Lifecycle management for ai / ml for air interface enhancements

Abstract

Various aspects of the present disclosure relate to a UE, a processor for wireless communication, a network entity, a method, and a computer readable medium for supporting lifecycle management (LCM) for AI / ML for air interface enhancements. The UE sends an indication to the network entity about whether an AI / ML function supports model identification. The UE performs a function-based LCM procedure for the AI / ML function. If the AI / ML function supports model identification, the function-based LCM procedure includes a model ID-based LCM procedure for at least one AI / ML model associated with the AI / ML function. In this way, one unified LCM framework is proposed to support all potential use cases with AI / ML.

Description

Technical Field

[0001] This disclosure relates to wireless communications, and more specifically to user equipment (UE), network entities, processors, methods, and computer-readable media for supporting lifecycle management (LCM) for artificial intelligence / machine learning (AI / ML) enhancements for air interface. Background Technology

[0002] A wireless communication system may include one or more network communication devices (such as base stations), which may also be referred to as eNodeB (eNB), next-generation NodeB (gNB), or other suitable terms. Each network communication device (such as a base station) may support wireless communication with one or more user communication devices, which may also be referred to as user equipment (UE), or other suitable terms. The wireless communication system may support wireless communication with one or more user communication devices by utilizing the resources of the wireless communication system (e.g., time resources (e.g., symbols, time slots, subframes, frames, etc.) or frequency resources (e.g., subcarriers, carriers)). Furthermore, the wireless communication system may support wireless communication across a variety of wireless access technologies, including third-generation (3G) wireless access technology, fourth-generation (4G) wireless access technology, fifth-generation (5G) wireless access technology, and other suitable wireless access technologies other than 5G (e.g., sixth-generation (6G)).

[0003] Artificial intelligence (AI) / machine learning (ML) is used to train neural networks using massive amounts of data to learn and perform certain tasks, and it has been successfully applied in the fields of computer vision (CV) and natural language processing (NLP). As a subset of ML, deep learning (DL) utilizes multi-layered neural networks (NNs) as "AI models" to learn how to solve problems and optimize performance from large amounts of data. Because many academic papers and field test results have shown promising benefits, AI / ML-based methods can achieve better performance than traditional methods if the AI / ML models are well trained.

[0004] Within 3GPP, discussions are underway regarding the introduction of AI / ML into the air interface for select use cases in NR Releases 18 and 19, including enhanced CSI feedback, beam management, and improved location accuracy, with the adoption of agreed-upon evaluation methods and results. A general framework to support AI / ML in the air interface is expected to be specified in the normative work (i.e., NR Rel-19). Therefore, in the specification study and subsequent work items, it is necessary to investigate and clarify the lifecycle management (LCM) characteristics of AI / ML models based on investigation and evaluation. Summary of the Invention

[0005] This disclosure relates to user equipment (UE), network entities, processors, methods, and computer-readable media for supporting LCM for AI / ML enhancements in the air interface. Embodiments of this disclosure propose a unified LCM framework to support all potential use cases with AI / ML.

[0006] In a first aspect, a UE is provided. The UE includes a processor; and a transceiver coupled to the processor, wherein the processor is configured to: send an indication to a network entity via the transceiver regarding whether an artificial intelligence / machine learning (AI / ML) function supports model identification; and perform a function-based lifecycle management (LCM) process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0007] In a second aspect, a network entity is provided, comprising: a processor; and a transceiver coupled to the processor, wherein the processor is configured to: receive, via the transceiver, an indication from a user equipment (UE) regarding whether an artificial intelligence / machine learning (AI / ML) function supports model identification; and perform a function-based lifecycle management (LCM) process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0008] In a third aspect, a processor for wireless communication is provided. The processor includes at least one memory; and a controller coupled to the at least one memory and configured to: send an indication to a network entity regarding whether an artificial intelligence / machine learning (AI / ML) function supports model identification; and perform a function-based lifecycle management (LCM) process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0009] In a fourth aspect, a method is provided performed by a user equipment (UE), the method comprising: sending an indication to a network entity regarding whether an artificial intelligence / machine learning (AI / ML) function supports model identification; and performing a function-based lifecycle management (LCM) process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0010] In a fifth aspect, a method is provided to be performed by a network entity, the method comprising: receiving from a user equipment (UE) an indication as to whether an artificial intelligence / machine learning (AI / ML) function supports model identification; and performing a function-based lifecycle management (LCM) process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0011] In a sixth aspect, a computer-readable medium having instructions stored thereon is provided, which, when executed by a processor of a device, cause the device to perform the method according to the fourth or fifth aspect of this disclosure.

[0012] In some implementations of the methods, UEs, and network entities described herein, the indication may include: a bit indicating the presence of at least one AI / ML model associated with the AI / ML function, which will be identified.

[0013] In some implementations of the methods, UEs, and network entities described herein, the indication may include an identifier of at least one AI / ML model associated with the AI / ML function, wherein the identifier is a model ID or a temporary ID.

[0014] In some implementations of the methods, UEs, and network entities described in this paper, this instruction is included in the information regarding AI / ML functionality.

[0015] In some implementations of the methods, UEs, and network entities described in this paper, the indication may be sent to the network entity along with information about the AI / ML function during the function identification process.

[0016] In some implementations of the methods, UEs, and network entities described herein, information regarding AI / ML functionality may include at least one of the following: configuration of features or feature groups that enable AI / ML; and application conditions.

[0017] In some implementations of the methods, UEs, and network entities described in this paper, information about AI / ML capabilities may be included in dedicated UE capabilities or UE assistance information.

[0018] In some implementations of the methods, UEs, and network entities described herein, the function-based LCM procedure may also include a model identification procedure. The UE may determine the model ID of at least one AI / ML model based on this indication; and during the model identification procedure, the UE may exchange information about at least one AI / ML model with the network entity using the model ID. The network entity may determine the model ID of at least one AI / ML model based on this indication; and during the model identification procedure, the network entity may exchange information about at least one AI / ML model with the UE using the model ID.

[0019] In some implementations of the methods, UEs, and network entities described herein, information about at least one AI / ML model may include at least one of the following: description; configuration; or application conditions.

[0020] In some implementations of the methods, UEs, and network entities described in this paper, the model identification process can be enabled after the AI / ML function has been activated.

[0021] In some implementations of the methods, UEs, and network entities described in this paper, the model ID-based LCM process can be enabled after the model identification process.

[0022] In some implementations of the methods, UEs, and network entities described herein, each of the function-based LCM process and the model ID-based LCM process may include at least one of the following operations: activation; switching; selection; updating; or deactivation.

[0023] In some implementations of the methods, UEs, and network entities described herein, the UE may apply the same operation to at least one AI / ML model of the AI / ML function in response to one operation of the AI / ML function.

[0024] In some implementations of the methods, UEs, and network entities described in this paper, the AI / ML functionality is located on the UE side, and at least one AI / ML model is a UE-side model or a UE-part model of a dual-side model. Attached Figure Description

[0025] Figure 1 Examples of wireless communication systems in which some embodiments of the present disclosure may be implemented are illustrated.

[0026] Figures 2A-2C Examples of manageable units in some LCM schemes are illustrated.

[0027] Figures 3A-3CExamples of three types of model identification processes are illustrated.

[0028] Figures 4A-4C The diagram shows Figures 3A-3C The example shown is a signaling signaling process in the model identification process.

[0029] Figure 5 The illustration shows an example of a processing flow for a unified framework for LCM according to some example embodiments of the present disclosure.

[0030] Figure 6 Another example of the processing flow for a unified framework for LCM according to some example embodiments of the present disclosure is illustrated.

[0031] Figure 7 An example of a function identification process according to some exemplary embodiments of the present disclosure is illustrated.

[0032] Figure 8 The illustration shows an example of an indication of whether or not an AI / ML model with respect to AI / ML functionality will be identified and managed, according to some example embodiments of this disclosure.

[0033] Figure 9 An example of a function activation process according to some exemplary embodiments of the present disclosure is illustrated.

