aspect of the present application, a method performed by a UE for performing a beam management operation is provided. The method includes receiving, from a BS, a first message granting permission for the UE to report AI / ML functionalities for autonomous activation; and transmitting, to the BS, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.

[0019] Aspects of the present disclosure are best understood from the following detailed disclosure when read with the accompanying drawings. Various features are not drawn to scale. Dimensions of various features may be arbitrarily increased or reduced for clarity of discussion.

[0020] FIG. 1 is a flowchart illustrating a method / process performed by a UE for performing an AI / ML operation, according to an example implementation of the present disclosure.

[0021] FIG. 2 is a flowchart illustrating a method / process performed by a BS for performing an AI / ML operation, according to an example implementation of the present disclosure.

[0022] FIG. 3 is a block diagram illustrating a node for wireless communication, according to an example implementation of the present disclosure.

[0023] The following contains specific information related to implementations of the present disclosure. The drawings and their accompanying detailed disclosure are merely directed to implementations. However, the present disclosure is not limited to these implementations. Other variations and implementations of the present disclosure will be obvious to those skilled in the art.

[0024] Unless noted otherwise, like or corresponding elements among the drawings may be indicated by like or corresponding reference numerals. Moreover, the drawings and illustrations in the present disclosure are generally not to scale and are not intended to correspond to actual relative dimensions.

[0025] For the purposes of consistency and ease of understanding, like features may be identified (although, in some examples, not illustrated) by the same numerals in the drawings. However, the features in different implementations may be different in other respects and may not be narrowly confined to what is illustrated in the drawings.

[0026] References to “one implementation,” “an implementation,” “example implementation,” “various implementations,” “some implementations,” “implementations of the present application,” etc., may indicate that the implementation(s) of the present application so described may include a particular feature, structure, or characteristic, but not every possible implementation of the present application necessarily includes the particular feature, structure, or characteristic. Further, repeated use of the phrase “In some implementations,” or “in an example implementation,” “an implementation,” do not necessarily refer to the same implementation, although they may. Moreover, any use of phrases like “implementations” in connection with “the present application” are never meant to characterize that all implementations of the present application must include the particular feature, structure, or characteristic, and should instead be understood to mean “at least some implementations of the present application” includes the stated particular feature, structure, or characteristic. The term “coupled” is defined as connected, whether directly or indirectly through intervening components, and is not necessarily limited to physical connections. The term “comprising,” when utilized, means “including, but not necessarily limited to”; it specifically indicates open-ended inclusion or membership in the so-described combination, group, series, and the equivalent.

[0027] The expression “at least one of A, B and C” or “at least one of the following: A, B and C” means “only A, or only B, or only C, or any combination of A, B and C.” The terms “system” and “network” may be used interchangeably. The term “and / or” is only an association relationship for describing associated objects and represents that three relationships may exist such that A and / or B may indicate that A exists alone, A and B exist at the same time, or B exists alone. The character “ / ” generally represents that the associated objects are in an “or” relationship.

[0028] For the purposes of explanation and non-limitation, specific details, such as functional entities, techniques, protocols, and standards, are set forth for providing an understanding of the disclosed technology. In other examples, detailed disclosure of well-known methods, technologies, systems, and architectures are omitted so as not to obscure the present disclosure with unnecessary details.

[0029] Persons skilled in the art will immediately recognize that any network function(s) or algorithm(s) disclosed may be implemented by hardware, software, or a combination of software and hardware. Disclosed functions may correspond to modules which may be software, hardware, firmware, or any combination thereof.

[0030] A software implementation may include computer-executable instructions and / or Artificial Intelligence (AI) / Machine Learning (ML) module(s) stored on a computer-readable medium, such as memory or other type of storage devices. One or more microprocessors or general-purpose computers with communication processing capability may be programmed with corresponding computer-executable instructions and perform the disclosed network function(s), AI / ML module(s), or algorithm(s). The AI / ML module(s) may be implemented with a supervised learning approach, a semi-supervised learning approach, an unsupervised learning approach (e.g., Transductive approach and Inductive approach), a federated learning approach, or a reinforcement learning (RL) approach, but the present disclosure is not limited thereto. The computer-executable instructions associated with the AI module(s) and / or the ML module(s) may include but are not limited to, data management instructions (e.g., collection instructions, validation instructions…etc.), model monitoring and management instructions (e.g., NW key performance indicators (KPIs) monitoring, model input / output monitoring, model selection / switching / update / upload / download, model (de)activation, model identification, functionality selection…etc.), and / or pre-process input instructions.

[0031] The microprocessors or general-purpose computers may include Application-Specific Integrated Circuits (ASICs), programmable logic arrays, Central Processing Units (CPUs), Tensor Processing Units (TPUs), Graphics Processing Units (GPUs), General-purpose computing on GPUs (GPGPU, or less often GPGP), and / or using one or more Digital Signal Processors (DSPs). Although some of the disclosed implementations are oriented to software installed and executing on computer hardware, alternative implementations implemented as firmware, as hardware, or as a combination of hardware and software are well within the scope of the present disclosure. The computer-readable medium may include, but is not limited to, Random Access Memory (RAM), Dynamic Random Access Memory (DRAM), High Bandwidth Memory (HBM), Magnetoresistive Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Resistive Random Access Memory (RRAM), Read-Only Memory (ROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), flash memory (or other memory technology), Compact Disc Read-Only Memory (CD-ROM) , Digital Versatile Disks (DVD) (or other optical disk storage), magnetic cassettes, magnetic tape, magnetic disk storage (or other magnetic storage devices), or any other equivalent medium capable of storing computer-readable instructions.

[0032] A radio communication network architecture such as a Long-Term Evolution (LTE) system, an LTE-Advanced (LTE-A) system, an LTE-Advanced Pro system, or a 5G NR Radio Access Network (RAN), 5G-Advanced (5G-A) system, or an open radio access network (O-RAN) may typically include at least one base station (BS), at least one UE, and one or more optional network elements that provide connection within a network. The BS and one or more optional network elements enable the UE to access a radio network. The UE may communicate with the network, such as a Core Network (CN), an Evolved Packet Core (EPC) network, an Evolved Universal Terrestrial RAN (E-UTRAN), a Next-Generation Core (NGC), a 5G Core (5GC), or an internet via a RAN established by one or more BSs and the network elements / functions.

[0033] A UE may include, but is not limited to, a mobile station, a mobile terminal or device, or a user communication radio terminal. The UE may be a portable radio equipment that includes, but is not limited to, a mobile phone, a tablet, a wearable device, a sensor, a vehicle, a virtual reality (VR) device, an augmented (AR) device, an Internet of Things (IoT) device, an unmanned aerial vehicle (UAV), or a Personal Digital Assistant (PDA) with wireless communication capability. The UE may be configured to receive and transmit signals over an air interface to one or more cells in a RAN. In some implementations, the UE may be an AI / ML-enabled device and / or an AI / ML capable device that is equipped with AI module(s) and / or ML module(s).

[0034] The BS may be configured to provide communication services according to at least a Radio Access Technology (RAT), such as Worldwide Interoperability for Microwave Access (WiMAX), Global System for Mobile communications (GSM) that is often referred to as 2G, GSM Enhanced Data rates for GSM Evolution (EDGE) RAN (GERAN), General Packet Radio Service (GPRS), Universal Mobile Telecommunication System (UMTS) that is often referred to as 3G based on basic Wideband-Code Division Multiple Access (W-CDMA), High-Speed Packet Access (HSPA), LTE, LTE-A, evolved / enhanced LTE (eLTE) that is LTE connected to 5GC, NR (often referred to as 5G), 5G-A, and / or LTE-A Pro. However, the scope of the present disclosure is not limited to these protocols.

