Assistance for function selection at network

Abstract

The invention relates to assistance for function selection at a network. Methods, apparatuses, and computer program products provide means for function selection assistance to determine a selection of preferred functions for artificial intelligence / machine learning model features. An example method includes receiving a request for a supported function from a network node; identifying a supported function; determining function selection auxiliary information; and providing the function selection assistance information and an indication of the supported function to the network node. A method may optionally include receiving an indication of a selected one of supported functions based at least in part on function selection assistance information; and supporting the feature using the selected functionality.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. Provisional Application No. 63 / 518,498, filed on August 9, 2023, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The example embodiments described herein generally provide for selection of supported functions for an artificial intelligence / machine learning feature, and more specifically, provide a function selection aid to determine selection of preferred functions for an artificial intelligence / machine learning feature. Background Art

[0004] The mobile device operating on the network can perform a variety of functions. The functions performed on the mobile device support the features of the mobile device. These functions are supported on the local device based on the capabilities of the local device. These functions usually rely on the mobile network and the support provided via the network node to perform the functions of the supported features. These wireless communication services can be provided by a mobile network (also referred to as a cellular network). At least the last link in the mobile network is wireless, and via the mobile network, voice and / or data services are provided to multiple devices. The mobile network can be a third generation (3G), fourth generation (4G), and / or next generation (e.g., fifth generation or 5G) network.

[0005] Different devices have different capabilities, and networks support various subsets of capabilities. Compatibility between the network and the devices may be inefficient. Additionally, while a device may be able to perform certain functions, those functions may be inefficient or the device's capabilities may not adequately support those functions. These issues may result in delays and poor performance, leading to user dissatisfaction. Summary of the invention

[0006] Various embodiments generally relate to the selection of supported functions for artificial intelligence / machine learning model features, and more specifically, provide function selection assistance to determine the selection of preferred functions for artificial intelligence / machine learning model features. Some embodiments provided herein include an apparatus embodied by a user equipment (UE), the apparatus comprising at least one processor; and at least one memory comprising computer program code, the at least one memory and the computer program code being configured to, together with the at least one processor, cause the apparatus to: receive a request for supported functions and function selection assistance information from a network node; identify supported functions; determine function selection assistance information; and provide the function selection assistance information and an indication of supported functions to the network node. According to an embodiment, the apparatus is also configured to: receive an indication of a selected function among the supported functions based at least in part on the function selection assistance information; and use the selected function to support the feature.

[0007] According to some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions configured to perform features. Features of example embodiments include artificial intelligence (AI) / machine learning (ML) features. According to some embodiments, the AI / ML features include AI / ML-based location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. In some embodiments, the request for supported functions from the network node includes: a request for function selection assistance information.

[0008] According to some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0009] Another embodiment provided herein includes a computer program product having at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions including program code instructions to: receive a request for supported functions and function selection assistance information from a network node; identify supported functions; determine function selection assistance information; and provide the function selection assistance information and an indication of the supported functions to the network node. According to an embodiment, the computer program product program code instructions to: receive an indication of a selected function among the supported functions based at least in part on the function selection assistance information; and support a feature using the selected function.

[0010] According to some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions configured to perform features. Features of example embodiments include artificial intelligence (AI) / machine learning (ML) features. According to some embodiments, the AI / ML features include AI / ML-based location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. In some embodiments, the request for supported functions from the network node includes: a request for function selection assistance information.

[0011] According to some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0012] Another embodiment provided herein includes a method, comprising: receiving a request for supported functions and function selection assistance information from a network node; identifying supported functions; determining function selection assistance information; and providing the function selection assistance information and an indication of the supported functions to the network node. According to an embodiment, the apparatus is further configured to: receive an indication of a selected function among the supported functions based at least in part on the function selection assistance information; and support a feature using the selected function.

[0013] According to some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions configured to perform features. Features of example embodiments include artificial intelligence (AI) / machine learning (ML) features. According to some embodiments, the AI / ML features include AI / ML-based location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. In some embodiments, the request for supported functions from the network node includes: a request for function selection assistance information.

[0014] According to some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0015] Embodiments provided herein include an apparatus including: a component for receiving a request for supported functions and function selection assistance information from a network node; a component for identifying supported functions; a component for determining function selection assistance information; and a component for providing the function selection assistance information and an indication of supported functions to the network node. The example apparatus also includes: a component for receiving an indication of a selected function among the supported functions based at least in part on the function selection assistance information; and a component for supporting a feature using the selected function.

[0016] According to some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions configured to perform features. Features of example embodiments include artificial intelligence (AI) / machine learning (ML) features. According to some embodiments, the AI / ML features include AI / ML-based location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. In some embodiments, the request for supported functions from the network node includes: a request for function selection assistance information.

[0017] According to some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0018] Another embodiment provided herein includes a device comprising at least one processor; and at least one memory comprising computer program code, wherein the at least one memory and the computer program code are configured, together with the at least one processor, to cause the device to: request an identification of supported functions from the device; request function selection assistance information from the device; receive an identification of supported functions from the device; receive function selection assistance information from the device; select a selected function based on the identification of supported functions and the function selection assistance information; and provide the selected function to the device.

[0019] In some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions to perform the features. The features of some embodiments include artificial intelligence (AI) / machine learning (ML) features. The AI / ML features of some embodiments include location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. The device capabilities of some embodiments include one or more of the following: processing power, available memory, electrical power, device input / output conditions, or device connection status.