[0034] Figure 10 An example of an LCM process according to some exemplary embodiments of the present disclosure is illustrated, which includes operations on AI / ML functions and models.

[0035] Figure 11 Examples of devices suitable for implementing some embodiments of the present disclosure are illustrated.

[0036] Figure 12 Examples of processors suitable for implementing some embodiments of the present disclosure are illustrated.

[0037] Figure 13 The diagram illustrates a flowchart of a method performed by a user equipment according to various aspects of this disclosure.

[0038] Figure 14 The diagram illustrates a flowchart of a method performed by a network entity according to various aspects of this disclosure.

[0039] In all the accompanying drawings, the same or similar reference numerals denote the same or similar elements. Detailed Implementation

[0040] The principles of this disclosure will now be described with reference to some embodiments. It should be understood that these embodiments are described for illustrative purposes only and to assist those skilled in the art in understanding and implementing this disclosure, and do not impose any limitation on the scope of this disclosure. The disclosure described herein can be implemented in various ways other than those described below. In the following description and claims, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0041] References to "an embodiment," "an example embodiment," "an embodiment," "some embodiments," etc., in this disclosure indicate that the described embodiments(s) may include a particular feature, structure, or characteristic, but not every embodiment must include that particular feature, structure, or characteristic. Furthermore, these phrases do not necessarily refer to the same(s) embodiments(s). Moreover, when a particular feature, structure, or characteristic is described in connection with an embodiment, it should be understood that implementing such a feature, structure, or characteristic in conjunction with other embodiments (whether explicitly described or not) is within the knowledge of those skilled in the art.

[0042] It should be understood that although the terms “first” and “second” may be used herein to describe various elements, these elements should not be limited by these terms. These terms are used only to distinguish one element from another. For example, a first element may also be referred to as a second element without departing from the scope of the embodiments, and similarly, a second element may be referred to as a first element. As used herein, the term “and / or” includes any and all combinations of one or more of the listed terms. In some examples, values, processes, or apparatus are referred to as “best,” “lowest,” “highest,” “minimum,” “maximum,” etc. It should be understood that such descriptions are intended to indicate that a choice can be made among many functional alternatives used, and that such a choice need not be better, smaller, higher, or more desirable than other choices.

[0043] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the embodiments. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “the” used herein also include the plural forms. Furthermore, it should be understood that the terms “comprising,” “including,” “having,” “containing,” and / or “comprise”, when used herein, specify the presence of the stated features, elements, components, etc., but do not exclude the presence or addition of one or more other features, elements, components, and / or combinations thereof. For example, the term “comprising” and its variations should be understood as open terms meaning “including, but not limited to.” The term “based on” should be understood as “at least partially based on.” The terms “one embodiment” and “embodiment” should be understood as “at least one embodiment.” The term “another embodiment” should be understood as “at least one other embodiment.” The use of expressions such as “A and / or B” can mean “A only” or “B only” or “both A and B.” Other explicit and implicit definitions may be included below.

[0044] As mentioned above, it is necessary to study and clarify the characteristics of AI / ML model lifecycle management (LCM) based on investigation and evaluation. To date, two LCM schemes have been discussed for UE-side / UE-part models: function-based LCM and model ID-based LCM. The aim is to support AI / ML-related operations with different manageable units, functions, or the model itself.

[0045] Function-based LCM procedures may include a function identification process, which refers to the process / method of identifying AI / ML functions to achieve consensus between the network (NW) and the UE, wherein information about the AI / ML functions can be shared during function identification. Model ID-based LCM procedures may include a model identification process, which refers to the process / method of identifying AI / ML models to achieve consensus between the NW and the UE, wherein information about the AI / ML models can be shared during model identification.

[0046] To avoid redundant operations (e.g., activation, deactivation, toggling, rollback, etc.) and unnecessary specifications, a unified LCM framework is expected to be defined to support all potential use cases with AI / ML. However, several issues and problems exist when considering a unified design for LCM.

[0047] The first issue is that the processes discussed so far are independent for function-based LCM and model ID-based LCM, and they have different requirements during the identification process. The information shared and aligned during the identification process varies depending on the management purpose, such as general functional application conditions and model details.

[0048] The second issue is that the function-based LCM for the UE portion of the two-sided model has not been adequately discussed, even though it was previously discussed under the model ID-based LCM. The operation of the UE portion of the model discussed under the model ID-based LCM also needs to be designed and supported in the function-based LCM.

[0049] The third issue is that, although detailed interaction information during the identification process needs to be defined for each use case, the basic signaling and procedures used for enabling identification and LCM should be generic. Detailed operations in both LCM schemes (such as monitoring and rollback) are always discussed in each sub-use case, and some common operations (such as activation and deactivation) require further investigation.

[0050] This disclosure proposes a unified LCM framework to address the aforementioned issues through some specified air interface signaling, which can potentially be applied to all use cases. In some embodiments, the UE can send an indication to a network entity regarding whether the AI / ML function supports model identification. The network entity can receive this indication. The UE and the network entity can then perform a function-based LCM procedure for the AI / ML function on each other. If the indicated AI / ML function supports model identification, the function-based LCM procedure can include a model ID-based LCM procedure.

[0051] Various aspects of this disclosure are described in the context of wireless communication systems. Figure 1 An example of a wireless communication system 100 in which some embodiments of the present disclosure may be implemented is illustrated. The wireless communication system 100 may include one or more network entities 102 (also referred to as network devices (NEs)), one or more UEs 104, a core network 106, and a packet data network 108. The wireless communication system 100 may support various wireless access technologies. In some implementations, the wireless communication system 100 may be a 4G network, such as an LTE network or an Advanced LTE (LTE-A) network. In some other implementations, the wireless communication system 100 may be a 5G network, such as an NR network. In other implementations, the wireless communication system 100 may be a combination of 4G and 5G networks, or other suitable wireless access technologies, including IEEE 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), and IEEE 802.20. The wireless communication system 100 may support wireless access technologies other than 5G. Furthermore, the wireless communication system 100 may support technologies such as Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), or Code Division Multiple Access (CDMA).

[0052] One or more network entities 102 may be distributed throughout a geographic area to form a wireless communication system 100. One or more of the network entities 102 described herein may be, include, or may be referred to as network nodes, base stations, network elements, radio access networks (RANs), base transceiver stations, access points, NodeBs, eNodeBs (eNBs), next-generation NodeBs (gNBs), or other suitable terms. Network entities 102 and UEs 104 may communicate via a communication link 110, which may be a wireless or wired connection. For example, network entities 102 and UEs 104 may perform wireless communication (e.g., receive signaling, transmit signaling) via a Uu interface. In a 3GPP non-terrestrial network (NTN), a satellite-based network entity 102 may communicate directly with UEs 104 using an NR / LTE Uu interface. The satellite may be a transparent satellite or a regenerated satellite. For an NTN with transparent satellites, a base station on Earth may communicate with the UE via the satellite. For an NTN with regenerated satellites, a base station may be mounted on a satellite and communicate directly with the UE.

[0053] Network entity 102 can provide a geographic coverage area 112, and network entity 102 can support services (e.g., voice, video, packet data, messaging, broadcasting, etc.) for one or more UEs 104 within the geographic coverage area 112. For example, network entity 102 and UE 104 can support wireless communication of signals associated with services (e.g., voice, video, packet data, messaging, broadcasting, etc.) according to one or more radio access technologies. In some implementations, network entity 102 can be mobile, for example, a satellite associated with a non-terrestrial network. In some implementations, different geographic coverage areas 112 associated with the same or different radio access technologies can overlap, but different geographic coverage areas 112 can be associated with different network entities 102. The information and signals described herein can be represented using any of a variety of different technologies and techniques. For example, data, instructions, commands, information, signals, bits, symbols, and chips that may be referenced throughout the description can be represented by voltage, current, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.

[0054] One or more UEs 104 may be distributed throughout the geographic area of ​​the wireless communication system 100. UE 104 may include or be referred to as a mobile device, wireless device, remote device, remote unit, handheld device, subscriber device, or some other suitable term. In some implementations, UE 104 may be referred to as a unit, station, terminal, or client, etc. Alternatively or additionally, UE 104 may be referred to as an Internet of Things (IoT) device, an Internet of Everything (IoE) device, or a Machine-Type Communication (MTC) device, etc. In some implementations, UE 104 may be stationary within the wireless communication system 100. In some other implementations, UE 104 may be mobile within the wireless communication system 100.