[0035] The BS may include, but is not limited to, a node B (NB) in the UMTS, an evolved node B (eNB) in LTE or LTE-A, a radio network controller (RNC) in UMTS, a BS controller (BSC) in the GSM / GERAN, a next-generation eNB (ng-eNB) in an Evolved Universal Terrestrial Radio Access (E-UTRA) BS in connection with 5GC, a next-generation Node B (gNB) in the 5G-RAN or in the 5G Access Network (5G-AN), or any other apparatus capable of controlling radio communication and managing radio resources within a cell. The BS may serve one or more UEs via a radio interface. Although the gNB is used as an example in some implementations within the present disclosure, it should be noted that the disclosed implementations may also be applied to other types of base stations. In some implementations, the BS may be an AI / ML-enabled device and / or an AI / ML capable device that is equipped with AI module(s) and / or ML module(s).

[0036] The BS may be operable to provide radio coverage to a specific geographical area using multiple cells forming the RAN. The BS may support the operations of the cells. Each cell may be operable to provide services to at least one UE within its radio coverage.

[0037] Each cell (may often referred to as a serving cell) may provide services to one or more UEs within the cell’s radio coverage, such that each cell schedules the downlink (DL) (and optionally uplink (UL) resources) to at least one UE within its radio coverage for DL (and optionally UL packet transmissions from the UE). The BS may communicate with one or more UEs in the radio communication system via the cells.

[0038] A cell may allocate sidelink (SL) resources for supporting the Proximity Services (ProSe), LTE SL services, LTE / NR sidelink communication services, LTE / NR sidelink discovery services, and / or LTE / NR Vehicle-to-Everything (V2X) services. In addition, a cell may allocate DL and / or UL resources for supporting Multicast / Broadcast Service (MBS) services, Non-Terrestrial Networks (NTN) services, positioning services, power serving services and / or Network Energy Saving (NES) services.

[0039] In Multi-RAT Dual Connectivity (MR-DC) cases, the primary cell of a Master Cell Group (MCG) or a Secondary Cell Group (SCG) may be referred to as a Special Cell (SpCell). A Primary Cell (PCell) may include the SpCell of an MCG. A Primary SCG Cell (PSCell) may include the SpCell of an SCG. MCG may include a group of serving cells associated with the Master Node (MN), including the SpCell and optionally one or more Secondary Cells (SCells). An SCG may include a group of serving cells associated with the Secondary Node (SN), including the SpCell and optionally one or more SCells.

[0040] As discussed above, the frame structure for NR may support flexible configurations for accommodating various next generation (e.g., 5G) communication requirements, such as Enhanced Mobile Broadband (eMBB), Massive Machine Type Communication (mMTC), and Ultra-Reliable and Low-Latency Communication (URLLC), while fulfilling high reliability, high data rate, and low latency requirements. The Orthogonal Frequency-Division Multiplexing (OFDM) technology in the third generation partnership project (3GPP) may serve as a baseline for an NR waveform. The scalable OFDM numerology, such as adaptive sub-carrier spacing, channel bandwidth, and Cyclic Prefix (CP), may also be used.

[0041] Two coding schemes may be considered for NR, specifically, Low-Density Parity-Check (LDPC) code and Polar Code. The coding scheme adaption may be configured based on channel conditions and / or service applications.

[0042] At least the DL transmission data, a guard period, and UL transmission data should be included in a transmission time interval (TTI) of a single NR frame. The respective portions of the DL transmission data, the guard period, and the UL transmission data should also be configurable based on, for example, the network dynamics of NR. SL resources may also be provided in an NR frame to support ProSe services or V2X services.

[0043] Any two or more than two of the following paragraphs, (sub)-bullets, points, actions, behaviors, terms, or claims described in the present disclosure may be combined logically, reasonably, and properly to form a specific method.

[0044] Any sentence, paragraph, (sub)-bullet, point, action, behaviors, terms, or claims described in the present disclosure may be implemented independently and separately to form a specific method.

[0045] Dependency, e.g., “based on”, “more specifically”, “preferably”, “in one embodiment”, “in some implementations”, etc., in the present disclosure is just one possible example which would not restrict the specific method.

[0046] In some implementations, all the designs / embodiment / implementations introduced within this disclosure are not limited to be applied for dealing with the problems discussed within this disclosure. For example, the described embodiments may be applied to solve other problems that exist in the RAN of wireless communication systems. In some implementations, all of the numbers listed within the designs / embodiment / implementations introduced within this disclosure are just examples and for illustration, for example, of how the described methods are executed.

[0047] The terms, definitions, and abbreviations given in the present disclosure are either imported from existing documentation (e.g., European Telecommunications Standards Institute (ETSI), International Telecommunication Union (ITU), or elsewhere) or newly created by 3GPP experts whenever the need for precise vocabulary is identified.

[0048] AI / ML has been discussed to be applied to the NR system by the third generation partnership project (3GPP) in recent years. By collecting an adequate amount of data, the AI / ML model and / or functionality may achieve powerful capability of prediction that facilitates the air interface configuration. Aspects that may be empowered by leveraging AI / ML may include beam management, positioning accuracy enhancement, channel status information (CSI) prediction, CSI report compression, and mobility.

[0049] An AI / ML model may be used to generate the output of the desired prediction by taking some inputs, such as measurement results. A UE and / or a network (NW) provider may train multiple AI / ML models according to different situations / conditions at the UE- / NW-side. To protect the privacy of the model information, a functionality may be used to associate one or more models, which may correspond to one or more UE-side / NW-side situations / conditions. More specifically, the situation / condition where the training / monitoring / inference data is collected may be referred to as an additional condition (AC), and an additional condition may be classified as a UE-side AC or a NW-side AC. For example, a UE-side AC may be the UE’s mobility status and / or the UE’s power status; a NW-side AC may be the transmission power of the reference signals and / or the antenna configuration for the transmission. A functionality may be mapped to one or more models which apply to one or more NW-side and UE-side ACs. In some implementations, each AI / ML functionality may correspond to at least one AI / ML model. Each AI / ML model may correspond to at least one AC. Each AC may correspond to a condition under which data is collected for the corresponding AI / ML model.

[0050] The life cycle management (LCM) plays an important role in AI / ML operation in the NR system. However, it is still unclear what a functionality is and how the UE / NW uses the functionality to apply the AI / ML models for the specific use cases. As a result, in the present disclosure, the operation of functionality-based LCM is described, including the functionality state, how the UE / NW determines the functionality state, and how the functionality is actually used (e.g., how the UE / NW activates the functionalities).

[0051] Scenarios

[0052] In the present disclosure, use cases for which AI / ML LCM is applicable are considered, including, but not limited to, beam management, positioning, CSI prediction, CSI-compression, and mobility. A network may include one or more radio access network (RAN) nodes (e.g., evolved node B (eNB) and / or next-generation node B (gNB)), each of which may include one or more cells, each of which may be associated with one or more transmission and reception points (TRP) and / or transmission points (TP), a UE that may be capable of receiving positioning reference signal (PRS), synchronization signal block (SSB), and CSI-RS from one or more TRPs and / or TPs in the same or different cells, and a location management function (LMF) that may reside in the core network.

[0053] For positioning use cases, the signaling between the UE and the LMF may be based on the LTE Positioning Protocol (LPP), which may be facilitated via the RAN nodes, while the LPP signaling between the UE and the LMF may be transparent to the RAN nodes.

[0054] In the present disclosure, cases where the AI / ML models / functionalities reside at the UE side are considered. For example, the UE may store the model / functionalities and perform inference from the models / functionalities when the UE considers the model / functionalities to be activated.

[0055] Functionality States and Functionality Reporting

[0056] A supported functionality may refer to a functionality that the UE indicates by using capability information.

[0057] An applicable functionality may refer to a functionality that the UE is ready to apply for inference.

[0058] An activated functionality may refer to functionality that is already enabled for performing inference.