[0020] In some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0021] Another embodiment provided herein includes a computer program product having at least one non-transitory computer-readable storage medium having computer-executable program code instructions stored therein, the computer-executable program code instructions including program code instructions to: request an identification of supported functions from a device; request function selection assistance information from the device; receive an identification of supported functions from the device; receive function selection assistance information from the device; select a selected function based on the identification of supported functions and the function selection assistance information; and provide the selected function to the device.

[0022] In some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions to perform the features. The features of some embodiments include artificial intelligence (AI) / machine learning (ML) features. The AI / ML features of some embodiments include location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. The device capabilities of some embodiments include one or more of the following: processing power, available memory, electrical power, device input / output conditions, or device connection status.

[0023] In some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0024] Another embodiment provided herein includes a method comprising: requesting an identification of supported functions from a device; requesting function selection auxiliary information from the device; receiving an identification of supported functions from the device; receiving function selection auxiliary information from the device; selecting a selected function based on the identification of supported functions and the function selection auxiliary information; and providing the selected function to the device.

[0025] In some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions to perform the features. The features of some embodiments include artificial intelligence (AI) / machine learning (ML) features. The AI / ML features of some embodiments include location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. The device capabilities of some embodiments include one or more of the following: processing power, available memory, electrical power, device input / output conditions, or device connection status.

[0026] In some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0027] Embodiments provided herein include a network node, comprising: a component for requesting an identification of supported functions from a device; a component for requesting function selection auxiliary information from the device; a component for receiving an identification of supported functions from the device; a component for receiving function selection auxiliary information from the device; a component for selecting a selected function based on the identification of supported functions and the function selection auxiliary information; and a component for providing the selected function to the device.

[0028] In some embodiments, the supported functions include artificial intelligence (AI) / machine learning (ML) functions to perform the features. The features of some embodiments include artificial intelligence (AI) / machine learning (ML) features. The AI / ML features of some embodiments include location features, wherein each of the corresponding supported functions includes a different combination of device capabilities to perform the corresponding supported function. The device capabilities of some embodiments include one or more of the following: processing power, available memory, electrical power, device input / output conditions, or device connection status.

[0029] In some embodiments, the function selection assistance information includes one or more of: device preferences for supported functions; expected quality of service for supported functions; expected resource requirements for supported functions; support requirements for supported functions; likelihood of function switching between supported functions within a predetermined time period; or expected interruption of supported functions. In some embodiments, the function selection assistance information is based on at least one of resource availability of the device and resource requirements of supported functions.

[0030] The above invention summary is provided only to summarize some example embodiments to provide a basic understanding of some aspects of the present disclosure. Therefore, it should be understood that the above embodiments are merely examples and should not be interpreted as narrowing the scope or spirit of the present disclosure in any way. It should be understood that in addition to those summarized here, the scope also includes many potential embodiments, some of which will be further described below. Other features, aspects and advantages of the subject matter will become clear from the specification, drawings and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Having described certain example embodiments of the present disclosure in general terms above, non-limiting and non-exhaustive embodiments of the present disclosure will now be described with reference to the accompanying drawings, which are not necessarily drawn to scale. The components shown in the accompanying drawings may or may not be present in certain embodiments described herein. Some embodiments may include fewer (or more) components than those shown in the drawings.

[0032] Figure 1 illustrates an overview of an example mobile network (e.g., 5GS) according to various embodiments of the present disclosure;

[0033] Figure 2 A block diagram of an example apparatus for function selection assistance according to various embodiments of the present disclosure is provided;

[0034] Figure 3 illustrates a communication diagram of a process for function selection assistance between a network node and a user equipment device according to an example embodiment of the present disclosure;

[0035] Figure 4 is a flow chart of a method for function selection assistance between a network node and a user equipment device according to an example embodiment of the present disclosure; and

[0036] Figure 5 is another flow chart of a method for function selection assistance between a network node and a user equipment device according to an example embodiment of the present disclosure. DETAILED DESCRIPTION

[0037] Some embodiments of the present invention will be described more fully below with reference to the accompanying drawings, in which some but not all embodiments of the present invention are shown. In fact, various embodiments of the present invention can be embodied in many different forms and should not be construed as being limited to the embodiments described herein; on the contrary, these embodiments are provided so that the present disclosure meets applicable legal requirements. The same reference numerals always refer to the same elements. As used herein, the terms "data", "content", "information", "electronic information", "signal", "command" and similar terms can be used interchangeably to refer to data that can be captured, sent, received and / or stored according to various embodiments of the present disclosure. Therefore, the use of any such terms should not be regarded as limiting the spirit and scope of the embodiments of the present disclosure. In addition, when it is described herein that a first computing device receives data from a second computing device, it should be understood that the data can be received directly from the second computer device or indirectly via one or more intermediate computing devices (such as, for example, one or more servers, relays, routers, network access points, base stations, hosts, repeaters, etc., sometimes referred to as "networks" in this article). Similarly, where a first computing device is described herein as sending data to a second computing device, it should be understood that the data may be sent or transmitted directly to the second computing device, or may be sent or transmitted indirectly via one or more intermediate computing devices (such as, for example, one or more servers, remote servers, cloud-based servers (e.g., cloud utilities), relays, routers, network access points, base stations, hosts, repeaters, etc.).