[0055] One or more UEs 104 can be devices of different forms or with different capabilities. Figure 1 The diagram illustrates some examples of UE 104. UE 104 is capable of communicating with various types of devices, such as network entity 102, other UEs 104, or network devices (e.g., core network 106, packet data network 108, relay equipment, integrated access and backhaul (IAB) node, or another network device). Figure 1 As shown. Alternatively, UE 104 may support communication with other network entities 102 or UE 104 that may act as relays in wireless communication system 100.

[0056] UE 104 can also support direct wireless communication with other UE 104s via communication link 114. For example, UE 104 can support direct wireless communication with another UE 104 via a device-to-device (D2D) communication link. In some implementations, such as vehicle-to-vehicle (V2V) deployments, vehicle-to-everything (V2X) deployments, or cellular V2X deployments, communication link 114 may be referred to as a sidelink. For example, UE 104 can support direct wireless communication with another UE 104 via a PC5 interface.

[0057] Network entity 102 may support communication with core network 106 or with another network entity 102, or both. For example, network entity 102 may interface with core network 106 via one or more backhaul links 116 (e.g., via S1, N2, N3, or another network interface). Network entities 102 may communicate with each other via backhaul links 116 (e.g., via X2, Xn, or another network interface). In some implementations, network entities 102 may communicate directly with each other (e.g., between network entities 102). In some other implementations, network entities 102 may communicate with each other or indirectly (e.g., via core network 106). In some implementations, one or more network entities 102 may include sub-components, such as access network entities, which may be examples of access node controllers (ANCs). An ANC may communicate with one or more UEs 104 via one or more other access network transport entities (which may be referred to as radio headends, smart radio headends, or transmit-receive points (TRPs)).

[0058] In some implementations, network entity 102 can be configured with a decomposed architecture that can utilize protocol stacks physically or logically distributed across two or more network entities 102, such as an Integrated Access Backhaul (IAB) network, Open RAN (O-RAN) (e.g., a network configuration sponsored by the O-RAN Alliance), or Virtualized RAN (vRAN) (e.g., Cloud RAN (C-RAN)). For example, network entity 102 may include one or more of the following: Central Unit (CU), Distributed Unit (DU), Radio Unit (RU), RAN Intelligent Controller (RIC) (e.g., Near Real-Time RIC, Non-Real-Time RIC), Service Management and Orchestration (SMO) system, or any combination thereof.

[0059] An RU can also be referred to as a radio headend, intelligent radio headend, remote radio headend (RRH), remote radio unit (RRU), or transmit-receive point (TRP). In a decomposed RAN architecture, one or more components of network entity 102 may be co-located, or one or more components of network entity 102 may be located in distributed locations (e.g., separate physical locations). In some implementations, one or more network entities 102 in a decomposed RAN architecture may be implemented as virtual units (e.g., virtual CU (VCU), virtual DU (VDU), virtual RU (VRU)).

[0060] The functional decomposition between CU, DU, and RU can be flexible and can support different functions based on the functions performed at the CU, DU, or RU (e.g., network layer functions, protocol layer functions, baseband functions, radio frequency functions, and any combination thereof). For example, a protocol stack functional decomposition can be used between the CU and DU, allowing the CU to support one or more layers of the protocol stack and the DU to support one or more different layers of the protocol stack. In some implementations, the CU can host upper-layer protocol layer (e.g., Layer 3 (L3), Layer 2 (L2)) functions and signaling (e.g., Radio Resource Control (RRC), Serving Data Adaptation Protocol (SDAP), Packet Data Convergence Protocol (PDCP)). The CU can connect to one or more DUs or RUs, and one or more DUs or RUs can host lower-layer protocol layer functions and signaling, such as Layer 1 (L1) (e.g., Physical (PHY) layer) or L2 (e.g., Radio Link Control (RLC) layer, Media Access Control (MAC) layer), and each can be at least partially controlled by the CU.

[0061] Alternatively, a functional split of the protocol stack can be employed between the DU and RU, allowing the DU to support one or more layers of the protocol stack and the RU to support one or more different layers of the protocol stack. The DU can support one or more different cells (e.g., via one or more RUs). In some implementations, the functional split between the CU and DU, or between the DU and RU, can be within a protocol layer (e.g., some functions of the protocol layer can be performed by one of the CU, DU, or RU, while other functions of the protocol layer are performed by different items in the CU, DU, or RU).

[0062] The CU can be further functionally divided into CU control plane (CU-CP) and CU user plane (CU-UP) functions. The CU can be connected to one or more DUs via midhaul communication links (e.g., F1, F1-c, F1-u), and the DUs can be connected to one or more RUs via fronthaul communication links (e.g., open fronthaul (FH) interfaces). In some implementations, the midhaul or fronthaul communication links can be implemented based on interfaces (e.g., channels) between layers of a protocol stack supported by the corresponding network entity 102 communicating via such communication links.

[0063] Core network 106 can support user authentication, access authorization, tracking, connectivity, and other access, routing, or mobility functions. Core network 106 can be an evolved packet core (EPC) or a 5G core (5GC), which may include control plane entities that manage access and mobility (e.g., Mobility Management Entity (MME), Access and Mobility Management Functions (AMF)) and user plane entities that route packets or interconnect with external networks (e.g., Serving Gateway (S-GW), Packet Data Network (PDN) Gateway (P-GW), or User Plane Functions (UPF)). In some implementations, control plane entities may manage non-access stratum (NAS) functions, such as mobility, authentication, and bearer management (e.g., data bearers, signaling bearers, etc.) for one or more UEs 104 served by one or more network entities 102 associated with core network 106.

[0064] Core network 106 can communicate with packet data network 108 via one or more backhaul links 116 (e.g., via S1, N2, N3, or another network interface). Packet data network 108 may include application server 118. In some implementations, one or more UEs 104 may communicate with application server 118. UE 104 may establish a session (e.g., Protocol Data Unit (PDU) session, etc.) with core network 106 via network entity 102. Core network 106 can use the established session (e.g., an established PDU session) to route services (e.g., control information, data, etc.) between UE 104 and application server 118. A PDU session may be an example of a logical connection between UE 104 and core network 106 (e.g., one or more network functions of core network 106).

[0065] In the wireless communication system 100, network entity 102 and UE 104 can use the resources of the wireless communication system 100 (e.g., time resources (e.g., symbols, time slots, subframes, frames, etc.) or frequency resources (e.g., subcarriers, carriers)) to perform various operations (e.g., wireless communication). In some implementations, network entity 102 and UE 104 can support different resource structures. For example, network entity 102 and UE 104 can support different frame structures. In some implementations, such as in 4G, network entity 102 and UE 104 can support a single frame structure. In some other implementations, such as in 5G and other suitable wireless access technologies, network entity 102 and UE 104 can support various frame structures (i.e., multiple frame structures). Network entity 102 and UE 104 can support various frame structures based on one or more sets of parameters.

[0066] One or more parameter sets may be supported in the wireless communication system 100, and the parameter sets may include subcarrier spacing and cyclic prefixes. A first parameter set (e.g., μ=0) may be associated with a first subcarrier spacing (e.g., 15 kHz) and a regular cyclic prefix. In some implementations, the first parameter set (e.g., μ=0) associated with the first subcarrier spacing (e.g., 15 kHz) may utilize one time slot per subframe. A second parameter set (e.g., μ=1) may be associated with a second subcarrier spacing (e.g., 30 kHz) and a regular cyclic prefix. A third parameter set (e.g., μ=2) may be associated with a third subcarrier spacing (e.g., 60 kHz) and a regular cyclic prefix or an extended cyclic prefix. A fourth parameter set (e.g., μ=3) may be associated with a fourth subcarrier spacing (e.g., 120 kHz) and a regular cyclic prefix. A fifth parameter set (e.g., μ=4) may be associated with a fifth subcarrier spacing (e.g., 240 kHz) and a regular cyclic prefix.