[0059] A configured functionality may refer to a functionality that is configured by the network and may be considered to be ready to apply for inference. It should be noted that in the present disclosure, a configured functionality may be an activated functionality, and vice versa.

[0060] An available functionality may refer to a functionality that the UE considers to be ready to apply for inference without the NW-side information (e.g., associated ID or NW-side AC). It should be noted that in the present disclosure, an available functionality may be an applicable functionality, and vice versa.

[0061] In some implementations, for the communication between the UE and the BS (e.g., a gNB), an associated ID may be used to represent one or more gNB-specific NW-side ACs / TRP-specific NW-side ACs. For example, one or more associated IDs may be associated with a functionality, and one or more functionalities may be associated with an associated ID. More specifically, the mapping between associated IDs and functionalities may be one-to-one mapping, many-to-one mapping, one-to-many mapping, and / or many-to-many mapping. In some implementations, if the UE considers that a functionality supports the NW-side ACs corresponding to an associated ID, the UE may associate the functionality with the associated ID.

[0062] In some implementations, the gNB may transmit a first signaling (e.g., a message) to the UE to request the UE’s capability related to AI / ML support.

[0063] More specifically, the first signaling may be an RRC signaling.

[0064] More specifically, in the first signaling, the gNB may include an Information Element (IE) indicating the NW-side ACs that may be supported by the gNB.

[0065] In some implementations, the IE may be a list of ENUMERATED values. Each of the ENUMERATED values may indicate a NW-side AC supported by the gNB.

[0066] In some implementations, the IE may be a list of integer values. Each of the integer values may indicate a NW-side AC supported by the gNB.

[0067] In some implementations, the IE may be one or more associated IDs.

[0068] In some implementations, the IE may be a list of Reference Signal (RS) configurations (e.g., PRS configuration, CSI-RS configuration). Each of the RS configurations may be associated with one or more associated IDs.

[0069] More specifically, in the first signaling, the gNB may include an IE indicating the NW-side AC(s) that is / are currently applied by the gNB.

[0070] In some implementations, the IE may be a list of ENUMERATED values. Each of the ENUMERATED values may indicate a NW-side AC currently applied by the gNB.

[0071] In some implementations, the IE may be a list of integer values. Each of the integer values may indicate a NW-side AC currently applied by the gNB.

[0072] In some implementations, the IE may be one or more associated IDs.

[0073] In some implementations, the IE may be a list of RS configurations (e.g., PRS configuration, CSI-RS configuration). Each of the RS configurations may be associated with one or more associated IDs. In addition, each of the one or more associated IDs may be associated with a NW-side AC currently applied by the gNB.

[0074] In some implementations, the UE may be explicitly provided with the mapping between the associated IDs and the NW-side ACs via RRC signaling. In some implementations, the mapping between the associated IDs and the NW-side ACs may be NW implementation.

[0075] In some implementations, the PCell / PSCell may provide respective associated IDs for corresponding SCells. In some implementations, the initial active Bandwidth Part (BWP) may provide respective associated IDs for other BWPs. In some implementations, the dormant BWP may have a separate associated ID from the non-dormant BWP. In some implementations, the source cell may provide respective associated IDs for other candidate cells via an RRC message / RRC signaling.

[0076] More specifically, in the first signaling, the gNB may include an IE indicating the request of the UE’s capability related to AI / ML. For example, the gNB may request whether the UE has the capability related to AI / ML by transmitting the IE to the UE.

[0077] In some implementations, the IE may be an ENUMERATED format to indicate whether UE’s capability related to AI / ML is requested. If the value in the ENUMERATED format is present, the gNB requests UE’s capability to AI / ML. If the value in the ENUMERATED format is absent, the gNB does not request UE’s capability to AI / ML.

[0078] In some implementations, the IE may be an ENUMERATED format to indicate whether UE-side additional condition(s) related to AI / ML is requested.

[0079] In some implementations, the IE may be one or more ENUMERATED values. Each of the ENUMERATED values may indicate whether an aspect of UE’s capability related to AI / ML is requested. In some implementations, an aspect may be associated with an AI / ML use case. For example, an ENUMERATED value may represent the request for whether the UE supports the AI / ML for beam management, and another ENUMERATED value may represent the request for whether the UE supports the AI / ML for CSI prediction, and another ENUMERATED value may represent the request for whether the UE supports the AI / ML for positioning. In some implementations, an aspect may be associated with one-side model or two-side model. For example, an ENUMERATED value may represent the request for whether the UE supports one-side model, and another ENUMERATED value may represent the request for whether the UE supports the two-side model, and another ENUMERATED value may represent the request for whether the UE supports both one-side model and two-side model.

[0080] In some implementations, the IE may be a bitmap. Each of the bits in the bitmap may indicate whether an aspect of UE’s capability related to AI / ML is requested. For example, if the aspect is associated with an AI / ML use case, the first bit (e.g., the leftmost bit) may represent the request for whether the UE supports the AI / ML for beam management, the second bit may represent the request for whether the UE supports the AI / ML for CSI prediction, the third bit may represent the request for whether the UE supports the AI / ML for positioning, but not limited thereto.

[0081] In some implementations, the bit value ‘0’ may represent that the UE does not support the AI / ML for the corresponding use case, and the bit value ‘1’ may represent that UE supports the AI / ML for the corresponding use case.

[0082] In some implementations, the bit value ‘1’ may represent that the UE does not support the AI / ML for the corresponding use case, and the bit value ‘0’ may represent that UE supports the AI / ML for the corresponding use case.

[0083] In some implementations, when the UE receives the first signaling from the gNB, the UE may perform one or more of the following actions:

[0084] (a) If the IE indicating whether UE’s capability related to AI / ML is requested is present, the UE may consider that the gNB requests for the UE’s capability related to AI / ML. More specifically, the UE may consider that gNB requests for the UE’s capability related to specific AI / ML aspect (e.g., beam management, positioning, CSI prediction, CSI compression, etc.) according to the indication in the IE.

[0085] (b) If the IE indicating the NW-side ACs that may be supported by the gNB is present in the first signaling, the UE may consider / expect that the gNB supports the NW-side ACs that are indicated by the IE.

[0086] More specifically, the UE may consider / expect the gNB supports the NW-side ACs corresponding to the ENUMERATED or integer values in the IE according to a predefined / preconfigured table.

[0087] More specifically, the table may be preconfigured via RRC signaling (e.g., via System Information Block (SIB), RRCReconfiguration, etc.). In some implementations, the table may include multiple tuples, each of which may further include an integer value and an ENUMERATED value indicating the NW-side ACs. For example, if a tuple includes an integer value 1 and an ENUMERATED value indicating a first NW-side AC and another tuple includes an integer value 2 and an ENUMERATED value indicating a second NW-side AC, the UE may comprehend that the first NW-side AC is associated with value 1 and the second NW-side AC is associated with value 2.

[0088] In some implementations, the UE may check the value of the IE in the preconfigured / predefined table and the IE in the first signaling. For example, if the IE includes integers 1 and 2, the UE may comprehend that the first and the second NW-side ACs are supported by the gNB.

[0089] Alternatively, the UE may consider / expect the gNB supports the associated IDs indicated in the first signaling.

[0090] (c) If the IE indicating the NW-side ACs that is currently applied by the gNB is present in the first signaling, the UE may consider / expect that the gNB supports and is currently applying the NW-side AC(s) that are indicated by the IE.

[0091] More specifically, the UE may consider / expect the gNB supports and is currently applying the NW-side ACs which correspond to the ENUMERTED or integer values in the IE according to a predefined / preconfigured table.

[0092] More specifically, the table may be preconfigured via RRC signaling (e.g., via SIB, RRCReconfiguration, etc.). In some implementations, the table may include multiple tuples, each of which may further include an integer value and an ENUMERATED value indicating the NW-side ACs. For example, if a tuple includes an integer value 1 and an ENUMERATED value indicating a first NW-side AC and another tuple includes an integer value 2 and an ENUMERATED value indicating a second NW-side AC, the UE may comprehend that the first NW-side AC is associated with value 1 and the second NW-side AC is associated with value 2.