[0038] The term "comprising" is meant to include, but not be limited to, the manner in which it is commonly used in the context of this patent, and should be interpreted accordingly. The use of broader terms such as "including," "comprising," and "having" should be understood to support narrower terms such as "consisting of," "consisting essentially of," and "consisting essentially of." Furthermore, where the terms "including" and "having" and variations thereof are used in the detailed description or claims, these terms are intended to be inclusive in a manner similar to the term "comprising."

[0039] The phrases "in one embodiment," "according to one embodiment," "in some embodiments," "in various embodiments," etc. generally mean that the particular feature, structure, or characteristic following the phrase may be included in at least one embodiment of the present disclosure, but not necessarily in all embodiments of the present disclosure. Thus, the particular feature, structure, or characteristic may be included in multiple embodiments of the present disclosure, such that these phrases do not necessarily refer to the same embodiment.

[0040] As used herein, the terms "example," "exemplary," and the like are used to mean "serving as an example, instance, or illustration." Any implementation, aspect, or design described herein as "example" or "exemplary" is not necessarily to be construed as preferred over other implementations, aspects, or designs. Rather, the use of the terms "example," "exemplary," and the like are intended to present concepts in a concrete manner.

[0041] If the specification specifies that a component or feature "may", "can", "might", "should", "will", "preferably", "likely", "usually", "optionally", "for example", "often" or "might" (or other such language) be included or have a characteristic, then the particular component or feature need not be included or have that characteristic. Such a component or feature may optionally be included in some embodiments or may be excluded.

[0042] As used herein, the term "computer-readable medium" refers to non-transitory storage hardware, non-transitory storage devices, or non-transitory computer system memory that can be accessed by a controller, microcontroller, computing system, or module of a computing system to encode computer-executable instructions or software programs thereon. Non-transitory "computer-readable medium" can be accessed by a computing system or module of a computing system to retrieve and / or execute computer-executable instructions or software programs encoded on the medium. Examples of non-transitory computer-readable media may include, but are not limited to, one or more types of hardware memory, non-transitory tangible media (e.g., one or more magnetic storage disks, one or more optical disks, one or more USB flash drives), computer system memory, or random access memory (such as DRAM, SRAM, EDO RAM), etc.

[0043] In addition, as used herein, the term "circuitry" refers to (a) a pure hardware circuit implementation (e.g., an implementation in an analog circuit system and / or a digital circuit system); (b) a combination of circuits and (multiple) computer program products, including software and / or firmware instructions stored on one or more computer-readable memories, which work together to cause the device to perform one or more functions described herein; and (c) circuits such as, for example, (multiple) microprocessors or (multiple) microprocessor portions, which require software or firmware to operate even if the software or firmware is not physically present. This definition of "circuitry" applies to all uses of the term herein, including in any claims. As another example, as used herein, the term "circuitry" also includes: an implementation including one or more processors and / or (multiple) portions thereof and accompanying software and / or firmware. As another example, as used herein, the term "circuitry" also includes, for example, a baseband integrated circuit or application processor integrated circuit for a mobile phone, or a similar integrated circuit in a server, a cellular network device, other network equipment (such as a core network device), a field programmable gate array, and / or other computing devices.

[0044] Wireless communication networks and their user equipment (UE) devices have growing capabilities. In addition, advances in artificial intelligence (AI) and machine learning (ML) provide opportunities for AI / ML capabilities. UE devices can benefit from AI / ML capabilities. For example, the positioning of UE devices can adopt AI / ML models. The positioning of UE devices based on available wireless signals and positioning services (e.g., global navigation satellite system (GNSS)) can be challenging. Positioning in environments where GNSS is weak, uncertain, or unavailable can rely on other processes, such as wireless access point fingerprint identification. AI / ML models that take into account UE capabilities can be used to meet the function of identifying the location of UE devices. For example, the AI / ML model can receive wireless fingerprints read at UE devices, available access points or nodes, and other possible capabilities to generate and provide outputs of high-precision locations of UE devices within the environment.

[0045] In addition to positioning, AI / ML models can also be used for other features. AI / ML models can be used for beam management. By using machine learning models, the setup of Layer 1 (PHY) and Layer 2 (MAC) processes for establishing and maintaining the best beam pair for connection can be enhanced and improved. Channel state information (CSI) feedback can be enhanced by using AI / ML models, such as space-frequency domain CSI compression using bilateral (network and UE) AI models or time domain CSI prediction using UE-side AI models. Other functions can be enhanced and improved by using the AI / ML models described in this article, and other functions can be further developed through other applications of AI / ML models.

[0046] AI / ML models can be processing intensive and require a lot of resources. In the case where multiple functions of the UE device enable AI / ML features, it can be determined to use the AI / ML functions at the UE device. The determination of which AI / ML functions are enabled can take into account the impact of function selection on the following: AI / ML resources at the UE, signaling overhead, and performance of the AI / ML functions. The embodiments described in this document support the identification and selection of functions for the UE that are most suitable for the UE, and generate less signaling overhead when implementing AI / ML features accurately and efficiently. AI / ML features can be implemented using many different AI / ML functions, each of which implements different performance for the same AI / ML feature.

