Method, apparatus and computer program

Abstract

A user equipment comprising: means for sending a request for discovery of one or more network entities to provide one or more artificial intelligence / machine learning services for the user equipment; means for receiving a response to the request for discovery, the response indicating that the request has been verified; and means for, in response to receiving the response to the request for discovery, sending a request for the providing of one or more artificial intelligence / machine learning model services for the user equipment.

Description

METHOD, APPARATUS AND COMPUTER PROGRAM

[0001] Technical Field

[0002] The present disclosure relates to apparatus, methods and computer program. In particular, but not exclusively, the present disclosure relates to apparatus, methods and computer programs where a user equipment consumes artificial intelligence / machine learning services.

[0003] Background

[0004] A communication network can be seen as a facility that enables communications between two or more terminal devices or provides terminal devices access to at least a data network. A mobile or wireless communication network is one example of a communication network. A terminal device may be provided with a service or application.

[0005] Such communication networks operate in according with standards such as those provided by 3 GPP (Third Generation Partnership Project) or ETSI (European Telecommunications Standards Institute). Examples of standards are the so-called 5G (5th Generation) or so-called 6G (6th Generation) under definition standards provided by 3GPP.

[0006] It has been proposed that terminal devices be involved in artificial intelligence / machine learning (AI / ML) services.

[0007] Summary

[0008] Some example embodiments of this disclosure will be described with respect to certain aspects. These aspects are not intended to indicate key or essential features of the embodiments of this disclosure, nor are they intended to be used to limit the scope of thereof. Other features, aspects, and elements will be readily apparent to a person skilled in the art in view of this disclosure.

[0009] According to first aspect of the invention, there is provided a user equipment comprising: means for sending a request for discovery of one or more network entities to provide one or more artificial intelligence / machine learning services for the user equipment; means for receiving a response to the request for discovery, the response indicating that the request has been verified; and means for, in response to receiving the response to the request for discovery, sending a request for the providing of one or more artificial intelligence / machine learning model services for the user equipment.

[0010] Other features may be seen from the claims dependent on claim 1.

[0011] According to second aspect, there is provided a method comprising: sending a request for discovery of one or more network entities to provide one or more artificial intelligence / machine learning services for the user equipment; receiving a response to the request for discovery, the response indicating that the request has been verified; and in response to receiving the response to the request for discovery, sending a request for the providing of one or more artificial intelligence / machine learning model services for the user equipment.

[0012] The one or more artificial intelligence / machine learning services may comprise a service in which the network entity assists in the training of a machine learning model for the user equipment.

[0013] The response may comprise information identifying a network entity which is able to provide the one or more artificial intelligence / machine learning services for the user equipment.

[0014] The request for the providing of the one or more artificial intelligence / machine learning services for the user equipment may be sent to the network entity identified by the information identifying the network entity in the response.

[0015] The response to the request for discovery may comprise an access token.

[0016] The request for assistance in the training of the machine learning model for the user equipment may comprise the access token.

[0017] The response to the request may comprise one or more of: information about one or more artificial intelligence / machine learning services provided by the network entity; an identifier of the user equipment; information about a manufacturer of the user equipment; location information associated with the providing of the one or more artificial intelligence / machine learning services provided by the network entity; or information about a vendor of the user equipment.

[0018] The request for discovery of one or more network entities may comprise one or more of: information indicating that a service for training of a machine learning model is required; a user equpment identifier; an analytics type identifier; or information about one or more privacy requirements associated with data to be used in the machine learning model training.

[0019] The request for assistance may comprise one of more of: the analytics typeidentifier; or data for use in the training of the machine learning model.

[0020] The method may comprise receiving information from the network entity relating to a trained machine learning model.

[0021] The information may comprise information about a unform resource indicator from which trained machine learning model can be downloaded.

[0022] The method may comprise providing, to a network, information relating to one or more artificial intelligence / machine learning capabilities of the user equipment.

[0023] The information relating to one or more artificial intelligence / machine learning capabilities of the user equipment may comprise information about one or more of one or more datasets stored by the user equipment; one or more datasets generated by the user equipment; a manufacturer identity associated with the user equipment; a vendor identity associated with the user equipment; a serving network operator of the user equpment; end-to-end encryption support provided by the user equipment; or one or more artificial intelligence / machine learning service subscriptions.

[0024] The method may be performed by an apparatus.

[0025] The apparatus may be a user equipment or provided in a user equipment.

[0026] The apparatus may comprise at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to provide one or more of the methods of the second aspect.

[0027] According to a third aspect, there is provided an apparatus comprising means for: receiving a request from a user equipment to discover one or more network entities capable of providing one or more artificial intelligence / machine learning services to the user equipment; verifying the request using information about one or more network entities which are able to provide the one or more artificial intelligence / machine learning services to the user equipment; and sending to the user equipment, a response to the request for discovery, when the request has been verified.

[0028] Other features may be seen from the claims dependent on claim 14.

[0029] The apparatus may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function.

[0030] According to a fourth aspect, there is provided method comprising: receiving a request from a user equipment to discover one or more network entities capable ofproviding one or more artificial intelligence / machine learning services to the user equipment; verifying the request using information about one or more network entities which are able to provide the one or more artificial intelligence / machine learning services to the user equipment; and sending to the user equipment, a response to the request for discovery, when the request has been verified.

[0031] The method may comprise obtaining information about one or more artificial intelligence / machine learning capabilities of the user equipment and further using the obtained information about the user equipment to verify the request.

[0032] The one or more artificial intelligence / machine learning services may comprise a service in which the network entity assists in the training of a machine learning model for the user equipment.

[0033] The response may comprise information identifying a network entity which is able to provide the one or more artificial intelligence / machine learning services for the user equipment.

[0034] The response to the request for discovery may comprise an access token.

[0035] The method may be performed by an apparatus.

[0036] The apparatus may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function.

[0037] The apparatus may comprise at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to provide one or more of the methods of the fourth aspect.

[0038] According to fifth aspect of the invention, there is provided a user equipment comprising: means for providing, to a network, information relating to one or more artificial intelligence / machine learning capabilities of the user equipment; means for receiving information for authorizing access to one or more artificial intelligence / machine learning services of the user equipment, and a request for one of the one or more artificial intelligence / machine learning services supported by the user equipment; and means for providing the one of the one or more artificial intelligence / machine learning services which have been requested.

[0039] According to sixth aspect, there is provided a method comprising: providing, to a network, information relating to one or more artificial intelligence / machine learningcapabilities of the user equipment; receiving information for authorizing access to one or more artificial intelligence / machine learning services of the user equipment; receiving a request for one of the one or more artificial intelligence / machine learning services supported by the user equipment; and providing the one of the one or more artificial intelligence / machine learning services which have been requested.

[0040] The information for authorizing access to one or more artificial intelligence / machine learning services of the user equipment comprises one or more of: a token verification policy; and an authorization policy.

[0041] The method may comprise using the information for authorizing access to validate the request.

[0042] The method may comprise receiving an access token associated with the request for the one of the one or more artificial intelligence / machine learning services and using the information for authorizing access and said access token to validate the request.

[0043] The artificial intelligence / machine learning service may comprise one or more of: training of an artificial intelligence / machine learning model; providing training data for an artificial intelligence / machine learning model; or providing inference for an artificial intelligence / machine learning model.