[0093] In some implementations, the UE may check the value of the IE in the preconfigured / predefined table and the IE in the first signaling. For example, if the IE includes integers 1 and 2, the UE may comprehend that the first and the second NW-side ACs are supported by the gNB.

[0094] Alternatively, the UE may consider / expect the gNB supports and is applying the associated IDs indicated in the first signaling.

[0095] (d) The UE may include the supported functionalities in a second signaling, which may be transmitted to the gNB. In some implementations, the UE may include the supported functionalities (e.g., in the second signaling) according to the information included in the first signaling.

[0096] In some implementations, the UE may include an indication of all the supported functionalities in the second signaling. More specifically, if (only) the IE indicating that the UE’s capability related to AI / ML is included in the first signaling and the IE indicates that the UE’s capability related to AI / ML is requested, the UE may include all the supported functionalities in the second signaling. In some implementations, the indication of the supported functionalities may be one or more lists, each of which may include one or more integers.

[0097] In some implementations, a list may correspond to a specific aspect of the AI / ML use cases. More specifically, the number of lists may be equal to the number of the aspects requested in the first signaling.

[0098] In some implementations, an integer in a list may correspond to a functionality (e.g., a functionality ID) that the UE supports. In other words, the functionality IDs present in the list may indicate that the corresponding functionalities are supported functionalities.

[0099] In some implementations, the indication of the supported functionalities may be one or more bitmaps.

[0100] In some implementations, a bitmap may correspond to a specific aspect of the AI / ML use cases. More specifically, the number of bitmaps may be equal to the number of the aspects requested in the first signaling.

[0101] In some implementations, a bit in a bitmap may correspond to a functionality and may indicate whether a functionality associated with the aspect is supported. For example, whether the UE supports a first functionality associated with a first aspect may be indicated by the first bit (e.g., the leftmost bit) in the first bitmap; whether the UE supports a third functionality associated with a second aspect may be indicated by the third bit in the second bitmap.

[0102] In some implementations, the bit value ‘0’ may represent that UE does not support the corresponding functionality, and the bit value ‘1’ may represent that UE supports the corresponding functionality.

[0103] In some implementations, the bit value ‘1’ may represent that UE does not support the corresponding functionality, and the bit value ‘0’ may represent that UE supports the corresponding functionality.

[0104] The order of the bitmap corresponding to the functionalities may be determined by the presentation order of the functionalities in the message indicating the supported functionalities.

[0105] In some implementations, the UE may include the supported functionalities related to one or more specific aspects. More specifically, if (only) the IE indicating that the UE’s capability related to AI / ML is included in the first signaling and the IE indicates that the UE’s capability related to AI / ML in one or more aspects is requested, the UE may include all the supported functionalities related to the one or more aspects in the second signaling.

[0106] In some implementations, the UE may include the supported functionalities associated with one or more NW-side ACs in the second signaling.

[0107] More specifically, if the IE indicating the NW-side ACs that may be supported by the gNB is present in the first signaling and the IE indicates one or more NW-side ACs, the UE may include the supported functionalities associated with the one or more NW-side ACs in the second signaling.

[0108] Alternatively, if the IE indicating the NW-side ACs that are currently applied by the gNB is present in the first signaling and the IE indicates one or more NW-side ACs, the UE may include the supported functionalities associated with the one or more NW-side ACs in the second signaling.

[0109] Alternatively, if one or more associated ID is included in the first signaling, the UE may include the supported functionalities associated with the one or more associated IDs in the second signaling.

[0110] (e) The UE may include the mapping relation between the supported functionalities and the NW-side ACs in a second signaling, which may be transmitted to the gNB.

[0111] In some implementations, the mapping relation may be included in the IE indicating the supported functionalities.

[0112] For example, for each IE indicating a supported functionality, if a NW-side AC is mapped to / associated with the functionality, the UE may include the NW-side AC in the IE indicating the supported functionality.

[0113] For example, for each IE indicating a supported functionality, if an associated ID is supported by the supported functionality, the UE may include the associated ID in the IE indicating the supported functionality.

[0114] In some implementations, the mapping relation may be an IE separate from the IE indicating the supported functionalities. For example, the mapping relation IE may include multiple sequences, where a sequence may indicate a functionality and the NW-side ACs supported by the functionality. For each supported functionality, if the functionality is associated with / mapped to one or more NW-side ACs, the UE may include an IE indicating the functionality and an IE indicating the one or more NW-side ACs in the same sequence.

[0115] The UE may transmit the second signaling to the gNB. More specifically, the second signaling may be an RRC signaling.

[0116] In some implementations, upon receiving the second signaling from the UE, the gNB may perform one or more of the following actions:

[0117] (a) The gNB may consider the functionalities indicated in the second signaling to be supported functionalities.

[0118] (b) The gNB may include an IE to indicate one or more NW-side ACs that may be supported by the gNB in a third signaling, where the component in the IE may be the same as the counterpart of the first signaling. Alternatively, the gNB may use one or more associated ID to represent the one or more NW-side ACs.

[0119] (c) The gNB may include an IE to indicate one or more NW-side ACs that are currently applied by the gNB in a third signaling, where the component in the IE may be the same as the counterpart of the first signaling. Alternatively, the gNB may use one or more associated ID to represent the one or more NW-side ACs.

[0120] (d) The gNB may include one or more IEs indicating one or more configurations for data collection / inference / monitoring in the third signaling. In some implementations, an IE may be a reference signal set (e.g., a PRS resource set, an SSB resource set, or a CSI-RS resource set) on which the reference signals may be transmitted by the gNB and may be measured by the UE.

[0121] In some implementations, the IE may further include one or more IEs indicating one or more NW-side ACs to indicate that the configuration is associated with the one or more IEs indicating the one or more configurations.

[0122] In some implementations, the IE may further include one or more associated IDs to indicate that the configuration is associated with the associated IDs.

[0123] In some implementations, the IE may include one or more configurations for data collection / inference / monitoring. In addition, each configuration may include one or more associated IDs.

[0124] (e) The gNB may transmit the third signaling to the UE.

[0125] In some implementations, the third signaling may be an RRC signaling.

[0126] In some implementations, the third signaling may be an RRCReconfiguration signaling including an otherConfig IE.

[0127] In some implementations, upon receiving the third signaling from the gNB, the UE may perform one or more of the following actions:

[0128] (a) If an IE indicating one or more NW-side ACs that may be supported by the gNB is included in the third signaling, the UE may consider / expect that the gNB supports the one or more NW-side ACs. Therefore, the UE may perform the AI / ML inference based on the one or more NW-side ACs.

[0129] (b) If an IE indicating one or more NW-side ACs that are currently applied by the gNB is included in the third signaling, the UE may consider / expect that the gNB supports and is currently applying the one or more NW-side ACs. Therefore, the UE may perform the AI / ML inference based on the one or more NW-side ACs.

[0130] (c) If one or more IEs indicating one or more configurations for data collection / inference / monitoring are included in the third signaling, for each IE, the UE may consider / expect the configuration to be used for data collection / inference / monitoring for AI / ML.

[0131] More specifically, if the IE includes one or more IEs indicating one or more NW-side ACs, the UE may consider / expect that the configuration to be used for data collection / inference / monitoring is associated with the NW-side ACs indicated by the one or more IEs.

[0132] More specifically, if the IE includes one or more associated IDs, the UE may consider / expect that the configuration to be used for data collection / inference / monitoring is associated with the one or more associated IDs.

[0133] (d) The UE may include an IE indicating the applicable functionalities in the fourth signaling.

[0134] In some implementations, if an IE indicating one or more NW-side ACs that may be supported by the gNB is included in the third signaling, the UE may include the applicable functionalities associated with the one or more NW-side ACs.