[0047] According to an example embodiment, the UE assists the network in selecting / reselecting a function for the UE to enable an AI / ML feature. The assistance is provided by sending function (re)selection assistance information from the UE to the network, wherein for each supported specific combination of UE conditions for enabling an AI / ML feature (e.g., supported functions), various aspects of enabling the AI / ML feature are considered. These aspects include UE preferences, expected quality of service (QoS) performance, expected AI / ML resource requirements, AI / ML model LCM support requirements, the likelihood of a function switching request in a specific time window, and expected interruption and interruption duration in a specific time window.

[0048] UE preference considerations indicate the UE's preference in selecting the associated function, in terms of preference level (e.g., low, medium, high) or preference order (number) compared to other supported functions. Expected QoS performance indicates the expected QoS (e.g., latency, accuracy) performance of the AI / ML-enabled feature when the associated function is selected. Expected AI / ML resource requirements indicate the expected AI / ML resource (e.g., processing power, memory, electrical power, etc.) usage of the supported functions. Consideration of AI / ML model lifecycle management (LCM) support requirements indicates that when selecting functions for the UE to enable AI / ML features, the expected LCM operation (e.g., monitoring) and the cycle and duration of its LCM operation and / or the number of expected data samples for the operation are considered. The UE device may indicate that it needs more network support to enable the function. For example, the UE may report the accuracy of the function, and the network may identify the additional support required. The UE may determine and report to the network the cost of selecting a specific function, thereby assisting in the selection of appropriate functions for the UE. An embodiment may optionally consider the possibility of a function switching request within a specific time window, which indicates how likely the UE device is to send a function switching request within a specific duration. The expected outages and duration of outages within a specific time window are considered, which indicate the potential downtime of the functionality.

[0049] According to the example embodiments described herein, a UE device is configured to receive a request for AI / ML function selection assistance information from a network node (e.g., a gNodeB, a location management function (LMF), etc.). The AI / ML function selection assistance information is determined at least in part based on: the UE's AI / ML capabilities, AI / ML resource availability, AI / ML resource requirements, and LCM requirements of the associated underlying AI / ML model of the supported functions, wherein the resource availability / requirements include UE characteristics such as processing power, memory, electrical power, etc. The UE then sends the AI / ML function selection assistance information to the network node, wherein for each supported function, the assistance information contains information about one or more of the following: UE preferences, expected QoS performance, AI / ML model LCM support requirements, the possibility of function switching requirements within a time window, expected interruptions within a time window, etc. The UE then receives an indication of the selected function from the network node.

[0050] According to some embodiments, the network node is configured to request the AI / ML function selection assistance information from the UE. The AI / ML function is selected based at least in part on the AI / ML function selection assistance information. An indication of the selected AI / ML function is provided to the UE.

[0051] Reference now Figure 1 , illustrates an example mobile network 100. Mobile network 100 (also referred to as a cellular network) is a network in which at least the last link is wireless and provides voice and / or data services to multiple devices. Mobile network 100 may be a third generation (3G), fourth generation (4G), and / or next generation (e.g., fifth generation or 5G) network.

[0052] The mobile network 100 is illustrated as providing communication services to UE 110. UE 110 may be enabled for voice services, data services, machine-to-machine (M2M) or machine-type communication (MTC) services, Internet of Things (IoT) services, and / or other services. UE 110 may be an end-user device such as a mobile phone (e.g., a smartphone), a tablet or PDA, a computer with a mobile broadband adapter, etc.

[0053] The mobile network 100 includes one or more radio access networks (RAN 120) that communicate with the UE 110 through a radio interface. The RAN 120 of an example embodiment may support Evolved UMTS Terrestrial Radio Access Network (E-UTRAN) access, Wireless Local Area Network (WLAN) access, fixed access, satellite radio access, new radio access technology (RAT), etc. As an example, the RAN 120 may include an E-UTRAN or a Next Generation RAN (NG-RAN), which includes one or more base stations 124 dispersed over a geographical area. The base station 124 may include an entity that communicates with the UE on a licensed spectrum using a radio communication technology and interfaces the UE with the core network 130. The base station 124 in the E-UTRAN may be referred to as an evolved NodeB (eNodeB). The base station 124 in the NG-RAN may be referred to as a gNodeB (NR base station) and / or an ng-eNodeB (LTE base station supporting a 5G core network). As another example, the RAN 120 may include a WLAN that includes one or more wireless access points (WAPs). A WLAN is a network in which a UE can connect to a local area network (LAN) through a wireless (radio) connection. A WAP is a node that communicates with a UE over an unlicensed spectrum using radio communication technology and provides the UE with access to a core network. An example of a WAP is a Wi-Fi access point operating on the 2.4 GHz or 5 GHz radio band. The term "base station" may then refer to an eNodeB, gNodeB, ng-eNodeB, WAP, etc.

[0054] UE 110 is able to attach to a cell of RAN 120 to access core network 130. Thus, RAN 120 represents a radio interface between UE 110 and core network 130. Core network 130 is a central part of mobile network 100 that provides various services to customers connected through RAN 120. One example of core network 130 is an evolved packet core (EPC) network described by 3GPP for LTE. Another example of core network 130 is a 5G core (5GC) network described by 3GPP. Core network 130 includes network elements 132, which may include servers, devices, apparatuses, or equipment (including hardware) that provide services to UE 110. Network elements 132 in the EPC network may include a mobility management entity (MME), a serving gateway (S-GW), a packet data network gateway (P-GW), and the like. The network elements 132 in the 5G network may include access and mobility management function (AMF), session management function (SMF), user plane function (UPF), policy control function (PCF), unified data management (UDM), etc.