[0067] The time intervals of resources (e.g., communication resources) can be organized according to frames (also called radio frames). Each frame can have a duration, for example, 10 milliseconds (ms). In some implementations, each frame can include multiple subframes. For example, each frame can include 10 subframes, and each subframe can have a duration, for example, 1 ms. In some implementations, each frame can have the same duration. In some implementations, each subframe of a frame can have the same duration.

[0068] Alternatively or concurrently, the time intervals of resources (e.g., communication resources) can be organized according to time slots. For example, a subframe may include a certain number (e.g., quantity) of time slots. The number of time slots in each subframe may also depend on one or more parameter sets supported in the wireless communication system 100. For example, a first parameter set, a second parameter set, a third parameter set, a fourth parameter set, and a fifth parameter set (i.e., μ=0, μ=1, μ=2, μ=3, μ=4) associated with corresponding subcarrier intervals of 15 kHz, 30 kHz, 60 kHz, 120 kHz, and 240 kHz may utilize one time slot per subframe, two time slots per subframe, four time slots per subframe, eight time slots per subframe, and 16 time slots per subframe, respectively. Each time slot may include a certain number (e.g., quantity) of symbols (e.g., OFDM symbols). In some implementations, the number (e.g., quantity) of time slots in a subframe may depend on the parameter set. For a regular cyclic prefix, a time slot may include 14 symbols. For an extended cyclic prefix (e.g., for a 60 kHz subcarrier spacing), a time slot can include 12 symbols. The relationship between the number of symbols per time slot, the number of time slots per subframe, and the number of time slots per frame for both regular and extended cyclic prefixes can depend on the parameter set. It should be understood that references to the first parameter set (e.g., μ=0) associated with the first subcarrier spacing (e.g., 15 kHz) can be used interchangeably between subframes and time slots.

[0069] In the wireless communication system 100, the electromagnetic (EM) spectrum can be divided into various categories, frequency bands, frequency channels, etc., based on frequency or wavelength. For example, the wireless communication system 100 can support one or more operating frequency bands, such as frequency range names FR1 (410MHz-7.125GHz), FR2 (24.25GHz-52.6GHz), FR3 (7.125GHz-24.25GHz), FR4 (52.6GHz-114.25GHz), FR4a or FR4-1 (52.6GHz-71GHz), and FR5 (114.25GHz-500GHz). In some implementations, network entity 102 and UE 104 can perform wireless communication on one or more operating frequency bands. In some implementations, FR1 can be used by network entity 102 and UE 104, along with other equipment or devices, for cellular communication services (e.g., control information, data). In some implementations, FR2 can be used by network entity 102 and UE 104, along with other equipment or devices, for short-range, high data rate capabilities.

[0070] FR1 can be associated with one or more parameter sets (e.g., at least three parameter sets). For example, FR1 can be associated with: a first parameter set (e.g., μ=0) comprising a subcarrier spacing of 15 kHz; a second parameter set (e.g., μ=1) comprising a subcarrier spacing of 30 kHz; and a third parameter set (e.g., μ=2) comprising a subcarrier spacing of 60 kHz. FR2 can be associated with one or more parameter sets (e.g., at least two parameter sets). For example, FR2 can be associated with: a third parameter set (e.g., μ=2) comprising a subcarrier spacing of 60 kHz; and a fourth parameter set (e.g., μ=3) comprising a subcarrier spacing of 120 kHz.

[0071] Typically, function-based LCM and model ID-based LCM on the UE side / UE part of the model serve the same purpose: supporting AI / ML-related operations. However, they have different manageable units. The manageable unit of function-based LCM is a 'function,' which is defined as the set of configuration parameters for features / FGs that enable AI / ML. In contrast, the manageable unit of model ID-based LCM is the AI / ML model and the relevant configuration parameters of the applied AI / ML model. Therefore, different approaches can be used to apply LCM: function-only LCM, model ID-only LCM, and a combination of both.

[0072] Figure 2A The illustration shows an example of a manageable unit for function-based LCM. In this approach, a manageable unit is a 'function', which refers to an AI / ML enabled feature / feature group (FG) enabled by (multiple) configurations, and these configurations are supported based on conditions indicated by UE capabilities. Figure 2A As shown, each function, such as function A, has a set of configuration parameters and a set of application conditions for supporting AI / ML-based methods. During the function identification process, the UE can report such information to the NW for management purposes.

[0073] With this LCM scheme, the NW (New Wireless Terminal) cannot see the deployed AI / ML models used for inference at the UE, because the NW does not need to know the deployed and currently used models. The NW can provide auxiliary information on applicable functions based on the configuration provided during the identification process, and the UE can select the model itself. In this sense, the model's proprietary information can be well protected, and this approach is beneficial for one-sided models, especially those from locally trained models.

[0074] Figure 2B The illustration shows an example of a manageable unit for an LCM based on a model ID. In this case, the manageable unit is the 'model,' whose description and application conditions are reported during model identification. Figure 2BAs shown, in each model, such as Model 1, the deployed AI / ML model has a corresponding description, configuration, and applicable conditions. During the model identification process, the UE can report such information to the NW for management purposes.

[0075] Through this LCM scheme, the NW can see the deployed AI / ML model (e.g., at least a descriptive representation) used for inference at the UE, and, if needed, the NW can provide auxiliary information and directly manipulate the model. The NW can select which models to activate, monitor, and deactivate. In this sense, the model's information is well exposed. This approach is beneficial for both two-sided and one-sided models undergoing model transformation from the NW.

[0076] Figure 2C The illustration shows examples of managed units in a combined function-based and model ID-based LCM, where models are grouped by function. In this case, a managed unit can be a 'model group' identified by a 'function', whose description, configuration, and application conditions are represented hierarchically, with some in function identifiers and others in individual model identifiers. For example... Figure 2C As shown, in each model, such as Model 1, the deployed AI / ML model has a corresponding description, configuration, and applicable conditions. During the model identification process, the UE can report such information to the NW for management purposes.

[0077] Through this LCM scheme, the NW can see the deployed AI / ML models used for inference at the UE, and the NW can provide auxiliary information and directly manipulate the models, such as... Figure 2B As shown in the diagram. The difference lies in the amount of information collected and the amount of signaling used to operate the model. In this sense, NW can select a group of models within a function for monitoring. This approach is advantageous for two-sided models and multiple models.

[0078] like Figures 3A-3C As shown, the general model identification process for the UE side and / or UE portion of the dual-side model can be divided into Type A and Type B1 / B2. For Type A, as... Figure 3A As shown, the model is identified without air interface signaling; for types B1 and B2, as... Figure 3B and Figure 3C As shown, the model is identified through the air interface, and its initiator is either NW or UE.

[0079] The following discusses the potential specification impacts and information to be shared via the air interface, including both one-sided (UE-side) and two-sided models. For the one-sided (UE-side) model, the primary objective is to share relevant information with the NW during the identification process.

[0080] Figure 4AThe diagram illustrates an example of signaling during the model identification process for Type A. For Type A, since both models are identified without over-the-air signaling, the model ID can be directly used to follow the model ID-based LCM. For example, as... Figure 4A As shown, the applicable model is reported using the model ID, and the model can be activated directly based on the model ID.

[0081] Figure 4B The diagram illustrates a signaling example during the Type B1 model identification process. For Type B1, this process includes a model ID request initiated by the UE and an allocation from the NW, as well as... Figure 4B The corresponding model description is shown. This type of basic process may include a UE requesting a model ID on a local model, which can be achieved by sending a request with some temporary IDs and some model descriptions. The NW then confirms the identification / registration, assigns the (multiple) model IDs to the requesting model, and requests additional, potentially other information. After these steps, the model identifier used to assign the (multiple) model IDs and the relevant information about the model (i.e., the model description) have been aligned for future model ID-based LCMs.