[0135] In some implementations, if an IE indicating one or more NW-side ACs that is currently applied by the gNB is included in the third signaling, the UE may include the applicable functionalities associated with the one or more NW-side ACs.

[0136] In some implementations, if one or more IEs indicating one or more configurations for data collection / inference / monitoring are included in the third signaling, and if one or more of the IEs includes one or more associated IDs, the UE may include the applicable functionalities associated with the one or more associated IDs.

[0137] In some implementations, the indication of the applicable functionalities may be one or more lists, each of which may include one or more integers.

[0138] In some implementations, a list may correspond to an associated ID (or an NW-side AC). More specifically, the number of the lists may be equal to the number of the associated IDs (or the NW-side ACs) in the third signaling.

[0139] In some implementations, an integer in a list may correspond to a functionality (e.g., a functionality ID) that the UE considers to be applicable. In other words, the functionality IDs present in the list may indicate that the corresponding functionalities are applicable functionalities.

[0140] In some implementations, the indication of the applicable functionalities may be one or more bitmaps.

[0141] In some implementations, a bitmap may correspond to an association ID (or an NW-side AC). More specifically, the number of bitmaps may be equal to the number of the association IDs (or the NW-side ACs) in the third signaling.

[0142] In some implementations, a bit in a bitmap may correspond to a functionality and may indicate whether a functionality associated with the associated ID is applicable. For example, whether the UE considers a first functionality associated with a first associated ID to be applicable may be indicated by the first bit (e.g., the leftmost bit) in the first bitmap; whether the UE considers a third functionality associated with a second associated ID to be applicable may be indicated by the third bit in the second bitmap.

[0143] In some implementations, the bit value ‘0’ may represent that UE does not consider the corresponding functionality to be applicable, and the bit value ‘1’ may represent that UE considers the corresponding functionality to be applicable.

[0144] In some implementations, the bit value ‘1’ may represent that UE does not consider the corresponding functionality to be applicable, and the bit value ‘0’ may represent that UE considers the corresponding functionality to be applicable.

[0145] The order of the bitmap corresponding to the functionalities may be determined by the presentation order of the functionalities in the message indicating the supported functionalities.

[0146] (e) The UE may transmit the fourth signaling to the gNB.

[0147] In some implementations, the fourth signaling may be an RRC signaling. More specifically, the fourth signaling may be an UEAssistanceInformation, RRCReconfigurationComplete, or RRCResumeComplete message.

[0148] In some implementations, the fourth signaling may be a UE Assistance Information message.

[0149] In some implementations, in case that the mapping relation of functionalities and the associated ID changes (e.g., the UE performs a cross-set validation to add or remove some of the mapping relations), the UE may also transmit a signaling to inform the gNB of the change. In some implementations, the signaling may be a second signaling that includes the mapping relation. In some implementations, the signaling may be a fourth signaling that includes the mapping relation.

[0150] Autonomous Activation of Functionality

[0151] In some implementations, the gNB may include an additional IE to indicate that the UE is allowed to indicate one or more functionalities to be autonomously activated in the third signaling. In some implementations, the IE may take ENUMERATED format with value in {‘true’, ‘false’}, where ‘true’ may represent that the UE is allowed to indicate the functionalities to be autonomously activated and ‘false’ may represent that the UE is not allowed to indicate the functionalities to be autonomously activated. For example, the gNB may transmit, to the UE, the third signaling granting permission for the UE to report AI / ML functionalities for autonomous activation.

[0152] In some implementations, when the gNB includes the configuration to be used for data collection / inference / monitoring for a functionality in the third signaling, the UE may assume that the functionality may be autonomously activated.

[0153] In some implementations, when the UE transmits the second signaling, the UE may assume that the functionalities indicated in the second signaling are autonomously activated.

[0154] In some implementations, the UE may transmit the fourth signaling with an additional IE indicating the functionalities to be autonomously activated. For example, the UE may transmit, to the gNB, the fourth signaling indicating at least one AI / ML functionality to be autonomously activated by the UE.

[0155] More specifically, the UE may transmit the fourth signaling with the additional IE if the third signaling includes an IE indicating that the UE is allowed to indicate one or more functionalities to be autonomously activated.

[0156] In some implementations, the additional IE may include a list of IEs indicating the functionalities to be autonomously activated, where each IE in the list may indicate a functionality. For example, an IE may be a functionality ID.

[0157] In some implementations, the additional IE may include a list of IEs indicating the functionalities to be autonomously activated, where each IE in the list may indicate an associated ID. The mapping between the associated IDs and the functionalities may be UE implementations or may be explicitly configured / indicated to the UE via RRC signaling.

[0158] In some implementations, the UE may activate the functionalities whose IDs are indicated in the additional IE in the fourth signaling upon transmitting the fourth signaling to the gNB.

[0159] In some implementations, the UE may activate the functionalities whose corresponding associated IDs are indicated in the additional IE in the fourth signaling upon transmitting the fourth signaling to the gNB.

[0160] In some implementations, the gNB may consider / expect that the functionalities whose IDs are indicated in the additional IE in the fourth signaling to be activated functionalities upon receiving the fourth signaling from the UE.

[0161] In some implementations, the gNB may consider / expect that the functionalities whose corresponding associated IDs are indicated in the additional IE in the fourth signaling to be activated functionalities upon receiving the fourth signaling from the UE.

[0162] In some implementations, the UE may activate a functionality if the functionality is the only one functionality indicated in the fourth signaling upon transmitting the fourth signaling to the gNB.

[0163] In some implementations, the gNB may consider / expect a functionality to be activated if the functionality is the only one functionality indicated in the fourth signaling upon receiving the fourth signaling from the UE.

[0164] Explicit and Implicit Activation of Functionality

[0165] In some implementations, after the UE transmits the fourth signaling to the gNB, the UE may receive a fifth signaling from the gNB, where the fifth signaling may include one or more IEs indicating one or more configurations for data collection / inference / monitoring. In some implementations, the IE may be a configuration for a reference signal set (e.g., a PRS resource set, an SSB resource set, or a CSI-RS resource set) on which the reference signals may be transmitted by the gNB and may be measured by the UE. In some implementations, the fifth signaling may be an RRC signaling. More specifically, the fifth signaling may be an RRC reconfiguration message. In some implementations, upon receiving the fifth signaling, the UE may perform one or more of the following actions:

[0166] (a) The UE may store the configurations received in the fifth signaling.

[0167] (b) For each configuration, if the IE for the configuration includes one or more functionality IDs, the UE may activate the functionalities indicated by the one or more functionality IDs. In some implementations, the configuration may include an IE (e.g., activation status) associated with the functionality ID, where the IE may indicate whether the functionality is activated upon reception of the fifth signaling.

[0168] (c) The UE may consider / expect that the configuration is associated with the functionality(ies) that is / are indicated in the fourth signaling. In some implementations, the UE may activate the functionality(ies) that are indicated in the fourth signaling upon receiving the fifth signaling from the gNB.

[0169] More specifically, the functionalities that UE activates after receiving the fifth signaling may be all the functionalities indicated in the fourth signaling.

[0170] More specifically, the functionalities that UE activates after receiving the fifth signaling may be the functionalities that are indicated to be autonomously activated in the fourth signaling.

[0171] More specifically, the functionality that UE activates after receiving the fifth signaling may be the (only) one functionality indicated in the fourth signaling.

[0172] (d) For each configuration, if the IE for the configuration includes one or more associated IDs, the UE may consider / expect that the configuration is associated with the associated IDs.

[0173] (e) For each configuration, if the IE for the configuration includes one or more associated IDs, the UE may activate the functionalities that are associated with the associated IDs.

[0174] (f) For each configuration, if the IE for the configuration includes one or more associated IDs, the UE may select a functionality from the functionalities that are associated with the associated IDs and activate the selected functionality.