[0055] Reference now Figure 2, an example apparatus 300 is provided. In some cases, the apparatus 300 may be an embodiment of the UE 110 and / or may be embodied by or otherwise associated with the UE 110. Alternatively, the apparatus 300 may be an embodiment of the network element 132, or may be embodied by or otherwise associated with the network element 132 (such as the PCF 260). In an example embodiment, the apparatus 300 is embodied by a UE device for implementing AI / ML functions. Embodiments of the present disclosure provide a method that supports the identification and selection of such UE functions that are most suitable for the UE, and generates less signaling overhead when implementing AI / ML features.

[0056] The apparatus 300 may include a processor 302, a memory 304, and a network interface 306. The apparatus 300 may be configured to perform the operations described herein. Although these components are described with respect to the performance of various functions, it should be understood that a particular implementation necessarily includes the use of specific hardware. It should also be understood that some of these components may include similar or general hardware. For example, two groups of circuits may utilize the same processor, network interface, storage medium, etc. to perform their related functions, so that there is no need to repeat hardware for each group of circuits.

[0057] In some embodiments, the processor 302 (and / or a coprocessor or auxiliary processor or any other processing circuit system otherwise associated with the processor) can communicate with the memory 304 via a bus to transfer information between components of the device. The memory 304 is non-transitory and can include, for example, one or more volatile and / or non-volatile memories. In other words, for example, the memory 304 can be an electronic storage device (e.g., a computer-readable storage medium). The memory 304 can be configured to store information, data, content, applications, instructions, etc. to enable the device to perform various functions according to the example embodiments disclosed herein.

[0058] The processor 302 can be implemented in a variety of different ways, and for example, can include one or more processing devices configured to execute independently. In some non-limiting embodiments, the processor 302 can include one or more processors configured in series via a bus to implement independent execution of instructions, pipelining and / or multithreading. The use of the term "processor" can be understood to include a single-core processor, a multi-core processor, multiple processors within a device, and / or a remote or "cloud" processor.

[0059] In some embodiments, the processor 302 may be configured to execute instructions stored in the memory 304 and / or in a circuit system accessible to the processor 302. In some embodiments, the processor 302 may be configured to perform hard-coded functions. Therefore, whether configured by hardware or software methods, or by a combination thereof, the processor 302 may represent an entity (e.g., physically embodied in a circuit system) that is capable of performing operations according to the embodiments disclosed herein when configured accordingly. Alternatively, as another example, when the processor 302 is embodied as an executor of software instructions, the instructions may specifically configure the processor 302 to perform the algorithms and / or operations described herein when the instructions are executed.

[0060] The network interface 306 can be any component, such as a device or circuit system embodied in hardware or a combination of hardware and software, which is configured to receive and / or send data from / to a network and / or any other device, circuit system, or module that communicates with the device 300. In this regard, the network interface 306 may include, for example, a network interface for implementing communication with a wired or wireless communication network. For example, the network interface 304 may include one or more network interface cards, antennas, buses, switches, routers, modems, and supporting hardware and / or software, or any other device suitable for implementing communication via a network. Additionally or alternatively, the network interface 306 may include a circuit system for interacting with an antenna to transmit a signal via the antenna or process the reception of a signal received via the antenna.

[0061] It should also be noted that all or some of the information discussed herein may be based on data received, generated, and / or maintained by one or more components of the device 300. In some embodiments, one or more external systems (such as remote cloud computing and / or data storage systems) may also be utilized to provide at least some of the functionality discussed herein.

[0062] As described herein, a function refers to an AI / ML-enabled feature enabled by a configuration that is supported by conditions based on UE capabilities. A function is characterized as a set of unique UE conditions that enable a specific AI / ML feature. More than one function may be defined within an AI / ML-enabled feature. Different functions for implementing an AI / ML-enabled feature may have different levels of accuracy, resource consumption, efficiency, and have different UE condition requirements. The embodiments described herein provide a mechanism by which the UE provides assistance for function selection at the network. Thus, the UE can determine which function is most appropriate given a set of environmental or UE conditions.

[0063] The UE conditions are indicated by UE capabilities from the UE. The network can use these conditions to identify and select functions for the UE based on the indicated UE conditions to enable the desired AI / ML features (e.g., direct AI / ML positioning at the UE). The UE can then use its AI / ML resources to support the line of sight of the AI / ML features according to the indicated functions, such as processing resources (e.g., in the form of a processor 302), memory resources (e.g., in the form of a memory 304), and power resources (e.g., in the form of remaining battery life), as well as available AI / ML models. The network may not be fully aware of the details of the AI / ML models. In addition, the network (or UE) performs function monitoring and performs function switching in the event of a change in function performance (such as performance degradation).

[0064] Depending on the deployment environment (e.g., city, highway, etc.), each function may have different requirements for AI / ML resources and LCM requirements. Therefore, the functions selected for the UE to enable specific AI / ML features have an impact on AI / ML resource usage and signaling overhead, as well as the complexity associated with LCM (such as performance monitoring). AI / ML resource availability can be dynamic at the UE. The UE can participate in different AI / ML tasks at the same time. The impact of AI / ML resources may result in increased latency for model LCM, such as inference, training, etc., depending on the selected function. In addition, function switching generates signaling overhead, and therefore if function selection results in frequent function switching, the function selection affects the signaling overhead.