[0044] The information relating to one or more artificial intelligence / machine learning capabilities may comprises information about one or more artificial intelligence / machine learning roles supported by the user equipment.

[0045] Each artificial intelligence / machine learning role may be is associated with respective set of information.

[0046] One or more of the artificial intelligence / machine learning roles is associated with one or more different profiles, each profile being associated with a respective set of information.

[0047] The one or more one or more artificial intelligence / machine learning roles supported by the user equipment may comprise one or more of: a role in which the user equipment is capable of training a machine learning model; a role in which the user equipment is capable of producing data for use in the training of a machine learning model; a role in which the user equipment is capable of being a participant in a distributed artificial intelligence / machine learning service; or a role in which the user equipment is capable ofgenerating inferences for an artificial intelligence / machine learning model.

[0048] The information relating to one or more artificial intelligence / machine learning capabilities may comprise information about one or more analytic types supported by the user equipment for one or more of the one or more artificial intelligence / machine learning roles.

[0049] The information relating to one or more artificial intelligence / machine learning capabilities may comprise one or more of: information about one or more consumers permitted to use one or more artificial intelligence / machine learning services supported by the user equipment; information about which information the one or more consumers are permitted to consume; or information about which information the one or more consumers are not permitted to consume.

[0050] The information which the one or more consumers are permitted to consume or not permitted to consume may relate to one or more of: location of the user equipment; tracking area identity; privacy information; or one or more security mechanisms which are supported.

[0051] The information relating to one or more artificial intelligence / machine learning capabilities may comprise one or more of: information about a manufacturer identity associated with the user equipment; information about one or more datasets available at the user equipment; or information about a one or more datasets which the user equipment is capable of providing.

[0052] The information relating to one or more artificial intelligence / machine learning capabilities may comprise one or more of: information indicating that end to end encryption is supported; or information indicating that end to end encryption is not supported.

[0053] The method may be performed by an apparatus.

[0054] The apparatus may be a user equipment or provided in a user equipment.

[0055] The apparatus may comprise one or more means for performing any of the methods of the sixth aspect.

[0056] The apparatus may comprise at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to provide one or more of the methods of the sixth aspect.

[0057] According to a seventh aspect, there is provided an apparatus comprising means for: receiving a request from a consumer to discover one or more user equipment capable of providing one or more artificial intelligence / machine learning services; obtaining information relating to one or more artificial intelligence / machine learning capabilities of one or more user equipment; and using the obtained information relating to one or more artificial intelligence / machine learning capabilities of one or more user equipment to authorize the request with respect to one or more user equipment.

[0058] The apparatus may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function.

[0059] According to an eighth aspect, there is provided a method comprising: receiving a request from a consumer to discover one or more user equipment capable of providing one or more artificial intelligence / machine learning services; obtaining information relating to one or more artificial intelligence / machine learning capabilities of one or more user equipment; and using the obtained information relating to one or more artificial intelligence / machine learning capabilities of one or more user equipment to authorize the request with respect to one or more user equipment.

[0060] The method may comprise obtaining information about the consumer and further using the obtained information about the consumer to verify the request.

[0061] The method may comprise sending to one or more of the user equipment for which the request has been authorized information for authorizing access by the consumer to one or more artificial intelligence / machine learning services of the user equipment.

[0062] The information for authorizing access to one or more artificial intelligence / machine learning services of the user equipment may comprise one or more of: a token verification policy; and an authorization policy.

[0063] The method may comprise providing to the consumer a list of one or more user equipment from which the consumer can request one or more artificial intelligence / machine learning services.

[0064] The artificial intelligence / machine learning service may comprise one or more of: training of an artificial intelligence / machine learning model; providing training data for an artificial intelligence / machine learning model; or providing inference for an artificial intelligence / machine learning model.

[0065] The information relating to one or more artificial intelligence / machine learning capabilities may comprises information about one or more artificial intelligence / machine learning roles supported by the user equipment.

[0066] Each artificial intelligence / machine learning role may be is associated with respective set of information.

[0067] One or more of the artificial intelligence / machine learning roles is associated with one or more different profiles, each profile being associated with a respective set of information.

[0068] The one or more one or more artificial intelligence / machine learning roles supported by the user equipment may comprise one or more of: a role in which the user equipment is capable of training a machine learning model; a role in which the user equipment is capable of producing data for use in the training of a machine learning model; a role in which the user equipment is capable of being a participant in a distributed artificial intelligence / machine learning service; or a role in which the user equipment is capable of generating inferences for an artificial intelligence / machine learning model.

[0069] The information relating to one or more artificial intelligence / machine learning capabilities may comprise information about one or more analytic types supported by the user equipment for one or more of the one or more artificial intelligence / machine learning roles.

[0070] The information relating to one or more artificial intelligence / machine learning capabilities may comprise one or more of: information about one or more consumers permitted to use one or more artificial intelligence / machine learning services supported by the user equipment; information about which information the one or more consumers are permitted to consume; or information about which information the one or more consumers are not permitted to consume.

[0071] The information which the one or more consumers are permitted to consume or not permitted to consume may relate to one or more of: location of the user equipment; tracking area identity; privacy information; or one or more security mechanisms which are supported.

[0072] The information relating to one or more artificial intelligence / machine learning capabilities may comprise one or more of: information about a manufactureridentity associated with the user equipment; information about one or more datasets available at the user equipment; or information about a one or more datasets which the user equipment is capable of providing.

[0073] The information relating to support by the user equipment for artificial intelligence / machine learning services may comprise one or more of: information indicating that end to end encryption is supported; or information indicating that end to end encryption is not supported.

[0074] The method may be performed by an apparatus.

[0075] The apparatus may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function.

[0076] The apparatus may comprise one or more means to perform any of the method of the eighth aspect.

[0077] The apparatus may comprise at least one processor; and at least one memory storing instructions that, when executed by the at least one processor, cause the apparatus at least to provide one or more of the methods of the eight aspect.

[0078] According to another aspect, there is provided a computer readable medium comprising program instructions stored thereon for performing at least one of the above methods.

[0079] According to an aspect, there is provided a non-transitory computer readable medium comprising program instructions stored thereon for performing at least one of the above methods.

[0080] According to an aspect, there is provided a non-volatile tangible memory medium comprising program instructions stored thereon for performing at least one of the above methods.

[0081] In the above, many different aspects have been described. It should be appreciated that further aspects may be provided by the combination of any two or more of the aspects described above.

[0082] Various other aspects are also described in the following detailed description and in the attached claims.

[0083] Brief Description of the Drawings

[0084] Example embodiments will become more fully understood from the detaileddescription given herein below and the accompanying drawings, which are given by way of illustration only and thus are not limiting of this disclosure.

[0085] Fig. 1 shows an example of a communication network to which examples disclosed herein may be applied.

[0086] FIG. 2 shows a schematic representation of a communication system for example a 5G communication system (5GS).

[0087] Fig. 3 shows an example of a first procedure;

[0088] Fig. 4 shows, an example of a second procedure;

[0089] Fig. 5 show an example of a first method;

[0090] Fig. 6 shows an example of a second method;

[0091] Fig. 7 shows an example of a third method;

[0092] Fig. 8 shows an example of a fourth method; and

[0093] Fig.9 shows an example of an apparatus.