[0175] In some implementations, when the UE is configured with one or more configurations for data collection / inference / monitoring, the UE may receive a sixth signaling from the gNB for activation / deactivation of one or more functionalities.

[0176] In some implementations, the sixth signaling may be an RRC message, a MAC CE, or a DCI.

[0177] In some implementations, the sixth signaling may include one or more fields indicating one or more associated IDs, one or more fields indicating one or more functionality IDs, and / or a field indicating whether the signaling is for activation or deactivation.

[0178] In some implementations, the field indicating an associated ID may indicate an integer value corresponding to an associated ID.

[0179] In some implementations, the field indicating a functionality ID may indicate an integer value corresponding to a functionality ID.

[0180] In some implementations, the field indicating whether the signaling is for activation or deactivation may take a binary value, where one of the values (e.g., 0) stands for the deactivation and the other value (e.g., 1) stands for the activation.

[0181] In some implementations, the sixth signaling may include a bitmap indicating one or more associated IDs, one or more fields indicating one or more functionality IDs, and / or the bitmap indicating whether the corresponding functionality is for activation or deactivation.

[0182] In some implementations, the first bit in the bitmap may be used to indicate whether the functionality associated with the functionality ID#1 is activated or deactivated. In some implementations, the bit indicating ‘0’ may mean ‘deactivated’, and the bit indicating ‘1’ may mean ‘activated’. In some implementations, the bit indicating ‘1’ may mean ‘deactivated’, and the bit indicating ‘0’ may mean ‘activated’.

[0183] In some implementations, the first bit in the bitmap may be used to indicate whether the functionalit(ies) associated with the associated ID#1 is activated or deactivated. In some implementations, the bit indicating ‘0’ may mean ‘deactivated’, and the bit indicating ‘1’ may mean ‘activated’. In some implementations, the bit indicating ‘1’ may mean ‘deactivated’, and the bit indicating ‘0’ may mean ‘activated’. In some implementations, there may be some reserved bits in the bitmap reserved for future use. These reserved bits may not be associated with any functionality IDs or any associated IDs.

[0184] In some implementations, the number of bits in the bitmap may be greater than or equal to the number of functionality IDs (e.g., the number of supported functionalities, the number of the applicable functionalities, the number of functionalities associated with a specific AI / ML use case, etc.), or the number of associated IDs.

[0185] In some implementations, upon receiving the sixth signaling, the UE may activate and / or deactivate one or more functionalities.

[0186] More specifically, if the sixth signaling includes a field indicating that the sixth signaling is for activation, the UE may activate the one or more functionalities upon receiving the sixth signaling.

[0187] Alternatively, if the one or more functionalities are deactivated before the UE receives the sixth signaling, the UE may activate the one or more functionalities upon receiving the sixth signaling.

[0188] More specifically, if the sixth signaling includes one or more fields indicating one or more functionality IDs, the UE may activate the one or more functionalities corresponding to the one or more functionality IDs.

[0189] Alternatively, if the sixth signaling includes one or more fields indicating one or more associated IDs, the UE may activate the one or more functionalities associated with the one or more associated IDs. It should be noted that the network may guarantee that different AI / ML use cases (e.g., beam management, CSI prediction, positioning, etc.) are associated with different associated ID to avoid the confusion between use cases.

[0190] Alternatively, if the sixth signaling includes one or more fields indicating one or more associated IDs, for each indicated associated ID, the UE may select a functionality associated with that indicated associated ID and activate the selected functionality.

[0191] In some implementations, upon receiving the sixth signaling, the UE may deactivate one or more functionalities.

[0192] More specifically, if the sixth signaling includes a field indicating that the signaling is for deactivation, the UE may deactivate the one or more functionalities upon receiving the sixth signaling.

[0193] Alternatively, if the one or more functionalities are activated before the UE receives the sixth signaling, the UE may deactivate the one or more functionalities upon receiving the sixth signaling.

[0194] More specifically, if the sixth signaling includes one or more fields indicating one or more functionality IDs, the UE may deactivate the one or more functionalities corresponding to the one or more functionality IDs.

[0195] Alternatively, if the sixth signaling includes one or more fields indicating one or more associated IDs, the UE may deactivate the one or more functionalities associated with the one or more associated IDs.

[0196] More specifically, if the one or more functionalities are already activated but are not indicated by sixth signaling, the UE may deactivate the one or more functionalities upon receiving the sixth signaling.

[0197] In some implementations, when a UE activates a functionality, the UE may apply the stored configuration associated with the associated ID that is associated with the functionality for data collection / inference / monitoring. In some implementations, applying the stored configuration may include performing measurement on the reference signals on the resources configured in the stored configuration.

[0198] In some implementations, a UE may receive a signaling from the network (e.g., an RRC message), and the UE may release one or more stored configurations for data collection / inference / monitoring upon receiving the signaling. Upon releasing the one or more stored configurations, the UE may consider the activated functionalities associated with the one or more configurations to be deactivated. Upon releasing the one or more stored configurations, the UE may consider the functionalities associated with the one or more configurations to be non-applicable.

[0199] Associated ID Update

[0200] In some implementations, a UE may acquire the associated ID during a functionality identification procedure. In the functionality identification procedure, the UE may collect the data and label the collected dataset with the associated ID. In some implementations, the UE may transfer the collected dataset and the corresponding associated ID to a UE-side server, which may be responsible for data storage and / or model training / update. In some implementations, the UE-side server may transfer the dataset and / or the trained model with the associated ID to the UE, and the UE may know a functionality is applicable for some associated IDs even if the UE attaches to the cell at the first time.

[0201] In some implementations, the gNB may change the mapping (or relation, or association) between the associated IDs and the configurations for data collection / inference / monitoring. In some implementations, the gNB may inform the UE of the change of the mapping via a signaling.

[0202] More specifically, the signaling may be a DCI. In some implementations, the DCI may include a field indicating the change of the mapping. In some implementations, the DCI may be addressed by a group-based Radio Network Temporary Identifier (RNTI). In some implementations, the DCI may be addressed by an RNTI specific to AI / ML use cases. In some implementations, an AI / ML-specific DCI format may include a field indicating the change of the mapping. The AI / ML-specific DCI format may be addressed by C-RNTI, SI-RNTI, group common RNTI, and / or AI / ML-specific RNTI.

[0203] Alternatively, the signaling may be a MAC CE. In some implementations, the MAC CE may include a field indicating the change of the mapping. Alternatively, the MAC CE may include empty field and may take a specific logical channel ID (LCID) in the MAC header / subheader.

[0204] Alternatively, the signaling may be an RRC message. In some implementations, the RRC message may include an IE indicating the change of the mapping. Alternatively, the RRC message may include an IE indicating the version ID of the mapping and / or an IE indicating the updated mapping. In some implementations, the association ID may be broadcast by a new SIB and the change of association ID may follow the modification period procedure.

[0205] In some implementations, upon receiving signaling indicating the change of mapping from the gNB, the UE may consider that the mapping between the associated IDs and the configurations is changed.

[0206] More specifically, the UE may consider the mapping to be changed if the signaling is a DCI including a field indicating the change of the mapping.

[0207] Alternatively, the UE may consider the mapping to be changed if the signaling is a MAC CE including a field indicating the change of the mapping.

[0208] Alternatively, the UE may consider the mapping to be changed if the signaling is a MAC CE with a specific LCID in the MAC header / subheader.

[0209] Alternatively, the UE may consider the mapping to be changed if the signaling is an RRC message including an IE indicating the change of the mapping.

[0210] Alternatively, the UE may consider the mapping to be changed if the signaling is an RRC message including a version ID and the version ID is not equal to a stored version ID.