[0065] A specific function is characterized by a range of parameter values ​​for a specific set of UE conditions. For the positioning use case, a function represents a specific configuration of a unique set of UE conditions that achieves a specific positioning feature. A specific positioning feature can be achieved by several functions, where each function is configured to use a certain combination of UE conditions (and thus UE capabilities).

[0066] Functions are characterized by a specific unique set of UE conditions that implement specific AI / ML features. The set of specific UE conditions is associated with one or more underlying AI / ML models. Each function is supported by one or more underlying AI / ML models. The details of the underlying model (such as the AI / ML resource requirements of the model) may not be clear enough at the network, and at least in some aspects (if not all), the model LCM may be transparent to the network. In addition, different AI / ML models may have different complexities, AI / ML resource requirements, and support different QoS (e.g., accuracy) levels and generation performance. In addition, AI / ML resource availability can change dynamically at the UE. Therefore, especially when multiple functions are supported at a UE that enables AI / ML features, the determination of which function to select must take into account the impact of function selection on the AI / ML resources, signaling overhead, and performance of the AI / ML features at the UE.

[0067] According to example embodiments described herein, AI / ML resources at a UE may include: processing power (e.g., graphics processing unit (GPU), compute processing unit (CPU), tensor processing unit (TPU), etc.), available memory (e.g., random access memory (RAM), read-only memory (ROM), etc.), power (e.g., battery status), device input / output conditions (e.g., sensor blocking, sensor data loss, etc.), and device connection status (e.g., upload / download throughput, link quality, etc.).

[0068] The UE device assists the network in (re)selecting a function by sending function (re)selection assistance information. The assistance information message includes a list of supported UE condition sets, which for simplicity is referred to as functions in the following, even though the function can be identified after the UE condition set is reported. For each function, one or more assistance information elements are provided. These include UE preference order, expected QoS, AI / ML model LCM support requirements, the possibility of function switching requests within a specific duration, or expected interruptions within a specific time window and the duration of the interruption. Each item is described in detail below.

[0069] The UE preference order may include situations where the UE device supports multiple functions and may prefer one function over another given its AI / ML resource availability. For example, when power availability is relatively low, the UE may prefer functions that consume less power. For different functions, the UE device may use different AI / ML models, and therefore the AI / ML resource requirements may be different. UE preferences may be established by various factors. For example, the UE may need to establish the UE's location, although the accuracy of the location required may be lower in certain circumstances. The UE may have competing demands for other resource-consuming services, making the UE more willing to use fewer resources for UE positioning. In this case, the UE preference may indicate that a specific function with relatively low resource requirements is more preferred for obtaining AI / ML location features.

[0070] For expected QoS (e.g., latency, accuracy) performance, there may be additional delays in performing AI / ML tasks (e.g., model inference) depending on the AI / ML resource availability at the UE device and the AI / ML resource requirements of the underlying model. This may result in additional delays or waiting times. In addition, the underlying models associated with different functions may have different accuracy performance.

[0071] AI / ML model LCM support requirements may include the frequency with which performance monitoring is performed, or the amount of data (e.g., the number of data samples) required for AI / ML model LCM purposes. The (multiple) underlying AI / ML models associated with a function may have different generalization performance. Therefore, given the mobility of UE devices, performance monitoring requirements may vary depending on mobility patterns in certain regions or certain environments.

[0072] The likelihood of a function switching request within a specific duration may reflect that given AI / ML resource availability, the underlying model of the UE device, and mobility patterns, the UE may calculate and indicate the likelihood of the need for a function switching within a given time window.

[0073] The expected outage within a specific time window and the duration of the outage may reflect that the UE device may be temporarily unable to support the function due to AI / ML resource limitations. In this case, if there is any expected potential downtime for a function taking into account the available AI / ML resources and the expected use of AI / ML resource consumption by the function and other tasks being / planned to be performed at the UE device, the expected outage and duration may be indicated.

[0074] Figure 3An example embodiment of a message flow for UE assistance for function (re)selection at the network is illustrated. As shown, in order to support use cases such as AI / ML assisted positioning involving UE, the network node (e.g., gNB, LMF) needs to (re)select appropriate functions for the UE. At 410, the network node 405 requests a list of supported functions from the UE. The request for supported functions identifies the functions that the UE is able to support AI / ML features. The request 410 also includes: a request for function selection assistance information from the UE.

[0075] At 420, the UE identifies supported functions and determines function selection assistance information for each supported function. The supported functions are based on the capabilities of the UE, such as battery life, processing power, available nodes, or other factors that determine the resources available to the UE to perform the function. Among the supported functions, the function selection assistance information is related to the above conditions, such as UE preference, expected QoS, expected AI / ML resource requirements, required AI / ML model LCM support, possibility of function switching, and expected interruption. The UE function selection assistance information for each supported function is based on factors such as AI / ML resource availability, AI / ML resource requirements, and LCM requirements of the underlying AI / ML model associated with the supported function.