[0094] Description

[0095] The following embodiments are exemplary. Although the specification may refer to “an”, “one”, or “some” embodiment(s) in several locations of the text, this does not necessarily mean that each reference is made to the same embodiment s), or that a particular feature only applies to a single embodiment. Single features of different embodiments may also be combined to provide other embodiments. Further, when a particular feature, structure, or characteristic is described in connection of an embodiment, it is within the knowledge of one skilled in the art to apply such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described. It shall be understood that although the terms “first,” “second” and the like may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another.

[0096] For the purposes of the present disclosure, the phrases “at least one of A or B”, “at least one of A and B”, and “A and / or B” means (A), (B), or (A and B). For the purposes of the present disclosure, the phrase “A, B, and / or C” means (A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C).

[0097] Embodiments described may be implemented in a communication network, such as any of the following radio access technologies (RATs): Worldwide Interoperabilityfor Micro-wave Access (WiMAX), Global System for Mobile communications (GSM, 2G), GSM EDGE radio access Network (GERAN), General Packet Radio Service (GRPS), Universal Mobile Telecommunication System (UMTS, 3G) based on basic wideband-code division multiple access (W-CDMA), high-speed packet access (HSPA), Long Term Evolution (LTE), LTE-Advanced, and enhanced LTE (eLTE), 5G (also called NR), or any future RAT such as 6G. Moreover, communication within the communication network may utilize any proper wireless communication technology, comprising but not limited to: Code Division Multiple Access (CDMA), Frequency Division Multiple Access (FDMA), Time Division Multiple Access (TDMA), Frequency Division Duplex (FDD), Time Division Duplex (TDD), Multiple-Input Multiple-Output (MIMO), Orthogonal Frequency Division Multiple (OFDM), and / or Discrete Fourier Transform spread OFDM (DFT-s-OFDM).

[0098] As used herein, the term “network device” or “network node” refers to a node in a communication network via which user equipment may access the network and / or which is capable of controlling radio communication and managing radio resources within a cell. The network node or network device may be referred to as a base station (BS), an access point (AP) or an access node. The network device may be, depending on the applied technology, for example, a node B (NodeB or NB), an evolved NodeB (eNodeB or eNB), an NR NB (also referred to as a gNB), a Remote Radio Unit (RRU), a radio head (RH), a remote radio head (RRH), a relay, an Integrated Access and Backhaul (IAB) node, a low power node, a non-terrestrial network (NTN) or non-ground network device such as a satellite network device, a low earth orbit (LEO) satellite and a geosynchronous earth orbit (GEO) satellite, or an aircraft network device.

[0099] Moreover, in connection of split radio access network (RAN), the network device may refer to a centralised unit (CU) of a base station and / or a distributed unit (DU) of a base station. An interface between CU and DU may be referred to as an Fl interface in NR. In the split RAN architecture, node operations may be carried out, at least partly, in the central / centralized unit, CU, (e.g. server, host or node) operationally coupled to the DU, (e.g. a radio head / node). One CU may control one or more DUs, acting at least as transmit / receive (Tx / Rx) nodes. In some embodiments, the DUs may comprise e.g. a radio link control (RLC), medium access control (MAC) layer and a physical (PHY) layer, whereas the CU may comprise the layers above RLC layer, such as a packet data convergence protocol(PDCP) layer, a radio resource control (RRC) and an internet protocol (IP) layers. Other functional splits are possible too. In practice, any processing task may be performed in either the CU or the DU and the boundary where the responsibility is shifted between the CU and the DU may depend on the applied implementation.

[0100] The term “terminal device” refers to any end device that may be capable of wireless communication. By way of example, a terminal device may be referred to as a communication device, user equipment (UE), a Subscriber Station (SS), or a Mobile Station (MS). The terminal device may include a mobile phone, a cellular phone, a smart phone, voice over IP (VoIP) phones, wireless local loop phones a tablet, a wearable terminal device, a personal digital assistant (PDA), portable computers, desktop computer, image capture terminal devices such as digital cameras, gaming terminal devices, music storage and playback appliances, vehicle-mounted wireless terminal devices, USB dongles, an Internet of Things (loT) device, a watch or other wearable, a head-mounted display (HMD), a vehicle, a drone, a medical device and applications (e.g., remote surgery), an industrial device and applications (e.g., a robot and / or other wireless devices operating in an industrial and / or an automated processing chain contexts), a consumer electronics device, a device operating on commercial and / or industrial wireless networks, and the like.

[0101] Fig. 1 illustrates an example of a communication network to which examples disclosed herein may be applied. The communication network or a cellular communication network may comprise a network node 110 providing one or more cells, such as cell 100, and a network node 112 providing one or more other cells, such as cell 102. Each cell may be, e.g., a macro cell, a micro cell, femto, or a pico cell, for example. The cell may define a coverage area or a service area of the corresponding access node.

[0102] The network node 110 may provide a user equipment (UE) 120 (one or more UEs) with wireless access to the communication network. The wireless access may comprise downlink (DL) communication from the network node to the UE 120 and uplink (UL) communication from the UE 120 to the network node.

[0103] There may be a plurality of UEs 120, 122 in the system. Each of them may be served by the same or by different network nodes 110, 112. UE may be configured with dual connectivity (DC), wherein the UE, e.g. UE 120, may be connected to multiple network nodes 110, 112. The UEs 120, 122 may communicate with each other, in case device-to-device (D2D) communication interface is established between them via a so-called sidelink (SL). Such D2D communications may be referred to as machine-to-machine, peer-to-peer (P2P) communications, or vehicle-to-vehicle (V2V), for example.

[0104] In the case of multiple network nodes in the communication network, the network nodes may be connected to each other via an interface. LTE specifications call such an interface as X2 interface. An interface between an LTE node and a 5G node, or between two 5G nodes may be called Xn interface.

[0105] The network nodes 110 and 112 may be further connected via another interface to a core network 116 of the communication network.

[0106] In the following various examples are explained with reference to communication devices capable of communication with a communication system. Before explaining in detail the various examples of this disclosure, a 5th generation communication system (5GS), an access network and a core network (5GC) thereof, and communication devices are briefly explained with reference to FIG. 2.

[0107] FIG. 2 shows a schematic representation of a communication system for example a 5G communication system (5GS). The 5GS may comprise a user equipment (UE) or Terminal 120, a 5G core network 202 (which may be an example of core network 116 described previously), and one or more application functions 203. An application function 203 may be deployed in the 5GS as trusted application function or may be deployed or host on one or more application servers of the data network (DN) 204. Such application functions may be untrusted application functions. The 5GS connects the UE to a data network, the access network, and the 5GC 202 (e.g., a UPF of the 5GC).

[0108] The 5GC may comprise the following network functions: Network Slice Selection Function (NSSF); Network Exposure Function (NEF) 205; Network Repository Function (NRF); Policy Control Function (PCF); Unified Data Management (UDM) 206; Application Function (AF) 203; Authentication Server Function (AUSF) 207; an Access and Mobility Management Function (AMF) 208; Session Management Function (SMF) 209; and a user plane function (UPF) 210.