[0082] Figure 4C The diagram illustrates a signaling example during the Type B2 model identification process. For Type B2, this process may include a model request initiated by the UE and a model transfer from the NW, as well as... Figure 4C The diagram shows (multiple) model IDs and model descriptions. This type of basic process may include a UE requesting several AI / ML models for some applications, along with related requirements and capabilities. The NW can confirm the request and, if necessary, transmit the model with (multiple) model IDs along with the model description. Following these steps, the model identifier used to assign the (multiple) model IDs and related information about the model (i.e., the model description) have been aligned for future model ID-based LCM.

[0083] This disclosure presents a signaling set for supporting a unified LCM framework to enable AI / ML in the air interface. Within this framework, 'function identifier' and 'function-based LCM' are selected as baselines, with explicit indications regarding whether 'model identifier' and 'model ID-based LCM' are supported during function identifier selection. Signals to support this hierarchical process are further proposed.

[0084] Figure 5 An example of a processing flow 500 for a unified framework for LCM according to some exemplary embodiments of this disclosure is illustrated. Processing flow 500 may involve a UE 501 and a network entity (e.g., a base station, such as a gNB) 502. Processing flow 500 can be applied to reference... Figure 1The wireless communication system 100, for example, UE 501 can be any of UE 104, and network entity 502 can be any of network entity 102. It should be understood that the processing flow 500 can be applied to other communication scenarios.

[0085] At 510, UE 501 may send an indication 515 to network entity 502 regarding whether the AI / ML function supports model identification. Therefore, at 520, network entity 502 may receive the indication 515 from UE 501. The AI / ML function may refer to an AI / ML-enabled function / function group (FG) enabled by (multiple) configurations, where (multiple) configurations are supported based on conditions indicated by the UE capability. Accordingly, the function-based LCM may operate based on at least one configuration of the AI / ML-enabled feature / FG or a specific configuration of the AI / ML-enabled feature / FG.

[0086] AI / ML functions can be associated with multiple AI / ML models. In some embodiments, indication 515 may include a bit indicating the presence of at least one AI / ML model associated with the AI / ML function, which will be identified. Alternatively, indication 515 may include multiple identifiers of the multiple AI / ML models associated with the AI / ML function. The multiple identifiers may be multiple model IDs aligned to network entity 502, or multiple temporary IDs that can be used to deduce the multiple model IDs during model identification for type B1.

[0087] In some embodiments, indication 515 may be included in information about AI / ML functionality and sent to network entity 502 during the functionality identification process. For example, the UE may send information about AI / ML functionality and the aforementioned indication in dedicated UE capabilities or UE assistance information. Information about AI / ML functionality may include configurations of features or feature groups that enable AI / ML. Additionally or alternatively, information about AI / ML functionality may include application conditions.

[0088] At 530, UE 501 and network entity 502 can perform a function-based LCM procedure for the AI / ML function. According to embodiments of this disclosure, the LCM for the AI / ML function can select a function identifier, and a basic function-based LCM (e.g., operation on the function) is selected as the baseline. If the AI / ML function supports a model identifier, and a corresponding indication is explicitly sent to network entity 502, then UE 501 and network entity 502 can perform a model ID-based LCM procedure for at least one AI / ML model associated with the AI / ML function in the function-based LCM procedure 530. Reference will be made below. Figures 6 to 10 Describe the details.

[0089] Figure 6 The illustration shows another example of a processing flow 600 for a unified framework for LCM according to some exemplary embodiments of the present disclosure. Processing flow 600 can be as follows: Figure 5 The example implementation of processing flow 500 is shown. Processing flow 600 is based on a framework of function identifier 610 and function-based LCM 620, in which model identifier 623 and model ID-based LCM 624 are embedded.

[0090] As shown in the figure, processing flow 600 includes a function identification process 610. As defined in 3GPP, this is a process / method for identifying AI / ML functions to achieve consensus between the NW and UE, and relevant information about the AI / ML function can be shared during function identification. During function identification process 610, the configuration of features or feature groups enabling AI / ML, as well as the application conditions associated with the AI / ML function, can be sent between UE 501 and network entity 502. In some embodiments, during function identification process 610, indications regarding whether model identification and model ID-based LCM are supported can be included in the information about the AI / ML function.

[0091] Following the function identification process 610, the processing flow 600 may further include a function-based LCM 620. A basic LCM with a manageable unit for the function is used, as discussed in 3GPP. If support for model identification and model ID-based LCM is indicated, model identification and subsequent LCM for the AI / ML model associated with that function can be triggered.

[0092] Function-based LCM may include a function activation procedure 621. Function activation procedure 621 may be initiated by UE 501 or network entity 502. Function activation procedure 621 is initiated if predefined and aligned conditions(s) are met during function identification. For the NW-side model, function activation may not be specified. For the UE-side model, function activation requires indication or configuration by network entity 502, or by UE 501, but the UE may need to report to network entity 502. For the dual-side model, function activation requires indication or configuration by network entity 502, or by UE 501.

[0093] Processing flow 600 may also include function-based operations 622, such as function switching, selection, and updating, using a basic LCM with functional manageable units, as discussed in 3GPP.

[0094] Processing flow 600 may further include AI / ML model identification process 623 and LCM based on AI / ML model ID 624. For activated functions, if indication is required, the models in the function need to be further identified, at least where more than one model can be deployed for a function; that is, at least one AI / ML model in the function is identified to reach consensus between network entity 502 and UE 501. For example, for a UE-side model or UE portion of a dual-side model, UE 501 may also report the number of models and their corresponding descriptions in the function. The description may include at least the required model inputs and potential model outputs, which may be reported by model or by function. Additional conditions associated with the model (e.g., preferred reference signal (RS) configuration) may be reported by UE 501, for example, via uplink control information (UCI) or media access control (MAC) control element (MAC-CE). Based on the reported additional conditions, network entity 502 can configure appropriate configurations for UE 501 for inference. An example is the number of time-domain RS transmissions required for time-domain CSI or beamforming.

[0095] Regarding the LCM process 624 based on AI / ML model ID, a basic LCM with manageable units of models is used, such as model activation, model selection, model switching, model update, and model deactivation.

[0096] Processing flow 600 may also include an AI / ML function deactivation process 625. If the detected performance has degraded or is about to degrade, and / or applicable conditions are not further met, the function will be deactivated or disabled; if a model is activated in the function, the function will also deactivate all activated models. Function deactivation can be performed by the UE 501, but at least if ground truth is first available on the UE side, function deactivation needs to be reported to network entity 502. The detailed procedures with new designs for each process are described below.

[0097] Figure 7 An example of a function identification process 700 according to some exemplary embodiments of the present disclosure is illustrated. The function identification process 700 may be... Figure 6 The example shown is the function identification process 610. During the function identification process 700, some relevant information about AI / ML functions can be shared and aligned, mainly including the conditions used to activate AI / ML functions.

[0098] The function identification process 700 may include two signaling interactions. In optional step 0, network entity 502 queries the AI / ML function on the UE side. To identify the function on the UE side or a portion of the UE model, network entity 502 may send a query message containing information about the function via signaling, such as Radio Resource Control (RRC) signaling, including extended UE capability messages, etc. UECapabilityEnquiry Or new AI-related news, UEAICapabilityEnquiry Or for AI-related information requests UEAIInformationRequest .

[0099] At step 1, UE 501 can report AI / ML functions to network entity 502. These functions will be presented as AI-related feature groups (e.g., UE Assistance Information (UAI)). UECapabilityInformation RRC messages) or other signaling (e.g., dedicated RRC parameters or dedicated MAC CE format) to enable UE capabilities (e.g., UEAssistanceInformation The report is sent to network entity 502 in the form of an RRC message. In some cases, such as UAI, step 0 is not required and may therefore be optional in this disclosure.

[0100] The information to be reported during the function identification process 700 may include model identifiers and LCMs based on model IDs for whether the function is supported, along with other information to be shared during function identification (e.g., application conditions).

[0101] Figure 8 The illustration shows examples of indications regarding whether or not AI / ML models will be identified and managed, according to some exemplary embodiments of this disclosure. For example... Figure 8 As shown, there are N models in the AI / ML function, of which K models need to be identified and managed (or assisted) by the network, and the remaining models do not require such identification and management. The AI / ML models identified within the relevant function will be used for the following model ID-based LCM to share more information between network entity 502 and UE 501, as described below.