[0076] At 430, the UE reports not only the supported functions to the network 405, but also the function selection assistance information. The network node identifies the supported functions as a list of available functions that can be selected from, and at 440, uses the function selection assistance information as input to identify the most suitable function among the supported functions. The network identifies the most suitable function based on a combination of multiple factors, including one or more of the following: UE preference, QoS of supported functions, expected AI / ML resource requirements of supported functions, minimum AI / ML model LCM support requirements, low probability of necessity of function switching (e.g., within a predetermined time interval), or low probability of expected interruption (e.g., within a predetermined time interval). The most suitable function can optionally be determined based on internal network conditions (e.g., cell load, history of UE being able to support the provided functions, etc.). The network may choose to consider a portion of the function selection assistance information rather than relying exclusively on the information to select the most suitable function. The most suitable function is selected as the selected function, and the selected function is indicated to the UE at 450. At 460, the UE supports the AI / ML feature based on the selected function.

[0077] The example embodiments described herein take into account the UE itself when determining the functions to be supported by the UE. While the network may select functions for the UE device based on the capabilities of the UE device and the functions supported at the UE device, the embodiments provided herein improve this function selection by considering various factors specific to the UE when establishing the functions to be used at the UE. This can improve the performance of the UE, thereby providing a better user experience, which is beneficial to both the user and the service provider.

[0078] The embodiments described herein provide significant improvements to the functionality of the computer itself. Specifically, the example embodiments enable devices such as UE devices to identify supported functions and establish function selection assistance information. The function selection assistance information provides UE preference information predicted based on various features of the UE and the priority of the actions performed by the UE. The function selection assistance information also establishes the expected quality of service performance of the performance of the AI / ML-enabled features in terms of latency, accuracy, etc. when the associated function is selected. This enables the most efficient and effective functions to be selected to improve the quality of service by reducing latency and improving accuracy when available. The function selection assistance information also provides AI / ML model LCP support requirements, which indicate the exceptional LCM operations required for the function. The UE device can indicate that it needs more support from the network to enable the function, and provide an estimate of which specific functions the network will spend in terms of support. This enables the network to determine the available support capabilities to determine whether such a function is feasible. The function selection assistance information may also include the possibility of a function switching request during a specific time window. If the function is switched, inefficiency is introduced, so that the possibility of the function being switched may indicate a function that cannot be reliably implemented at the UE device. Function selection assistance information may also include expected interruptions over a period of time and the duration of the interruptions, which enables the selection of a function to minimize the expected interruptions, thereby improving the functionality of the AI / ML model features themselves. Therefore, the embodiments described herein are specifically directed to improving the functionality of the computing device itself through function selection assistance information.

[0079] Nevertheless, function selection assistance information is implemented in practical applications that support AI / ML model features, such as determining position / location, beam management, and CSI feedback enhancement. By using the AI ​​model for space-frequency domain CSID compression, the AI / ML model feature operation on the UE device is improved. Beam management, as a practical application, enables spatial downlink beam prediction for a set of beams based on the measurement results of another set of beams. Using the AI / ML model described in this article, the temporal downlink beam prediction of the first set of beams can be based on the historical measurement results of the second set of beams. In addition, the AI / ML model can provide positioning enhancement through direct AI / ML positioning, such as using the AI / ML model output for UE location. Certain example embodiments described herein enhance positioning by using fingerprint recognition based on channel observations as an input to the AI / ML model, and the output is a more accurate location. The embodiments described herein are implemented into various practical applications, and each of these applications benefits by using the unique provision of function selection assistance using UE capabilities.

[0080] Figure 4 and Figure 5 A flow chart illustrating an example embodiment of a method for providing function selection assistance to determine the selection of preferred functions for artificial intelligence / machine learning model features. Figure 4 As shown, at 510, a request for supported functions and function selection assistance information is received from a network node. The request may be received by a component such as, for example, a network interface 306 and / or a processor 302 of a UE device embodied by the apparatus 300. At 520, supported functions are identified, and at 530, function selection assistance information is determined. This may be performed by a component such as, for example, a processor 302 of a UE device. At 540, the function selection assistance information and an indication of supported functions are provided to the network node. This may be performed by a component such as, for example, a network interface 306 and / or a processor 302 of a UE device embodied by the apparatus 300. Optionally, as shown by the dashed line of element 550, at 550, an indication of a selected function among supported functions based at least in part on the function selection assistance information is received. The indication may be received at the UE device, such as by a component such as a network interface 306 or a processor 302 of the apparatus 300. At 560, the selected function is used to support the feature. This may be performed by a component such as, for example, a processor 302 of a UE device embodied by the apparatus 300. The selected functions may include functions that utilize the capabilities of the UE device to perform features, such as a positioning feature of the UE device using a UE-side AI / ML model.

[0081] Figure 5Another process for selecting a function using function selection assistance information described herein is illustrated. At 610, an identification of supported functions is requested. The request may be made by a component such as a processor 302 of a network node embodied by an apparatus 300 for generating a request and a network interface 306 for sending the request. At 620, function selection assistance information is requested. The request may also be made by a component such as a processor 302 of a network node embodied by an apparatus 300 for generating a request and a network interface 306 for sending the request. For example, the request may be directed to a UE device in which a function is executed to implement a feature. At 630, an indication of supported functions is received, and at 640, function selection assistance information is received. These may be received by a component such as a processor 302 and / or a network interface 306 of a network node embodied by an apparatus 300. At 650, a selection of a selected function is performed based on the identification of supported functions and the function selection assistance information. The selection may be performed by a component such as, for example, a processor 302 of a network node embodied by an apparatus 300. At 660 , the selected functionality is provided to the apparatus, which may include a UE device, and the selected functionality is provided by a component such as the processor 302 and / or the network interface 306 of the network node embodied by the apparatus 300 .