[0109] FIG. 2 also shows the various interfaces (Nl, N2 etc.) that may be implemented between the various elements of the system. It should be understood that not all of the above network functions are shown in FIG. 2, and that in some examples the 5GCmay comprise additional network functions other than those mentioned above.

[0110] The AMF 208 may handle termination of non-access stratum (NAS) signalling, NAS ciphering & integrity protection, registration management, connection management, mobility management, access authentication and authorization, security context management. The UPF 210 may support packet routing and forwarding, packet inspection and quality of service (QoS) handling, for example.

[0111] Artificial intelligence (Al) can be broadly defined as getting computers to perform tasks mimicking human brain. Machine learning (ML) is one category of Al techniques: computer algorithms able to automatically improve their performance without explicit programming. Al algorithms were first conceived in the 1950’s but only in recent years AI / MLML has become useful in vast area of real-world applications, partly due to advancements in computational power and in providing storage capacity for data.

[0112] AI / ML can help adjust and optimize radio access network (RAN) parameters and settings using (real-time) monitoring and prediction of network performance, quality, and demand. Additionally, AI / ML can identify and diagnose degradation in network performance, as well as provide protection from cyberattacks. AI / ML is usable in energy saving, load balancing, mobility optimization, link adaptation and security just to mention but a few.

[0113] It is envisioned that AI / ML will enable real-time analysis as well as automated operation and control in 5G and beyond RAN. This requires the availability of data streamed from wireless devices in a timely manner, especially in extremely time-critical applications such as real-time video monitoring and extended reality (XR). This may be reflected in network architecture, such as by placing and moving ML agents to the required locations in the network, for example for data collection. User devices (mobile devices) may assist network in decision-making in resource management, thus a user device may act as an infrastructure resource.

[0114] As the network evolves to programmable and flexible cloud native implementation, AI / ML-based network automation will be used to simplify network management and optimization. It is expected that parts of the air interface, in particular signal processing algorithms, are supported and eventually even replaced with machine learning models. Thus, a 6G wireless communication standard will natively support an Al-based airinterface.

[0115] Machine learning algorithms are usually classified into four different types: supervised learning, unsupervised learning, semi-supervised learning and reinforcement learning.

[0116] In supervised learning the algorithm learns from labelled data. For training, the algorithm receives input data and corresponding correct output labels. The algorithm is trained to predict accurate labels for new data.

[0117] In unsupervised learning the algorithm analyses unlabelled data. The aim is to discover patterns, relationships, or structures within the data, for example, unsupervised learning algorithms make groups of similar data points.

[0118] Semi-supervised learning is a hybrid machine learning approach that combines labelled and unlabelled data for training. A limited amount of labelled data and a larger set of unlabelled data is used to improve training. This approach is useful when acquiring labelled data is expensive or time-consuming as is the case in many real-world applications. Semi-supervised learning techniques can be applied to various tasks, such as classification, regression, and anomaly detection, allowing models to make more accurate predictions and generalize better in real-world scenarios.

[0119] Reinforcement learning is a machine learning algorithm which learns from trial and error. An ML agent interacts with environment and learns from experience aiming to maximize cumulative rewards. The ML agent receives feedback through rewards or penalties based on its actions. The agent learns to take actions that lead to the most favourable outcomes over time. The algorithm adapts to changing environments and achieves long-term goals through a sequence of actions.

[0120] An example of an ML algorithm found applicable to adjust and optimize the radio access network (RAN) parameters and settings is deep learning. Deep learning is a subset of machine learning algorithms using a neural network. Neural networks are also known as artificial neural networks (ANNs) or simulated neural networks (SNNs). Deep learning can be based on supervised, semi-supervised or unsupervised learning.

[0121] Artificial neural networks (ANNs) are comprised of an input layer, one or more hidden layers, and an output layer. Each node of a layer, or an artificial neuron, connects to another one and has an associated weight as well as a threshold value. If theoutput of an individual node is above the threshold value specified to this node, the node is activated and sending or passing data to the next layer of the neural network.

[0122] In the case the supervised learning is applied in the training of a neural network, the training is carried out by using examples, each of which contains a known "input" and "result", forming probability-weighted associations between them. The training comprises determining the difference between the output of the neural network (a prediction) for an input and a target output for the same input. The difference is called an error value. The neural network then adjusts its weighted associations according to a learning rule and using this error value. Successive adjustments make the neural network produce output that is approaching the target output. After a sufficient number of these adjustments, the training can be terminated based on a certain criteria.

[0123] Some embodiments may address one or more security issues which may arise when UE(s) are involved in AI / ML service operations.

[0124] Consider a first example where the UE is a service producer. The UE may for example be an AI / ML model training participant in the case of distributed learning, e.g. federated learning (FL). In another example, the UE is a data producing entity.

[0125] Some embodiments may provide a procedure by which consumers (such as network functions NF or application functions AF) may be authorized by a 5G or a 6G core to discover which UE(s) can provide a required AI / ML service.

[0126] Currently a discovery request to find a relevant UE service provider based upon the requested AI / ML operation is not authorized. This may lead to sharing, without authorization, unnecessary details regarding the service producers (e.g. the UEs) to potentially malicious entities. This may lead to an increase in the risk of denial-of-service attacks, since, in principle, every discovery request may be taken as a valid request and consequently accepted.

[0127] With some embodiments, providing some authorization during the discovery itself may be advantageous. If an AI / ML service consumer is deemed unauthorized for the discovery process, an unnecessary process of for example access token generation, and verification can be avoided.

[0128] Consider a second example where the UE is a service consumer requesting model training services from the core network. For example, one 6G Al ML use cases maybe for the provision of a ML model training offloading service to the UE. In one example, the UE may request the core network to train the ML model on behalf of the UE using the training data of the UE. The trained ML model may be sent back by the core network to the UE. Such a scenario may be supported by some embodiments. In some embodiments, the UE is authorized to request such a service from the core network. In some embodiments, the core network after having trained the ML model may ensure that the ML model is sent to a genuine UE.

[0129] In some embodiments, a procedure may be provided to allow one or more network entities to access UE resources where the UE provides an AI / ML service to the respective network entities. The respective network entity may be referred to as a consumer. The respective network entity wanting access to the UE resource may be a NF or AF.

[0130] In some embodiments, an authorization policy for a UE may be defined. In some embodiments, the UE may register AI / ML related UE information in the UDM (via NAS). The AI / ML related information may comprise information about one or more AI / ML related capabilities of the UE.

[0131] The authorization policy may be used by one or more network entities. The one or more network entities may comprise one or more of UDM, NF, AF or NEF The authorization policy may be used by one or more network entities to take an authorization decision before selecting and / or discovering the UE for the consuming NF or AF. The policy may be used for the granting authorization for one or more network entities to consume an AI / ML service provided by the UE.

[0132] The term AI / ML may refer to Al only, ML only, or Al and ML.

[0133] Reference is made to Figure 3 which shows a first example where a UE is providing one or more AI / ML services. The one or more AI / ML services are consumed by a consumer. The consumer may comprise one or more network entities.

[0134] The procedure of Figure 3 may provide authorization (including discovery and service authorization) of NF(s) and AF(s) which need to discover and request services from the UE(s) providing AI / ML services.