[0102] There are two methods for indicating information. In some embodiments, UE 501 may use a single bit to indicate the presence of a model to be identified. As an indication of further model identification, information for each model (including the model ID) will be provided further during the model identification process.

[0103] Alternatively, UE 501 can directly indicate the model to be identified. The model in the AI / ML function can be explicitly indicated along with an identifier (ID). If type A is considered, the model ID is aligned on both sides, and the identifier(s) can be the model ID. Otherwise (i.e., for types B1 and B2), a temporary ID can be used to locally identify the model, which will be assigned a model ID during the subsequent model identification process.

[0104] Figure 9 An example of a function activation process 900 according to some exemplary embodiments of the present disclosure is illustrated. The function activation process 900 may be... Figure 6 The example shown is a function identification process 621. To manage functions, it is necessary to activate them before any operation is performed. Function activation process 900 may include the following interaction between network entity 502 and UE 501.

[0105] At optional step 0, UE 501 can initiate activation of the function. If the UE's monitoring / evaluation meets the conditions for activating the function, UE 501 can request network entity 502 to assist or confirm function activation via dedicated signaling (e.g., RRC or MAC CE signaling).

[0106] In step 1, network entity 502 can activate the function. If the monitoring or evaluation of network entity 502 meets the conditions for activating the function, network entity 502 can request UE 501 to activate the function via dedicated signaling (e.g., RRC or MAC CE signaling).

[0107] At step 2, UE 501 can confirm activation. Upon receiving the activation signaling, the corresponding function can be activated if other relevant conditions are met, such as battery or compute load. This step can be implicitly included in other signaling after receiving activation from network entity 502. The function is then activated for LCM operations, which network entity 502 uses to manage functions and AI / ML models in UE 501.

[0108] After the functionality is activated, the model identification process (e.g., process 623) can be enabled. The model identification process may include references Figures 4A-4C The basic types A, B1, or B2 are identified. Furthermore, in accordance with the proposals in this disclosure, the model identification process may include additional features for each type.

[0109] In some embodiments, UE 501 may determine the model ID of the AI / ML model(s) ...

[0110] For Type A model procedures, information provided during function identification (e.g., the model ID(s) indicated from UE 501 to network entity 502) can be used for model identification. In this type, the model ID(s) are explicitly indicated during function identification. Then, during the model identification process, more granular information about the model, such as application conditions or model description information, can be exchanged using the model IDs.

[0111] For type B1, information provided during function identification, such as a one-bit indicating the presence of a model or multiple temporary IDs of the models(s) to be identified, can be used for model identification. In this type, models within the function need to be assigned temporary IDs, which can be determined during identification by the network entity based on the number of models. More granular information about the models, such as application conditions or model description information, can then be exchanged using the assigned model IDs. In the absence of an identifier in the indication, the UE and the network entity can perform model identification for all models within the function.

[0112] For type B2, there is no impact. Since this type is used for model transfer from network entities to the UE, the proposals in this disclosure regarding the UE-side / partial models have no impact on the model identification of this type.

[0113] In this manner, during the model identification process, the applicable conditions and model IDs for each model(s) in the function are aligned between network entity 502 and UE 501. Following model identification, the model ID-based LCM process is enabled. Model ID-based LCM-related operations (e.g., model switching, model selection, model update, and model deactivation) can be performed as discussed in 3GPP. In some embodiments, the unified LCM framework includes additional operations on AI / ML functions and related models, including function switching, updates, and deactivation.

[0114] Figure 10 An example of an LCM process 1000 according to some exemplary embodiments of the present disclosure is illustrated, the LCM process including operations on AI / ML functions and models. Figure 10 In this context, "Function Switching" means switching the applicable function if a change in application conditions is detected. Function switching can include two operations: deactivating the currently active function and activating another function. "Function Update" means updating relevant information about the function, such as applicable conditions and configuration. "Function Deactivation" means deactivating the active function. If multiple models are identified within a function, then when one function is deactivated, all models within that function can also be deactivated.

[0115] Regarding these operations, according to some embodiments of this disclosure, the following related adjustments are made. In some embodiments, in response to one operation of the AI / ML function, the same operation is applied to at least one AI / ML model of the AI / ML function. For example, after a function is switched or updated, it may be necessary to identify the models in the function as described above. Furthermore, if the function is switched back without being updated, i.e., the function is activated without being updated, it may not be necessary to identify the models in the function.

[0116] As another example, after a function is deactivated, the models within that function can be deactivated, regardless of whether they were previously activated. For instance, a function can be deactivated when a performance degradation is detected when using it, or when the application conditions for that function are not met, and the models within it are also deactivated. In short, any operation on a function also represents an operation on the models within that function.

[0117] According to reference Figure 2 to Figure 10 Some embodiments discussed present a unified LCM framework that uses a function-based LCM as a baseline and, upon activation, instructs models within the function for possible identification and model ID-based LCM. The relevant signaling supporting this unified LCM framework can be implemented in the specifications of all potential use cases via dedicated RRC signaling and / or MAC CE.

[0118] Figure 11 Examples of devices suitable for implementing some embodiments of this disclosure are illustrated. Device 1100 may be an example of UE 104 or network entity 102 as described herein. Device 1100 may support wireless communication with one or more network entities 102, UE 104, or any combination thereof. Device 1100 may include components for bidirectional communication, including components for transmitting and receiving communications (such as processor 1102, memory 1104, transceiver 1106, and optional I / O controller 1108). These components may communicate electronically or be otherwise coupled (e.g., operative ground, communication ground, functional ground, electronic ground, electrical ground) via one or more interfaces (e.g., buses).

[0119] Processor 1102, memory 1104, transceiver 1106, or various combinations thereof, or various components thereof, may be examples of components used to perform various aspects of the present disclosure described herein. For example, processor 1102, memory 1104, transceiver 1106, or various combinations thereof, or components thereof, may support methods for performing one or more of the operations described herein.

[0120] In some implementations, processor 1102, memory 1104, transceiver 1106, or various combinations or components thereof may be implemented in hardware (e.g., in a communication management circuitry system). The hardware may include a processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination thereof, configured to or otherwise supporting components for performing the functions described in this disclosure. In some implementations, processor 1102 and memory 1104 coupled to processor 1102 may be configured to perform one or more functions described herein (e.g., by executing instructions stored in memory 1104 by processor 1102).

[0121] For example, according to the examples disclosed herein, processor 1102 may support wireless communication at device 1100. Device 1100 may be an example of UE 104. In this case, processor 1102 may be configured to operate to support: components for sending an indication to a network entity regarding whether an AI / ML function supports model identification; and components for performing a function-based LCM process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0122] Device 1100 may be an example of network entity 102 (e.g., a network entity). In this case, processor 1102 may be configured to support: components for receiving from the UE an indication of whether the AI / ML function supports model identification; and components for performing a function-based LCM process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0123] Processor 1102 may include intelligent hardware devices (e.g., general-purpose processors, DSPs, CPUs, microcontrollers, ASICs, FPGAs, programmable logic devices, discrete gate or transistor logic components, discrete hardware components, or any combination thereof). In some implementations, processor 1102 may be configured to use a memory controller to operate a memory array. In some other implementations, the memory controller may be integrated into processor 1102. Processor 1102 may be configured to execute computer-readable instructions stored in memory (e.g., memory 1104) to cause device 1100 to perform various functions of this disclosure.

[0124] Memory 1104 may include random access memory (RAM) and read-only memory (ROM). Memory 1104 may store computer-readable and computer-executable code, including instructions that, when executed by processor 1102, cause device 1100 to perform the various functions described herein. The code may be stored in a non-transitory computer-readable medium such as system memory or another type of memory. In some implementations, the code may not be directly executed by processor 1102, but may cause a computer (e.g., when compiled and executed) to perform the functions described herein. In some implementations, memory 1104 may include a basic I / O system (BIOS) that controls basic hardware or software operations, such as interaction with peripheral components or devices.