[0082] Figure 4 and Figure 5A flowchart describing the operation according to an example embodiment of the present disclosure is illustrated. It should be understood that each block of the flowchart and the combination of blocks in the flowchart can be implemented in various ways, such as hardware, firmware, processors, circuit systems and / or other communication devices associated with the execution of software including one or more computer program instructions. For example, one or more of the above-mentioned processes or operations can be embodied by computer program instructions. In this regard, the computer program instructions embodying the above-mentioned processes or operations can be stored by the memory 304 of the device (e.g., device 300, UE 110) using an embodiment of the present invention, and executed by the processor 302. As will be understood, any such computer program instructions can be loaded onto a computer or other programmable device (e.g., hardware) to produce a machine so that the resulting computer or other programmable device implements the functions specified in the flowchart blocks. These computer program instructions can also be stored in a computer-readable memory, which can instruct a computer or other programmable device to operate in a specific manner so that the instructions stored in the computer programmable memory produce a product, the execution of which implements the functions specified in the flowchart blocks. Computer program instructions may also be loaded onto a computer or other programmable apparatus to cause a series of operations to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions executed on the computer or other programmable apparatus provide operations for implementing the functions specified in the flowchart blocks.

[0083] Therefore, the blocks of the flowchart support the combination of components for performing the specified functions and the combination of operations for performing the specified functions. It will also be understood that one or more blocks of the flowchart and the combination of blocks in the flowchart can be implemented by a computer system based on special-purpose hardware or a combination of special-purpose hardware and computer instructions that performs the specified functions.

[0084] Many modifications and other embodiments of the invention described herein will occur to those skilled in the art with the benefit of the teachings presented in the above description and the associated drawings. Therefore, it should be understood that the invention is not limited to the specific embodiments disclosed, and modifications and other embodiments are intended to be included within the scope of the appended claims.

[0085] Furthermore, although the foregoing description and the associated drawings describe example embodiments in the context of certain example combinations of elements and / or functions, it should be understood that alternative embodiments may provide different combinations of elements and / or functions without departing from the scope of the appended claims. In this regard, combinations of elements and / or functions that differ from those explicitly described above may also be contemplated, for example, as described in some of the appended claims. Although specific terms are used herein, they are used only in a general and descriptive sense and not for purposes of limitation.

Claims

1. A device for communication, comprising: means for receiving a request for supported functions and function selection assistance information from a network node; A component for identifying the supported functions; A component for determining the function selection auxiliary information; as well as Means for providing the function selection assistance information and the indication of supported functions to the network node.

2. The device according to claim 1, wherein the device further comprises: means for receiving an indication of a selected function among the supported functions, wherein the selected function is selected based at least in part on the function selection assistance information; as well as A component for supporting a feature using the selected functionality.

3. The apparatus according to claim 2, wherein the supported functions include: Artificial intelligence AI / machine learning ML functionality configured to perform the described features.

4. The apparatus according to claim 2 or 3, wherein the features include artificial intelligence (AI) / machine learning (ML) features.

5. The apparatus of claim 4, wherein the AI / ML features include AI / ML-based location features, wherein each corresponding supported function includes: Different combinations of device capabilities to perform the corresponding supported functions.

6. The apparatus of any one of claims 1 to 5, wherein the request from the network node for the supported functionality comprises: A request for auxiliary information for feature selection.

7. The device according to any one of claims 1 to 6, wherein the function selection auxiliary information comprises one or more of the following: device preferences for the supported functions; the expected quality of service of the supported functionality; the expected resource requirements of the supported functionality; Support requirements for the supported functions; the possibility of function switching between the supported functions within a predetermined time period; or Expected disruption of the supported functionality.

8. The apparatus according to any one of claims 1 to 7, wherein the function selection assistance information is based on at least one of resource availability of the apparatus or resource requirements of the supported functions.

9. An apparatus for communication, comprising: A component for requesting supported functions and function selection auxiliary information from a user device; means for receiving the supported functions and the function selection assistance information from the user equipment; means for selecting a function based on said supported functions and said function selection assistance information; as well as Means for causing the selected said functionality to be provided to said user equipment.

10. The apparatus of claim 9, wherein the supported functions include: Artificial intelligence AI / machine learning ML functions configured to perform artificial intelligence AI / machine learning ML features.

11. The apparatus of claim 10, wherein the AI / ML features include location features, wherein each corresponding supported function includes: Different combinations of device capabilities to perform the corresponding supported functions.

12. The device of claim 11, wherein the device capabilities include one or more of: processing power, available memory, electrical power, device input / output conditions, or device connection status.

13. The device according to any one of claims 9 to 12, wherein the function selection auxiliary information comprises one or more of the following: User device preferences for the supported functions; the expected quality of service of the supported functionality; the expected resource requirements of the supported functionality; Support requirements for the supported functions; the possibility of function switching between the supported functions within a predetermined time period; or Expected disruption of the supported functionality.

14. A method for communication, comprising: receiving a request for supported functions and function selection assistance information from a network node; Identify the supported functionality; Determining the function selection auxiliary information; as well as The function selection assistance information and the indication of the supported functions are provided to the network node.

15. A method for communication, comprising: Requesting supported functions and function selection assistance information from the user equipment; receiving the supported functions and the function selection assistance information from the user equipment; selecting a function based on the supported functions and the function selection assistance information; as well as The selected function is provided to the user equipment.

Images