[0135] Reference is made to the part of the procedure referenced la and lb.

[0136] As referenced la, the UE sends a message comprising information about one or more AI / ML capabilities of the user equipment. This message may be an NAS message.The message may be sent to the AMF.

[0137] As referenced lb, the AMF may provide the information about the one or more AI / ML capabilities of the user equipment to the UDM. The information may be provided in a registration message. The information may be provided via a UECM (UE context management) service supported by the UDM. The service API may be, for example, a Nudm_UECM_Regi strati on.

[0138] The information may comprise information about one or more AI / ML services or operations provided by the UE. The one or more AI / ML services or operations may comprise one or more of: providing training data for an AI / ML model (for example the UE is a data producer); providing inference for an AI / ML model; or training of an AI / ML model.

[0139] The information about the one or more AI / ML capabilities of the user equipment may comprise information about one or more AI / ML roles supported by the UE. The one or more AI / ML roles may comprise one or more of the following roles: a role in which the user equipment is capable of training a machine learning model; a role in which the user equipment is capable of producing data for use in the training of a machine learning model; a role in which the user equipment is capable of being a participant in a distributed artificial intelligence / machine learning service; or a role in which the user equipment is capable of generating the inferences for an artificial intelligence / machine learning model.

[0140] For example, a UE may be capable of training a ML model and may provide a service of being a participant in a distributed AI / ML mechanism (such as FL). In another example, the UE may only provide data and acts as a data producer.

[0141] One or more of the AI / ML roles may be associated with a respective set of information.

[0142] One or more of the AI / ML roles may be associated with one or more different profiles. Each profile may be associated with a respective set of information.

[0143] The information about the one or more AI / ML capabilities of the user equipment may comprise information about one or more analytics type supported by the UE.

[0144] Information about one or more analytics type supported by the UE may be provided for different ones of the AI / ML roles. By way of example, the analytics type mayrelate to UE mobility, UE communication pattern, network load, and / or the like.

[0145] Information about one or more analytics type supported by the UE may be provided for each one of the AI / ML roles.

[0146] The information about the one or more AI / ML capabilities of the user equipment may comprise identity information relating to the UE. The identity information may comprise information about a manufacturer identity. The identity information may comprise MSISDN (mobile station international subscriber directory number), GPSI (generic public subscription identifier), SUPI (subscription permanent identifier), or IMEI (iintemational mobile equipment identity).

[0147] The information about the one or more AI / ML capabilities of the user equipment may comprise information about a serving network operator.

[0148] The information about the one or more AI / ML capabilities of the user equipment may comprise one or more of: information about one or more datasets available at the UE; or information about one or more datasets which the UE is capable of providing. These datasets may be used in the AI / ML service provided by the UE.

[0149] The information relating to one or more AI / ML capabilities may comprise information about one or more entities or consumers permitted to use one or more AI / ML services supported by the user equipment.

[0150] The information relating to one or more AI / ML capabilities may comprise information about which information of the UE the one or more entities or consumers are permitted to consume.

[0151] The information relating to one or more AI / ML capabilities may comprise information about which information of the UE the one or more entities or consumers are not permitted to consume.

[0152] The information which is and / or is not permitted to be consumed may be dependent on the AI / ML role.

[0153] The information may be information relating to the location of the user equipment. This may be information which is or is not permitted to be consumed.

[0154] The information may be information relating to the tracking area identity (TAI). This may be information which is or is not permitted to be consumed.

[0155] The information may be privacy information. This may be information whichis or is not permitted to be consumed.

[0156] The information may be information relating to one or more security mechanisms which are supported.

[0157] The information may be information indicating that end to end encryption is supported.

[0158] The information may be information indicating that end to end encryption is not supported.

[0159] For example, a consumer of a UE AI / ML service may only be allowed to consume a UE AI / ML service if the UE is associated with an allowed TAI.

[0160] For example, a consumer of a UE AI / ML service may only be allowed to consume a UE AI / ML service based on the purpose of analytics.

[0161] For example, a consumer of a UE AI / ML service may only be allowed to consume a UE AI / ML service based on privacy and / or security mechanism support.

[0162] For example, a consumer of a UE AI / ML service may only be allowed to consume a UE AI / ML service based on the NF and / or AF information. The NF and / or AF information may comprise information about an NF instance ID and / or information about the type of NF and / or AF.

[0163] By way of example, the consumer information may containAllowed AF per Region= Region X (True), Region Y(False)Allowed NF per Region= Region X (True), Region Y(False)Allowed gNB(RAN) Node per Region,Allowed NF from HPLMN Per Purpose= network optimization (True), analytics (false)

[0164] As mentioned, the AI / ML capabilities of the UE may be defined on the basis of analytics type. An analytics type and / or Al ML purpose (e.g. role) may consist of different UE profiles. Based on the analytics type and / or AI / ML purpose, the AI / ML capabilities may differ.

[0165] In some embodiments, an analytics type may be associate with an analytics ID.

[0166] As referenced 2, the UDM stores the information received from the UE. This information may provide or be used to provide an authorization policy for the UE for AI / MLrelated operations or services. The authorization policy may comprise one or more of the information relating to one or more AI / ML capabilities such as previously discussed.

[0167] This authorization policy may be used for UE discovery and AI / ML service requests. For example, an AF or AI / ML enabled network entities may use this authorization policy to discover and request services from the UE(s). The authorization policy may define the AI / ML services and / or operation available to the respective AF or AI / ML enabled network entity.

[0168] The AI / ML enabled network entities may be NFs and / or gNB(RAN).

[0169] The AI / ML services or operations may be as previously discussed. For example, a UE may provide a passive service or operation where the UE provides training data for ML model training. For example, a UE may provide an active service or operation where the UE trains the ML model (e.g. in the case of FL).

[0170] The UDM may provide one or more operator specific authorization parameters. These one or more operator specific authorization parameters may override one one or more AI / ML capabilities provided by the UE.

[0171] As referenced 3a, a gNB(RAN) sends a discovery request to a AI / ML process coordinator.

[0172] The AI / ML process coordinator may be a standalone network function. In other embodiments, the AI / ML process coordinator may be co-located with one or more other network functions. For example, the AI / ML process coordinator may be implemented as a service co-located with a NWDAF. In some embodiments, the AI / ML process coordinator may be provided as part of a network function and / or application function. In some embodiments, the AI / ML process coordinator may be provided as part of the NRF. In some embodiments, the AI / ML process coordinator may be provided by two or more functions. The AI / ML process coordinator may also be considered a central service running on dedicated NF or 0AM (operations and management) function. The AI / ML process coordinator may manage the overall AI / ML workloads running in the system.

[0173] The discovery request may be a NGAP (next generation application protocol) message or a SBA (service based architecture) message.

[0174] The discovery request may be to discover the UE(s) eligible for a required AI / ML service. The discovery request may specify one or more AI / ML capabilities requiredof the eligible UEs. For example, the UE(s) may need be able to take part in an FL (federated learning) process and be associated with one or more TAIs.

[0175] As referenced 3b, a NWDAF, an AF or any other AI / ML enabled NF which requires a UE for a particular AI / ML service may sends a discovery request to the AI / ML process coordinator to discovery the UE(s). This may be as outlined in the part of procedure referenced 3 a.