[0125] I / O controller 1108 can manage the input and output signals of device 1100. I / O controller 1108 can also manage peripheral devices not integrated into device 1100. In some implementations, I / O controller 1108 can represent a physical connection or port to an external peripheral device. In some implementations, I / O controller 1108 can utilize an operating system such as iOS®, Android®, MS Windows®, OS / 2®, UNIX®, LINUX®, or other known operating systems. In some implementations, I / O controller 1108 can be implemented as part of a processor, such as processor 1102. In some implementations, a user can interact with device 1100 via I / O controller 1108 or via hardware components controlled by I / O controller 1108.

[0126] In some implementations, device 1100 may include a single antenna 1110. However, in other implementations, device 1100 may have more than one antenna 1110 (i.e., multiple antennas), including multiple antenna panels or antenna arrays capable of concurrently transmitting or receiving multiple wireless transmissions. Transceiver 1106 may communicate bidirectionally via one or more antennas 1110, wired or wireless links, as described herein. For example, transceiver 1106 may represent a wireless transceiver and may communicate bidirectionally with another wireless transceiver. Transceiver 1106 may also include a modem for modulating packets, providing modulated packets to one or more antennas 1110 for transmission, and demodulating packets received from one or more antennas 1110. Transceiver 1106 may include one or more transmit chains, one or more receive chains, or combinations thereof.

[0127] The transmission chain can be configured to generate and transmit signals (e.g., control information, data, packets). The transmission chain may include at least one modulator for modulating data onto a carrier signal, preparing the signal for transmission over a wireless medium. At least one modulator may be configured to support one or more techniques, such as amplitude modulation (AM), frequency modulation (FM), or digital modulation schemes such as phase shift keying (PSK) or quadrature amplitude modulation (QAM). The transmission chain may also include at least one power amplifier configured to amplify the modulated signal to an appropriate power level suitable for transmission over a wireless medium. The transmission chain may also include one or more antennas 1110 for transmitting the amplified signal over the air or wireless medium.

[0128] The receiver chain can be configured to receive signals (e.g., control information, data, packets) via a wireless medium. For example, the receiver chain may include one or more antennas 1110 for receiving signals over the air or via a wireless medium. The receiver chain may include at least one amplifier (e.g., a low-noise amplifier (LNA)) configured to amplify the received signal. The receiver chain may include at least one demodulator configured to demodulate the received signal and acquire transmitted data by reversing the modulation technique applied during signal transmission. The receiver chain may include at least one decoder for decoding the demodulated signal to receive the transmitted data.

[0129] Figure 12 An example of a processor 1200 suitable for implementing some embodiments of the present disclosure is illustrated. Processor 1200 may be an example of a processor configured to perform various operations according to the examples described herein. Processor 1200 may include a controller 1202 configured to perform various operations according to the examples described herein. Processor 1200 may optionally include at least one memory 1204. Additionally or alternatively, processor 1200 may optionally include one or more arithmetic logic units (ALUs) 1206. One or more of these components may be electronically communicated or otherwise coupled (e.g., operative ground, communicative ground, functional ground, electronic ground, electrical ground) via one or more interfaces (e.g., buses).

[0130] Processor 1200 may be a processor chipset and includes a protocol stack (e.g., a software stack) executed by the processor chipset to perform various operations (e.g., receive, acquire, retrieve, send, output, forward, store, determine, identify, access, write, read) according to the examples described herein. The processor chipset may include one or more cores, one or more caches (e.g., memory local to the processor chipset or included in the processor chipset (e.g., processor 1200)) or other memory (e.g., random access memory (RAM), read-only memory (ROM), dynamic RAM (DRAM), synchronous dynamic RAM (SDRAM), static RAM (SRAM), ferroelectric RAM (FeRAM), magnetic RAM (MRAM), resistive RAM (RRAM), flash memory, phase-change memory (PCM), etc.).

[0131] Controller 1202 can be configured to manage and coordinate various operations of processor 1200 (e.g., signaling, receiving, acquiring, retrieving, transmitting, outputting, forwarding, storing, determining, identifying, accessing, writing, and reading) such that processor 1200 supports these operations according to the examples described herein. For example, controller 1202 can operate as a control unit of processor 1200 to generate control signals for managing the operation of various components of processor 1200. These control signals include enabling or disabling functional units, selecting data paths, initiating memory accesses, and coordinating operation timing.

[0132] Controller 1202 can be configured to fetch (e.g., fetch, retrieve, receive) instructions from memory 1204 and determine subsequent instructions(s) to be executed, enabling processor 1200 to support various operations according to the examples described herein. Controller 1202 can be configured to track the memory addresses of instructions associated with memory 1204. Controller 1202 can be configured to decode instructions to determine the operations to be performed and the operands involved. For example, controller 1202 can be configured to interpret instructions and determine control signals to be output to other components of processor 1200, enabling processor 1200 to support various operations according to the examples described herein. Additionally or alternatively, controller 1202 can be configured to manage data flow within processor 1200. Controller 1202 can be configured to control data transfers between registers, arithmetic logic unit (ALU), and other functional units of processor 1200.

[0133] Memory 1204 may include one or more caches (e.g., memory or other memory, such as RAM, ROM, DRAM, SDRAM, SRAM, MRAM, flash memory, etc., local to or included in processor 1200). In some implementations, memory 1204 may reside within or on the processor chipset (e.g., local to processor 1200). In some other implementations, memory 1204 may reside outside the processor chipset (e.g., remote from processor 1200).

[0134] Memory 1204 may store computer-readable, computer-executable code, including instructions that, when executed by processor 1200, cause processor 1200 to perform the various functions described herein. The code may be stored in a non-transitory computer-readable medium, such as system memory or another type of memory. Controller 1202 and / or processor 1200 may be configured to execute computer-readable instructions stored in memory 1204 to cause processor 1200 to perform various functions (e.g., supporting transmit power priority functions or tasks). For example, processor 1200 and / or controller 1202 may be coupled to or coupled to memory 1204, and processor 1200, controller 1202, and memory 1204 may be configured to perform the various functions described herein. In some examples, processor 1200 may include multiple processors, and memory 1204 may include multiple memories. One or more of the multiple processors may be coupled to one or more of the multiple memories, which may be configured individually or collectively to perform the various functions described herein.

[0135] One or more ALU 1206s can be configured to support a variety of operations as described in the examples herein. In some implementations, one or more ALU 1206s may reside within or on a processor chipset (e.g., processor 1200). In some other implementations, one or more ALU 1206s may reside outside the processor chipset (e.g., processor 1200). One or more ALU 1206s can perform one or more calculations on data, such as addition, subtraction, multiplication, and division. For example, one or more ALU 1206s can receive input operands and an opcode that determines the operation to be performed. One or more ALU 1206s are configured with various logic and arithmetic circuitry, including adders, subtractors, shifters, and logic gates, to process and manipulate data according to the operations. Alternatively or concurrently, one or more ALU 1206 may support logical operations such as AND, OR, XOR, NOR, and NAND, enabling one or more ALU 1206 to handle conditional operations, comparisons, and bitwise operations.

[0136] Based on the examples disclosed herein, processor 1200 may support wireless communication. Processor 1200 may be implemented at UE 104. In this case, processor 1200 may be configured to support: components for sending an indication to a network entity regarding whether an AI / ML function supports model identification; and components for performing a function-based LCM process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0137] Processor 1200 may be implemented at network entity 102 (e.g., base station). In this case, processor 1200 may be configured to support: components for receiving an indication from the UE regarding whether the AI / ML function supports model identification; and components for performing a function-based LCM process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

[0138] Figure 13 A flowchart illustrating method 1300 performed by a UE according to various aspects of this disclosure is shown. Operation of method 1300 may be implemented by the device or components thereof described herein. For example, operation of method 1300 may be performed by UE 104 described herein. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Alternatively or concurrently, the device may use dedicated hardware to perform aspects of the described functions.