[0211] Alternatively, the UE may consider the mapping to be changed if the signaling is an RRC message including an IE indicating the updated mapping. In some implementations, the IE may be a list of sequences, and each sequence may include an IE indicating the new associated ID and a list of IEs indicating the old associated IDs that are mapped to the new associated ID. It should be noted that a sequence may include only an IE indicating the new associated ID. For example, the new associated ID may not be mapped to any of the old associated ID.

[0212] In some implementations, the UE may update the mapping when it considers that the mapping is changed, where the update may include one or more of the following actions:

[0213] (a) The UE may receive an RRC signaling including an IE indicating the updated mapping from the gNB.

[0214] More specifically, the resource of the RRC signaling may be indicated by the DCI including a field indicating the change of the mapping. More specifically, the UE may receive the RRC signaling including an IE indicating the updated mapping on the resource indicated by the DCI including a field indicating the change of the mapping.

[0215] Alternatively, the resource of the RRC signaling may be preconfigured to the UE (e.g., via system information block 1 (SIB1)). Alternatively, the UE may receive the RRC signaling including an IE indicating the updated mapping on the resource preconfigured to the UE (e.g., via SIB1).

[0216] Alternatively, a paging message may be used for indicating the change of the mapping and the UE may proceed with the SIB reception based on the pre-configured resource block.

[0217] (b) The UE may remove the stored mapping.

[0218] (c) The UE may store the mapping included in the signaling received from the gNB.

[0219] (d) The UE may re-initiate the functionality report procedure as described in the “Functionality State and Functionality Reporting” Section.

[0220] (e) For each new associated ID, if the new associated ID is associated with one or more old associated IDs, for each old associated ID, the UE may associate the new associated ID with the functionalities associated with the old associated ID. It should be noted that the old associated IDs are the associated IDs used by the gNB (or the network) before the update, and that the new associated IDs are the associated IDs used by the gNB (or the network) after the update.

[0221] In some implementations, after the UE updates the mapping, for each functionality, if the functionality is not associated with any new associated ID, the UE may deactivate the functionality or may consider the functionality not to be an applicable functionality. In some implementations, after the UE updates the mapping, the UE may initiate the functionality report to inform the gNB of the update mapping between the functionalities and the new associated IDs.

[0222] In some implementations, when a UE is operating AI / ML with an activated functionality, the UE may receive a fifth signaling including inference configurations. The UE may consider the fifth signaling to be the update of the inference configuration.

[0223] In some implementations, if the fifth signaling includes an inference configuration associated with the activated functionality (e.g., the inference configuration includes an associated ID associated with the activated functionality), the UE may keep the functionality activated and apply the inference configuration in the received fifth signaling.

[0224] In some implementations, if the fifth signaling includes an inference configuration associated with the activated functionality (e.g., the inference configuration includes an associated ID associated with the activated functionality), the UE may deactivate the functionality and store the inference configuration in the received fifth signaling. The UE may activate the functionality again upon receiving a sixth signaling indicating the activation of the functionality.

[0225] Example of AI / ML Operations

[0226] In some implementations, a UE may receive a first RRC signaling from the gNB for requesting the UE’s AI / ML capability, where the first signaling may include a list of IEs, each of which may indicate whether an aspect of UE’s AI / ML capability is requested.

[0227] In some implementations, upon receiving the first RRC signaling, the UE may transmit a second RRC signaling to the gNB, where the second RRC signaling may include an IE indicating the supported functionalities. In some implementations, the IE indicating the supported functionalities may be a list of sequences, each of which may include a functionality ID and a list of associated IDs that are associated with the functionality corresponding to the functionality ID. It should be noted that the association between a functionality and an associated ID may be determined by the UE either based on the UE’s implementation or based on the functionality identification prior to the functionality reporting procedure.

[0228] In some implementations, upon receiving the second RRC signaling, the gNB may transmit a third RRC signaling, where the third RRC signaling may include one or more associated IDs to indicate the NW-side ACs that may be applied by the gNB.

[0229] In some implementations, upon receiving the third RRC signaling, the UE may transmit a fourth RRC signaling to the gNB, where the fourth RRC signaling may include an IE indicating the functionalities that are applicable and associated with the associated IDs presented in the third RRC signaling.

[0230] In some implementations, after the UE transmits the fourth RRC signaling, the UE may receive a fifth signaling including a list of configurations for inference, where a configuration may include an associated ID. In some implementations, the UE may store the configurations in a UE storage. In some implementations, the UE may associate the associated ID with the configuration.

[0231] In some implementations, the UE may receive a MAC CE including a field indicating an associated ID, and the UE may activate the functionalities associated with the associated ID corresponding to the value of the field and apply the configuration associated with the associated ID for inference.

[0232] Alternatively, the UE may receive a MAC CE including a field indicating a functionality ID, and the UE may activate the functionality corresponding to the value of the field and apply the configuration(s) associated with the associated ID(s) that is / are associated with the functionality.

[0233] FIG. 1 is a flowchart illustrating a method / process 100 performed by a UE for performing an AI / ML operation, according to an example implementation of the present disclosure. In the action 102, the process 100 may start by receiving, from a BS, a first message granting permission for the UE to report AI / ML functionalities for autonomous activation. In the action 104, the process 100 may transmit, to the BS, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE. The process 100 may then end. The first message and the second message may correspond to the third signaling and the fourth signaling, respectively, described in the present disclosure. The first message may provide a general permission, whereas the second message may specify the particular functionalities to be activated.

[0234] In some implementations, the UE may activate the at least one AI / ML functionality indicated in the second message upon transmitting the second message.

[0235] In some implementations, the UE may receive, from the BS, a third message including a configuration associated with the at least one AI / ML functionality indicated in the second message. For example, the configuration included in the third message may include a configuration for data collection / inference / monitoring. The UE may activate the at least one AI / ML functionality indicated in the second message upon receiving the third message. The third message may correspond to the fifth signaling described in the present disclosure.

[0236] In some implementations, the UE may receive, from the BS, a third message including a configuration associated with the at least one AI / ML functionality indicated in the second message. The configuration may indicate an activation status associated with a functionality ID. The UE may activate an AI / ML functionality corresponding to the functionality ID based on the activation status upon receiving the third message. For example, the activation status may indicate whether the AI / ML functionality corresponding to the functionality ID is activated. The third message may correspond to the fifth signaling described in the present disclosure.

[0237] In some implementations, the first message, the second message, and the third message are transmitted via RRC signaling.

[0238] In some implementations, the UE may receive, from the BS, a fourth message including an ID field and an activation / deactivation field. The UE may activate or deactivate an AI / ML functionality corresponding to the ID field based on the activation / deactivation field. In some implementations, the ID field may indicate an associated ID or a functionality ID. In some implementations, the fourth message may include a MAC CE. The fourth message may correspond to the sixth signaling described in the present disclosure.

[0239] In some implementations, each AI / ML functionality may correspond to at least one AI / ML model, each AI / ML model may correspond to at least one AC, and each AC may correspond to a condition under which data is collected for the corresponding AI / ML model

[0240] FIG. 2 is a flowchart illustrating a method / process 200 performed by a BS for performing an AI / ML operation, according to an example implementation of the present disclosure. In the action 202, the process 200 may start by transmitting, to a UE, a first message granting permission for the UE to report AI / ML functionalities for autonomous activation. In the action 204, the process 200 may receive, from the UE, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE. The process 200 may then end. The method illustrated in FIG. 2 is similar to that in FIG. 1, except that it is described from the perspective of the BS (instead of the UE).

[0241] FIG. 3 is a block diagram illustrating a node 300 for wireless communication in accordance with various aspects of the present disclosure. As illustrated in FIG. 3, a node 300 may include a transceiver 320, a processor 328, a memory 334, one or more presentation components 338, and at least one antenna 336. The node 300 may also include a radio frequency (RF) spectrum band module, a BS communications module, a network communications module, and a system communications management module, Input / Output (I / O) ports, I / O components, and a power supply (not illustrated in FIG. 3).