[0176] The part of the procedures referenced 3a and 3b may be regarded as two different options. One or both of these options may be provided in a given procedure. Where both of these options are provided, they may be provided generally in a same time frame.

[0177] In both cases, a consumer (for example the gNB(RAN), the NWDAF, the AF or any other AI / ML enabled NF) may be interested in a list of UEs satisfying one or more AI / ML related capabilities such as: location of the UE, TAI associated with the UE, AI / ML services provided by the UE, and / or the like. The request may comprise a list or filter of one or more requirements for a respective UE.

[0178] Alternatively, a consumer may provide the UE list itself. In this case, the request may be for an access token and not for UE discovery.

[0179] As referenced 4, the AI / ML process coordinator verifies the request.

[0180] The AI / ML process coordinator may fetch profile information about the corresponding consumer, e.g. the gNB, or the NF / AF. The NRF may be queried for NF or gNB information. The NEF may be queried for AF information. The NRF and NEF may be regarded as registration entities.

[0181] The AI / ML process coordinator may fetch profile information about one or more UEs from the UDM. The profile information which is fetched is based on the list or filter of one or more requirements for the UE requested by the consumer. The UDM may thus be queried for UE information. The UDM may be regarded as a registration entity.

[0182] In some embodiments, the AI / ML process coordinator may be an authorization server. The AI / ML process coordinator may query the UDM for the UE information, the NRF for the NF and / or gNB information, or the NEF for the AF information.

[0183] The AI / ML may obtain the authorization policy for a given UE from the UDM.

[0184] As referenced 5, the AI / ML coordinator may authorize the request after analysing the properties of the consumer against the UE properties (for example as defined by the authorization policy).

[0185] If the consumer is authorized to discover and request the desired AI / ML service from a specific set of one or more UE(s), a positive response is sent to the consumer. Authorization may be guaranteed in the event of a positive response, i.e., only authorized UEs will be added to a list of UEs participants in the AI / ML operations or services.

[0186] If no UE is selected, then a negative response will be sent to the consumer. The following flow assumes that a positive response is to be sent.

[0187] As referenced 6, the Al ML process coordinator sends the list of UE(s) from which AI / ML operations or services can be requested to the consumer. This is the positive response. The part of the procedure referenced 6a is where the consumer is a gNB (RAN) and the part of the procedure referenced 6b is where the consumer is a NF or an AF.

[0188] As referenced 7, the consumer (NF, AF or gNB) further sends an access token request. The access token request may be sent to the AI / ML process coordinator (where the AI / ML process coordinator is part of the NRF) or to the NRF.

[0189] The access token request may comprise information about one or more of: the list of UE(s); specific UE(s) for which the AI / ML service is requested; a UE region for which the AI / ML service is requested; a AI / ML role of target producers (that is the UE); an analytics ID; dataset information (this may be where the UE is a data provider); a consumer instance; NFc type (e.g. AMF, SMF, NRF and / or the like), NFc TAI; or consumer AI / ML role.

[0190] Where the access token is generated by the NRF and the NRF does not have the respective UE information (e.g. authorization policy), the NRF may request the UDM to provide this UE information.

[0191] Based upon the UE information from the UDM (e.g. authorization profile) and the consumer profile information, the NRF verifies if the consumer (e.g. AF, NF or gNB) is authorized.

[0192] If the consumer is authorized, the NRF generates an access token. This may comprise one of more of the following information: one or more target UEs; an AI / ML role to be performed by a target UE; adataset identifier; an event identifier, a region of the UE; an AI / ML role of the consumer; or a region of the consumer.

[0193] The access token is then sent by the NRF to the corresponding consumer.

[0194] The UE is configured with the NRF public key to verify the signature and access token.

[0195] As referenced 8a, the AI / ML process coordinator provides to the AMF a token verification policy and / or the authorization policy.

[0196] As referenced 8b, the AMF provides to the UE the token verification policy and / or the authorization policy. Alternatively or additionally, the AMF may provide list of one or more authorized consumers. The IP address or the fully qualified domain name FQDN of the authorized consumers may be provided.

[0197] As referenced 9, a request for the AI / ML service is sent from the consumer to the UE. The part of the procedure referenced 9b is where the consumer is a gNB (RAN) and the part of the procedure referenced 9a is where the consumer is a NF or an AF.

[0198] As referenced 9c, the UE can validate the request. The UE may validate the request using the authorization policy or the access token. The validation of the request may be based on the list of authorized consumers.

[0199] When a request is validated, the UE may provide the requested AI / ML service or operation.

[0200] Some embodiments may relate to the authorization of UE to access core network resources to provide an AI / ML service for the UE.

[0201] In some embodiments, the authorization requirements both for the discovery and the providing of the service may be such that the core network or AF ensures that the UE is authorized to discover and request the AI / ML services provided by the core network or AF. For example, the service may be an ML model training service.

[0202] Reference is made to Figure 4 which shows an example where a UE requests core network resources to provide an AI / ML service for the UE. For example, the UE may request that the network resources train an AI / ML model for the UE.

[0203] As referenced la, the UE sends a message comprising information about one or more AI / ML capabilities of the user equipment. This message may be an NAS message. The message may be sent to the AMF.

[0204] As referenced lb, the AMF may provide the information about the one or more AI / ML capabilities of the user equipment to the UDM. The information may be provided in a registration message. The information may be provided in a UECM registration message, for example a Nudm_UECM_Regi strati on.

[0205] This information may comprise any of the information about the one or more AI / ML capabilities discussed previously.

[0206] In some embodiments, the AI / ML capability information may comprise information about one or more subscriptions which the UE has for AI / ML services such as AI / ML training services.

[0207] As referenced 2, the AI / ML service providing entity registers its AI / ML related capabilities. The AI / ML service providing entity may be a NWDAF, an AF, or any Native Al NF.

[0208] The AI / ML related capabilities may comprise one or more of:One or more AI / ML roles provided by the AI / ML service providing entity; vendor or manufacturer identity; interoperability indicator (e.g. vendor interoperability between the UE and AI / ML service providing entity); analytic type; type of training provided; information that end to end encryption is supported; information that end to end encryption is not supported; allowed consumers; blocked consumers; information defining which consumers are allowed to discover the AI / ML service providing entity; or information defining which consumers are not allowed to discover the AI / ML service providing entity.

[0209] One or more AI / ML roles provided by the AI / ML service providing entity may comprise one or more of: model training server; FL client / server; or training data producer.

[0210] Consumers allowed to discover the AI / ML service providing entity may be defined by one or more of: UE Manufacturer ID; vendor ID, UE identity, or TAI of the consumer.

[0211] As referenced 3a and 3b, the UE sends a message to the NRF or AI / ML process coordinator. The message is sent via the AMF. The message may be an NAS message. The message may be a request for discovery of the NF or AF providing the specific AI / ML service (for instance Al ML model training service) that the UE requires. The message may alternatively or additionally be a request for authorization to access to thosespecific services. For example, the request authorization to access to those specific services may be a request for an access token.

[0212] The AI / ML process coordinator may be as previously described.

[0213] The message may comprise information about one or more of an identifier of the UE, an analytics ID identifying an analytics type for which training is required; or privacy requirements for that analytics. For instance, data might need to be already anonymized before sharing, or special privacy mechanisms should be performed on data before training, and / or the like.