[0139] At 1310, the method may include sending an indication to the network entity regarding whether the AI / ML function supports model identification. The operation at 1310 can be performed according to the examples described herein. In some implementations, aspects of the operation at 1310 can be found in the references. Figure 1 The aforementioned UE 104 is used for execution.

[0140] At 1320, the method may include: performing a function-based lifecycle management (LCM) process for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM process includes a model ID-based LCM process for at least one AI / ML model associated with the AI / ML function. The operation at 1320 can be performed according to the examples described herein. In some implementations, aspects of the operation at 1320 may be derived from references... Figure 1 The aforementioned UE 104 is used for execution.

[0141] Figure 14 A flowchart illustrating a method 1400 performed by a network entity according to various aspects of this disclosure is shown. The operation of method 1400 may be implemented by a device or component thereof described herein. For example, the operation of method 1400 may be performed by a network entity 102 as described herein. In some implementations, the device may execute a set of instructions to control the functional elements of the device to perform the described functions. Alternatively or additionally, the device may use dedicated hardware to perform aspects of the described functions.

[0142] At 1410, the method may include: receiving an indication from the UE regarding whether the AI / ML function supports model identification. The operation of 1410 can be performed according to the examples described herein. In some implementations, aspects of the operation of 1410 may be derived from references... Figure 1 The network entity 102 performs the execution.

[0143] At 1420, the method may include: performing a function-based LCM procedure for the AI / ML function, wherein, if the AI / ML function supports model identification, the function-based LCM procedure includes a model ID-based LCM procedure for at least one AI / ML model associated with the AI / ML function. The operation at 1420 can be performed according to the examples described herein. In some implementations, aspects of the operation at 1420 may be derived from references... Figure 1 The network entity 102 performs the execution.

[0144] It should be noted that the methods described in this paper describe possible implementations, and the operations and steps can be rearranged or otherwise modified, and other implementations are also possible. Furthermore, aspects from two or more methods can be combined.

[0145] The various illustrative blocks and components disclosed herein can be implemented or executed using a general-purpose processor, DSP, ASIC, CPU, FPGA or other programmable logic device, discrete gate or transistor logic, discrete hardware component or any combination thereof designed to perform the functions described herein. The general-purpose processor may be a microprocessor, but alternatively, the processor may be any processor, controller, microcontroller or state machine. The processor may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration).

[0146] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored on or transmitted via a computer-readable medium as one or more instructions or code. Other examples and implementations are within the scope of this disclosure and the appended claims. For example, due to the nature of software, the functions described herein may be implemented using software executed by a processor, hardware, firmware, hardwired, or any combination thereof. Features implementing the functions may also be physically located in various locations, including being distributed such that portions of the functions are implemented in different physical locations.

[0147] Computer-readable media include both non-transitory computer storage media and communication media, with communication media including any medium that facilitates the transfer of a computer program from one place to another. Non-transitory storage media can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, non-transitory computer-readable media can include RAM, ROM, electrically erasable programmable ROM (EEPROM), flash memory, optical disc (CD) ROM or other optical disc storage, magnetic disk storage or other magnetic storage devices, or any other non-transitory medium that can be used to carry or store desired program code components in the form of instructions or data structures and can be accessed by a general-purpose or special-purpose computer, or a general-purpose or special-purpose processor.

[0148] As used herein, including in the claims, the article “a” preceding an element is a non-limiting article and should be understood to mean “at least one” or “one or more” of those elements. The terms “a,” “at least one,” “one or more,” and “at least one of one or more” are interchangeable. As used herein, including in the claims, the word “or” used in a list of items (e.g., a list of items beginning with phrases such as “at least one of…” or “one or more of…” or “one or two of…”) indicates an inclusive list, such that, for example, a list of at least one of A, B, or C means A or B or C or AB or AC or BC or ABC (i.e., A and B and C). Furthermore, as used herein, the phrase “based on” should not be construed as a reference to a closed set of conditions. For example, an example step described as “based on condition A” may be based on both condition A and condition B without departing from the scope of this disclosure. In other words, as used herein, the phrase “based on” should be interpreted in the same manner as the phrase “at least partially based on.” Furthermore, as used herein, including in the claims, a “set” may include one or more elements.

[0149] The description provided herein is intended to enable those skilled in the art to make or manufacture this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles defined herein may be applied to other variations without departing from the scope of this disclosure. Therefore, this disclosure is not limited to the examples and designs described herein, but should be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A user equipment (UE), comprising: processor; as well as The transceiver is coupled to the processor. The processor is configured as follows: The transceiver sends an indication to the network entity regarding whether the artificial intelligence / machine learning (AI / ML) function supports model identification; and The function-based lifecycle management (LCM) process for the AI / ML functions is executed. Where the AI / ML function supports the model identifier, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

2. The UE according to claim 1, wherein the indication includes: A bit indicating the presence of at least one AI / ML model associated with the AI / ML function, wherein the at least one AI / ML model will be identified.

3. The UE of claim 1, wherein the indication includes an identifier of at least one AI / ML model associated with the AI / ML function, wherein the identifier is a model ID or a temporary ID.

4. The UE of claim 1, wherein the indication is included in the information regarding the AI / ML function.

5. The UE of claim 4, wherein during the function identification process, the indication is sent to the network entity along with the information regarding the AI / ML function.

6. The UE of claim 1, wherein the function-based LCM process further includes a model identification process, and the processor is further configured to: Based on the indication, determine the model ID of the at least one AI / ML model; and During the model identification process, the model ID is used to exchange information about the at least one AI / ML model with the network entity.

7. The UE of claim 6, wherein the model identification process is enabled after the AI / ML function has been activated.

8. The UE according to claim 1, wherein the AI / ML function is located on the UE side, and the at least one AI / ML model is a UE-side model or a UE-part model of a dual-side model.

9. A network entity, comprising: processor; as well as The transceiver is coupled to the processor. The processor is configured as follows: Receive from the user equipment (UE) via the transceiver an indication regarding whether the artificial intelligence / machine learning (AI / ML) function supports model identification; and The function-based lifecycle management (LCM) process for the AI / ML functions is executed. Where the AI / ML function supports the model identifier, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

10. The network entity of claim 9, wherein the indication includes: A bit indicating the presence of at least one AI / ML model associated with the AI / ML function, which will be identified.

11. The network entity of claim 9, wherein the indication includes an identifier of at least one AI / ML model associated with the AI / ML function, wherein the identifier is a model ID or a temporary ID.

12. The network entity of claim 9, wherein the indication is included in the information regarding the AI / ML function.

13. The network entity of claim 12, wherein during the function identification process, the indication is received from the UE together with the information regarding the AI / ML function.

14. The network entity of claim 9, wherein the function-based LCM process further includes a model identification process, and the processor is further configured to: Based on the indication, determine the model ID of the at least one AI / ML model; and During the model identification process, the model ID is used to exchange information about the at least one AI / ML model with the UE.

15. The network entity of claim 14, wherein the model identification process is enabled after the AI / ML function has been activated.

16. The network entity of claim 9, wherein the AI / ML function is located on the UE side, and the at least one AI / ML model is a UE-side model or a UE-part model of a dual-side model.

17. A processor for wireless communication, comprising: At least one memory; as well as A controller, coupled to the at least one memory, and configured such that the controller: Send instructions to network entities regarding whether artificial intelligence / machine learning (AI / ML) functions support model identification; and The function-based lifecycle management (LCM) process for the AI / ML functions is executed. Where the AI / ML function supports the model identifier, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

18. A method performed by a user equipment (UE), the method comprising: Send instructions to network entities regarding whether artificial intelligence / machine learning (AI / ML) features support model identification; as well as The function-based lifecycle management (LCM) process for the AI / ML functions is executed. Where the AI / ML function supports the model identifier, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

19. A method performed by a network entity, the method comprising: Receive an indication from the user equipment (UE) regarding whether the artificial intelligence / machine learning (AI / ML) function supports model identification; as well as The function-based lifecycle management (LCM) process for the AI / ML functions is executed. Where the AI / ML function supports the model identifier, the function-based LCM process includes a model identifier (ID)-based LCM process for at least one AI / ML model associated with the AI / ML function.

Images