[0242] Each of the components may directly or indirectly communicate with each other over one or more buses 340. The node 300 may be a UE or a BS that performs various functions disclosed with reference to FIGS. 1 through 2.

[0243] The transceiver 320 has a transmitter 322 (e.g., transmitting / transmission circuitry) and a receiver 324 (e.g., receiving / reception circuitry) and may be configured to transmit and / or receive time and / or frequency resource partitioning information. The transceiver 320 may be configured to transmit in different types of subframes and slots including, but not limited to, usable, non-usable, and flexibly usable subframes and slot formats. The transceiver 320 may be configured to receive data and control channels.

[0244] The node 300 may include a variety of computer-readable media. Computer-readable media may be any available media that may be accessed by the node 300 and include volatile (and / or non-volatile) media and removable (and / or non-removable) media.

[0245] The computer-readable media may include computer-storage media and communication media. Computer-storage media may include both volatile (and / or non-volatile media), and removable (and / or non-removable) media implemented in any method or technology for storage of information such as computer-readable instructions, data structures, program modules, AI / ML module(s), or data.

[0246] Computer-storage media may include RAM, ROM, EPROM, EEPROM, flash memory (or other memory technology), CD-ROM, Digital Versatile Disks (DVD) (or other optical disk storage), magnetic cassettes, magnetic tape, magnetic disk storage (or other magnetic storage devices), etc. Computer-storage media may not include a propagated data signal. Communication media may typically embody computer-readable instructions (e.g., computer-readable instructions related to AI module(s) and / or the ML module(s)), data structures, program modules, or other data in a modulated data signal, such as a carrier wave, or other transport mechanisms and include any information delivery media.

[0247] The term “modulated data signal” may mean a signal that has one or more of its characteristics set or changed in such a manner as to encode information in the signal. Communication media may include wired media, such as a wired network or direct-wired connection, and wireless media, such as acoustic, RF, infrared, and other wireless media. Combinations of any of the above listed components should also be included within the scope of computer-readable media.

[0248] The memory 334 may include computer-storage media in the form of volatile and / or non-volatile memory. The memory 334 may be removable, non-removable, or a combination thereof. Example memory may include solid-state memory, hard drives, optical-disc drives, etc. As illustrated in FIG. 3, the memory 334 may store a computer-readable and / or computer-executable instructions 332 (e.g., software codes) that are configured to, when executed, cause the processor 328 to perform various functions disclosed herein, for example, with reference to FIGS. 1 through 2. Alternatively, the instructions 332 may not be directly executable by the processor 328 but may be configured to cause the node 300 (e.g., when compiled and executed) to perform various functions disclosed herein.

[0249] The processor 328 (e.g., having processing circuitry) may include an intelligent hardware device, e.g., a Central Processing Unit (CPU), a microcontroller, an ASIC, etc. The processor 328 may include memory. The processor 328 may process the data 330 and the instructions 332 received from the memory 334, and information transmitted and received via the transceiver 320, the baseband communications module, and / or the network communications module. The processor 328 may also process information to send to the transceiver 320 for transmission via the antenna 336 to the network communications module for transmission to a CN.

[0250] One or more presentation components 338 may present data indications to a person or another device. Examples of presentation components 338 may include a display device, a speaker, a printing component, a vibrating component, etc.

[0251] In view of the present disclosure, it is obvious that various techniques may be used for implementing the disclosed concepts without departing from the scope of those concepts. Moreover, while the concepts have been disclosed with specific reference to certain implementations, a person of ordinary skill in the art may recognize that changes may be made in form and detail without departing from the scope of those concepts. As such, the disclosed implementations are to be considered in all respects as illustrative and not restrictive. It should also be understood that the present disclosure is not limited to the particular implementations disclosed and many rearrangements, modifications, and substitutions are possible without departing from the scope of the present disclosure.

Claims

1. A User Equipment (UE) for performing an Artificial Intelligence (AI) / Machine Learning (ML) operation, the UE comprising: 　　　　at least one processor; and 　　　　at least one non-transitory computer-readable medium coupled to the at least one processor and storing one or more computer-executable instructions that, when executed by the at least one processor, cause the UE to: 　　　　　　　　receive, from a Base Station (BS), a first message granting permission for the UE to report AI / ML functionalities for autonomous activation; and 　　　　　　　　transmit, to the BS, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.

2. The UE of claim 1, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: 　　　　activate the at least one AI / ML functionality indicated in the second message upon transmitting the second message.

3. The UE of claim 1, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: 　　　　receive, from the BS, a third message comprising a configuration associated with the at least one AI / ML functionality indicated in the second message; and 　　　　activate the at least one AI / ML functionality indicated in the second message upon receiving the third message.

4. The UE of claim 1, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: 　　　　receive, from the BS, a third message comprising a configuration associated with the at least one AI / ML functionality indicated in the second message, the configuration indicating an activation status associated with a functionality identifier (ID); and 　　　　activate an AI / ML functionality corresponding to the functionality ID based on the activation status upon receiving the third message.

5. The UE of claim 4, wherein: 　　　　the first message, the second message, and the third message are transmitted via Radio Resource Control (RRC) signaling.

6. The UE of claim 1, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the UE to: 　　　　receive, from the BS, a fourth message comprising an identifier (ID) field and an activation / deactivation field; and 　　　　activate or deactivate an AI / ML functionality corresponding to the ID field based on the activation / deactivation field.

7. The UE of claim 6, wherein: 　　　　the fourth message comprises a Medium Access Control (MAC) Control Element (CE).

8. The UE of claim 1, wherein: 　　　　each AI / ML functionality corresponds to at least one AI / ML model, 　　　　each AI / ML model corresponds to at least one additional condition (AC), and 　　　　each AC corresponds to a condition under which data is collected for the corresponding AI / ML model.

9. A Base Station (BS) for performing an Artificial Intelligence (AI) / Machine Learning (ML) operation, the BS comprising: 　　　　at least one processor; and 　　　　at least one non-transitory computer-readable medium coupled to the at least one processor and storing one or more computer-executable instructions that, when executed by the at least one processor, cause the BS to: 　　　　　　　　transmit, to a User Equipment (UE), a first message granting permission for the UE to report AI / ML functionalities for autonomous activation; and 　　　　　　　　receive, from the UE, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.

10. The BS of claim 9, wherein: 　　　　the UE activates the at least one AI / ML functionality indicated in the second message upon transmitting the second message.

11. The BS of claim 9, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to: 　　　　transmit, to the UE, a third message comprising a configuration associated with the at least one AI / ML functionality indicated in the second message, wherein: 　　　　the UE activates the at least one AI / ML functionality indicated in the second message upon receiving the third message.

12. The BS of claim 9, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to: 　　　　transmit, to the UE, a third message comprising a configuration associated with the at least one AI / ML functionality indicated in the second message, the configuration indicating an activation status associated with a functionality identifier (ID), wherein: 　　　　the UE activates an AI / ML functionality corresponding to the functionality ID based on the activation status upon receiving the third message.

13. The BS of claim 12, wherein: 　　　　the first message, the second message, and the third message are transmitted via Radio Resource Control (RRC) signaling.

14. The BS of claim 9, wherein the one or more computer-executable instructions, when executed by the at least one processor, further cause the BS to: 　　　　transmit, to the UE, a fourth message comprising an identifier (ID) field and an activation / deactivation field, wherein: 　　　　the UE activates or deactivates an AI / ML functionality corresponding to the ID field based on the activation / deactivation field.

15. A method performed by a User Equipment (UE) for performing an Artificial Intelligence (AI) / Machine Learning (ML) operation, the method comprising: 　　　　receiving, from a Base Station (BS), a first message granting permission for the UE to report AI / ML functionalities for autonomous activation; and 　　　　transmitting, to the BS, a second message indicating at least one AI / ML functionality to be autonomously activated by the UE.

Images