[0214] As referenced 4, the NRF or AI / ML process coordinator (depending on the entity which receives the request) verifies the request.

[0215] The NRF or AI / ML process coordinator may fetch one or more of the subscription information about the corresponding UE, for example by sending a request to the UDM; or the AIML profile information of the NF / AF from the NRF which can serve the request, for example by providing AI / ML model training.

[0216] As referenced 5, in the case of successful verification, the NRF generates an access token. The access token may indicate the AI / ML service (for example the AI / ML model training service) the UE is authorized for. The access token may comprise information about one or more of attributes for the UE to be authorised to access the AI / ML service: an allowed UE manufacturer; allowed vendor information associated with the UE; an allowed assigned identity; an allowed location; or the like.

[0217] As referenced 6, this access token claim is sent to UE via AMF. The access token may be sent via an NAS message.

[0218] As referenced 7, the UE send a new AI / ML model training request to the AI / ML service providing entity. The request may comprise the access token. The request may comprise an analytics identity, and / or a corresponding MDT (minimization of drive test) file (in the case the training data is already not collected).

[0219] As referenced 8, in case of successful access token verification by the AI / ML service providing entity, the AI / ML model training is initiated. When the AI / ML model is trained, the AI / ML service providing entity may provide a URI (uniform resource identifier) of the trained AI / ML model to be downloaded by the UE.1

[0220] In one modification to the above-described procedure, no access token is used. In this example, after the part of the procedure referenced 4, in which the AI / ML process coordinator verifies that the UE is indeed authorized to request a particular service, details of the UE along with the requested and authorized service are pushed or otherwise provided to the corresponding AI / ML service providing entity which provides the requested service. The part of the procedure referenced 7 may be modified such that the AI / ML service providing entity is able to verify if the UE sending the request is indeed the same UE which has been authorized and whose information the AI / ML service providing entity has received. In the case of successful verification, the service is then provided to the UE.

[0221] Reference is made to FIGs. 5 to 8 which shows some methods of some example embodiments.

[0222] The respective methods may be performed by an apparatus.

[0223] The apparatus may comprise suitable means, such as circuitry for providing the method.

[0224] Alternatively or additionally, the apparatus may comprise at least one processor and at least one memory storing instructions that, when executed by the at least one processor cause the apparatus at least to provide the method below.

[0225] Alternatively or additionally, the apparatus may be such as discussed in relation to FIG. 9.

[0226] The respective methods may be provided by computer program code or computer executable instructions.

[0227] One or more of the methods may be modified to one include one or more steps such as previously described.

[0228] Reference is made to Figure 5. The apparatus may be or provided in a user equipment.

[0229] The method comprises as referenced Al, providing, to a network, information relating to one or more artificial intelligence / machine learning capabilities of the user equipment.

[0230] The method comprises as referenced A2, receiving information for authorizing access to one or more artificial intelligence / machine learning services of the user equipment.

[0231] The method comprises as referenced A3, receiving a request for one of the one or more artificial intelligence / machine learning services supported by the user equipment.

[0232] The method comprises as referenced A4, providing the one of the one or more artificial intelligence / machine learning services which have been requested.

[0233] Reference is made to Figure 6. The apparatus may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function.

[0234] The method comprises as referenced Bl, receiving a request from a consumer to discover one or more user equipment capable of providing one or more artificial intelligence / machine learning services.

[0235] The method comprises as referenced B2, obtaining information relating to one or more artificial intelligence / machine learning capabilities of one or more user equipment.

[0236] The method comprises as referenced B3, using the obtained information relating to one or more artificial intelligence / machine learning capabilities of one or more user equipment to authorize the request with respect to one or more user equipment.

[0237] Reference is made to Figure 7. The apparatus may be or provided in a user equipment.

[0238] The method comprises as referenced Cl, sending a request for discovery of one or more network entities to provide one or more artificial intelligence / machine learning services for the user equipment.

[0239] The method comprises as referenced C2, receiving a response to the request for discovery, when the request has been verified.

[0240] The method comprises as referenced C3, in response to receiving the response to the request for discovery, sending a request for the providing of one or more artificial intelligence / machine learning model services for the user equipment.

[0241] Reference is made to Figure 8. The apparatus may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function.

[0242] The method comprises as referenced DI, receiving a request from a userequipment to discover one or more network entities capable of providing one or more artificial intelligence / machine learning services to the user equipment.

[0243] The method comprises as referenced D2, verifying the request using information about one or more network entities which are able to provide the one or more artificial intelligence / machine learning services to the user equipment.

[0244] The method comprises as referenced D3 sending to the user equipment, a response to the request for discovery, when the request has been verified.

[0245] Fig. 9 shows, by way of example, a block diagram of an apparatus 10. The apparatus 10 comprises, for example, at least one processor 12 and at least one memory 14 storing instructions 15 that, when executed by the at least one processor, cause the apparatus 10 at least to perform the method or methods as disclosed herein, and any of the embodiments thereof. In an example, the at least one memory and the instructions (e.g. a computer program code, software), are configured, with the at least one processor, to cause the apparatus 10 to perform the method or methods as disclosed herein, and any of the embodiments thereof.

[0246] A processor 12 may comprise circuitry, or be constituted as circuitry or circuitries, the circuitry or circuitries being configured to perform phases of methods in accordance with example embodiments described herein. As used in this application, the term “circuitry” may refer to one or more or all of the following: (a) hardware-only circuit implementations, such as implementations in only analog and / or digital circuitry, and (b) combinations of hardware circuits and software, such as, as applicable: (i) a combination of analog and / or digital hardware circuit(s) with software / firmware and (ii) any portions of hardware processor(s) with software (including digital signal processor(s)), software, and memory(ies) that work together to cause an apparatus, such as a user equipment, to perform various functions) and (c) hardware circuit(s) and or processor(s), such as a microprocessor(s) or a portion of a microprocessor(s), that requires software (e.g., firmware) for operation, but the software may not be present when it is not needed for operation. This definition of circuitry applies to all uses of this term in this application, including in any claims. As a further example, as used in this application, the term circuitry also covers an implementation of merely a hardware circuit or processor (or multiple processors) or portion of a hardware circuit or processor and its (or their) accompanying software and / or firmware.The term circuitry also covers, for example and if applicable to the particular claim element, a baseband integrated circuit or processor integrated circuit for a mobile device or a similar integrated circuit in server, a cellular network device, or other computing or network device.

[0247] The memory 14 may be implemented using any suitable data storage technology. The memory may comprise a database for storing data. The memory 14 may be at least in part external to apparatus 10 but accessible to apparatus 10.

[0248] The instructions 15 may be comprised in a computer readable medium or a non-transitory computer readable medium. A term non-transitory, as used herein, is a limitation of the medium itself (i.e. tangible, not a signal) as opposed to a limitation on data storage persistency (e.g. random access memory, RAM, vs. read only memory, ROM).

[0249] For example, the apparatus 10 is a terminal device, such as the UE of Fig. 1 or 2. As another example, the apparatus is comprised in such a terminal device, e.g. as a chipset configured to control the terminal device. The apparatus 10 may be caused or configured to perform at least the method of Fig. 5 or 7 and / or any one or more of the embodiments described.

[0250] As another example, the apparatus 10 may provide an artificial intelligence / machine learning coordination function. The apparatus may be provided by a network function. The t apparatus may comprise a chipset . The apparatus 10 may be caused or configured to perform at least the method of Fig. 6 or 8 and / or any one or more of the embodiments described.

[0251] The apparatus 10 may optionally comprise a radio interface 16. The radio interface 16 may provide the apparatus 10 with communication capabilities. The radio interface 16 may comprise a receiver configured to receive information in accordance with at least one cellular or non-cellular standard. The radio interface 16 may comprise a transmitter configured to transmit information in accordance with at least one cellular or non-cellular standard. The receiver may comprise more than one receiver. The transmitter may comprise more than one transmitter. The radio interface 16 may comprise a transceiver configured to receive and transmit information in accordance with at least one cellular or non-cellular standard. The transceiver may comprise more than one transceiver.

[0252] The apparatus 10 may optionally comprise a user interface 18 comprising, for example, at least one of a keypad, a microphone, a touch display, a display, a speaker, etc.The user interface 18 may be used to control the apparatus by the user. The user interface 18 may be external to the apparatus 10. For example, the apparatus 10 may be connected to another device, such as a computer, either via wireless or wired connection, and the apparatus 10 is controlled by the user via the computer.

[0253] In an embodiment, at least some of the processes described herein may be carried out by an apparatus comprising means for carrying out at least some of the described processes. Means for performing method steps as disclosed herein may include software and / or hardware components of the apparatus 10. For example, the at least one processor 12, the memory 14, and the computer program code form means for carrying out the method or methods as disclosed herein, and any of the embodiments thereof. As used herein the term “means” is to be construed in singular form, i.e. referring to a single element, or in plural form, i.e. referring to a combination of single elements. Therefore, terminology “means for [performing A, B, C]”, is to be interpreted to cover an apparatus in which there is only one means for performing A, B and C, or where there are separate means for performing A, B and C, or partially or fully overlapping means for performing A, B, C. Further, terminology “means for performing A, means for performing B, means for performing C” is to be interpreted to cover an apparatus in which there is only one means for performing A, B and C, or where there are separate means for performing A, B and C, or partially or fully overlapping means for performing A, B, C.

[0254] Even though the invention has been described above with reference to an example according to the accompanying drawings, it is clear that the invention is not restricted thereto but can be modified in several ways within the scope of the appended claims. Therefore, all words and expressions should be interpreted broadly, and they are intended to illustrate, not to restrict, the embodiment. It will be obvious to a person skilled in the art that, as technology advances, the inventive concept can be implemented in various ways. Further, it is clear to a person skilled in the art that the described embodiments may, but are not required to, be combined with other embodiments in various ways.

Claims

32CLAIMS1. A user equipment compri sing : means for sending a request for discovery of one or more network entities to provide one or more artificial intelligence / machine learning services for the user equipment; means for receiving a response to the request for discovery, the response indicating that the request has been verified; and means for, in response to receiving the response to the request for discovery, sending a request for the providing of one or more artificial intelligence / machine learning model services for the user equipment.

2. The user equipment as claimed in claim 1, wherein the one or more artificial intelligence / machine learning services comprises a service in which the network entity assists in the training of a machine learning model for the user equipment.

3. The user equpment as claimed in claim 1 or 2, wherein the response comprises information identifying a network entity which is able to provide the one or more artificial intelligence / machine learning services for the user equipment.

4. The user equipment as claimed in claim 3, wherein the request for the providing of the one or more artificial intelligence / machine learning services for the user equipment is sent to the network entity identified by the information identifying the network entity in the response.

5. The user equipment as claimed in any preceding claim, wherein the response to the request for discovery comprises an access token.

6. The user equipment as claimed in claim 5, wherein the request for assistance in the training of the machine learning model for the user equipment comprises the access token.

337. The user equipment as claimed in any preceding claim, wherein the response to the request comprises one or more of: information about one or more artificial intelligence / machine learning services provided by the network entity; an identifier of the user equipment; information about a manufacturer of the user equipment; location information associated with the providing of the one or more artificial intelligence / machine learning services provided by the network entity; or information about a vendor of the user equipment.

8. The user equipment as claimed in any preceding claim, wherein the request for discovery of one or more network entities comprises one or more of: information indicating that a service for training of a machine learning model is required; a user equpment identifier; an analytics type identifier; or information about one or more privacy requirements associated with data to be used in the machine learning model training.

9. The user equipment as claimed in any preceding claim, wherein the request for assistance comprises one of more of: the analytics type identifier; or data for use in the training of the machine learning model.

10. The user equipment as claimed in claim 2 or any claim appended thereto, comprising receiving information from the network entity relating to a trained machine learning model.

11. The user equipment as claimed in claim 10, wherein the information comprises information about a unform resource indicator from which trained machine learning model can be downloaded.

12. The user equipment as claimed in any preceding claim, comprising means for providing, to a network, information relating to one or more artificial intelligence / machine learning capabilities of the user equipment.

13. The user equipment as claimed in claim 12, wherein the information relating to one or more artificial intelligence / machine learning capabilities of the user equipment comprises information about one or more of: one or more datasets stored by the user equipment; one or more datasets generated by the user equipment; a manufacturer identity associated with the user equipment; a vendor identity associated with the user equipment; a serving network operator of the user equpment; end-to-end encryption support provided by the user equipment; or one or more artificial intelligence / machine learning service subscriptions.

14. An apparatus comprising means for: receiving a request from a user equipment to discover one or more network entities capable of providing one or more artificial intelligence / machine learning services to the user equipment; verifying the request using information about one or more network entities which are able to provide the one or more artificial intelligence / machine learning services to the user equipment; and sending to the user equipment, a response to the request for discovery, when the request has been verified.

15. The apparatus as claimed in claim 14, wherein the means is for obtaining information about one or more artificial intelligence / machine learning capabilities of the user equipment and further using the obtained information about the user equipment to verify the request.

16. The user equipment as claimed in claim 14 or 15, wherein the one or more artificial intelligence / machine learning services comprises a service in which the network entity assists in the training of a machine learning model for the user equipment.

17. The user equipment as claimed in any of claims 14 to 16, wherein the response comprises information identifying a network entity which is able to provide the one or more artificial intelligence / machine learning services for the user equipment.

18. The user equipment as claimed in in any of claims 14 to 17, wherein the response to the request for discovery comprises an access token.

19. A method comprising: sending a request for discovery of one or more network entities to provide one or more artificial intelligence / machine learning services for the user equipment; receiving a response to the request for discovery, when the request has been verified; and in response to receiving the response to the request for discovery, sending a request for the providing of one or more artificial intelligence / machine learning model services for the user equipment.

20. A method comprising: receiving a request from a user equipment to discover one or more network entities capable of providing one or more artificial intelligence / machine learning services to the user equipment; verifying the request using information about one or more network entities which are able to provide the one or more artificial intelligence / machine learning services to the user equipment; and sending to the user equipment, a response to the request for discovery, when the request has been verified.

21. A computer program comprising computer executable instructions which when run provide the method of claim 19 or claim 20.

Images