Transmitting a request for updating an ai / ML model based on the fulfilment of certain configured triggered conditions

Abstract

A method of operating a communications device is provided. The method comprises receiving, from an information processing server via a radio access network, trigger configuration information for triggering an update of an artificial intelligence (AI) model used by the communications device to perform a task. The trigger configuration information comprises one or more trigger conditions for triggering the communications device to transmit a request to update the AI model. Each of the one or more trigger conditions is based on one or more performance metrics indicative of a suitability of the AI model to perform the task. The method comprises calculating the one or more performance metrics for each of the one or more trigger conditions. The method comprises determining, based on the calculated one or more performance metrics for each of the one or more trigger conditions, that one or more of the trigger conditions are met. In response, the method comprises transmitting, to the information processing server via the radio access network, the request to update the AI model.

Description

[0001] TRANSMITTING A REQUEST FOR UPDATING AN AI / ML MODEL BASED ON THE FULFILMENT OF CERTAIN CONFIGURED TRIGGERED CONDITIONS

[0002] BACKGROUND

[0003] Field of Disclosure

[0004] The present disclosure relates to methods, communications devices, infrastructure equipment of a radio access network and information processing servers.

[0005] The present application claims Paris Convention priority from GB patent application number 2402279.0, filed on 16 February 2024.

[0006] Description of Related Art

[0007] The “background” description provided herein is for the purpose of generally presenting the context of the disclosure. Work of the presently named inventors, to the extent it is described in this background section, as well as aspects of the description which may not otherwise qualify as prior art at the time of filing, are neither expressly or impliedly admitted as prior art against the present invention.

[0008] Previous generation mobile telecommunication systems, such as those based on the 3GPP defined UMTS and Long Term Evolution (LTE) architecture, are able to support a wider range of services than simple voice and messaging services offered by previous generations of mobile telecommunication systems. For example, with the improved radio interface and enhanced data rates provided by LTE systems, a user is able to enjoy high data rate applications such as mobile video streaming and mobile video conferencing that would previously only have been available via a fixed line data connection. The demand to deploy such networks is therefore strong and the coverage area of these networks, i.e. geographic locations where access to the networks is possible, is expected to continue to increase rapidly.

[0009] Current and future wireless communications networks are expected to routinely and efficiently support communications with an ever-increasing range of devices associated with a wider range of data traffic profiles and types than existing systems are optimised to support. For example, it is expected that future wireless communications networks will be expected to efficiently support communications with devices including reduced complexity devices, machine type communication (MTC) devices, high resolution video displays, virtual reality headsets, extended Reality (XR) and so on. Some of these different types of devices may be deployed in very large numbers, for example low complexity devices for supporting the “The Internet of Things” and may typically be associated with the transmissions of relatively small amounts of data with relatively high latency tolerance. Other types of devices, for example supporting high-definition video streaming, may be associated with transmissions of relatively large amounts of data with relatively low latency tolerance. Other types of devices, for example used for autonomous vehicle communications and for other critical applications, may be characterised by data that should be transmitted through the network with low latency and high reliability. A single device type might also be associated with different traffic profiles / characteristics depending on the application(s) it is running. For example, different consideration may apply for efficiently supporting data exchange with a smartphone when it is running a video streaming application (high downlink data) as compared to when it is running an Internet browsing application (sporadic uplink and downlink data) or being used for voice communications by an emergency responder in an emergency scenario (data subject to stringent reliability and latency requirements).

[0010] In view of this there is expected to be a desire for current wireless communications networks, for example those which may be referred to as 5G or new radio (NR) systems / new radio access technology (RAT) systems, or indeed future 6G wireless communications, as well as future iterations / releases of existing systems, to efficiently support connectivity for a wide range of devices associated with different applications and different characteristic data traffic profiles and requirements.

[0011] One example of a new service is referred to as Ultra Reliable Low Latency Communications (URLLC) services which, as its name suggests, requires that a data unit or packet be communicated with a high reliability and with a low communications delay. Another example of a new service is extended Reality (XR), which may be provided by various user equipment such as wearable devices. XR combines real- world and virtual environments, incorporating aspects such as augmented reality (AR), mixed reality (MR), and virtual reality (VR), and thus requires high quality and minimised interaction delay. Services such as URLLC and XR therefore represent a challenging example for both LTE type communications systems and 5G / NR communications systems, as well as future generation communications systems.

[0012] 5G NR has continuously evolved and the current work plan includes 5G-NR-advanced in which some further enhancements are expected, especially to support new use-cases / scenarios with higher requirements. The desire to support these new use-cases and scenarios gives rise to new challenges for efficiently handling communications in wireless communications systems that need to be addressed.

[0013] SUMMARY OF THE DISCLOSURE

[0014] The present disclosure can help address or mitigate at least some of the issues discussed above.

[0015] Respective aspects and features of the present disclosure are defined in the appended claims.

[0016] It is to be understood that both the foregoing general description and the following detailed description are exemplary, but are not restrictive, of the present technology. The described embodiments, together with further advantages, will be best understood by reference to the following detailed description taken in conjunction with the accompanying drawings.

[0017] BRIEF DESCRIPTION OF THE DRAWINGS

[0018] A more complete appreciation of the disclosure and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following detailed description when considered in connection with the accompanying drawings wherein like reference numerals designate identical or corresponding parts throughout the several views, and wherein:

[0019] Figure 1 schematically represents some aspects of an LTE-type wireless telecommunication system which may be configured to operate in accordance with certain embodiments of the present disclosure;

[0020] Figure 2 schematically represents some aspects of an NR-type wireless telecommunications system which may be configured to operate in accordance with certain embodiments of the present disclosure;

[0021] Figure 3 is a schematic block diagram of an example infrastructure equipment and communications device which may be configured to operate in accordance with certain embodiments of the present disclosure;

[0022] Figure 4 schematically illustrates a life cycle management (LCM) architecture for an artificial intelligence model;

[0023] Figure 5 schematically illustrates a deployment of AI / ML positioning;

[0024] Figure 6A is part schematic, part message flow diagram illustrating communications between a communications device, infrastructure equipment of a radio access network and an information processing server in accordance with example embodiments; Figure 6B is part schematic, part message flow diagram illustrating communications between a communications device, infrastructure equipment of a radio access network and an information processing server in accordance with example embodiments;

[0025] Figure 7 is a flow diagram illustrating model management in accordance with example embodiments; Figure 8 is a signaling diagram illustrating model management in accordance with example embodiments;

[0026] Figure 9 is a signaling diagram illustrating model management in accordance with example embodiments.

[0027] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] Long Term Evolution Advanced Radio Access Technology (4G)

[0029] Figure 1 provides a schematic diagram illustrating some basic functionality of a mobile telecommunications network / system 6 operating generally in accordance with LTE principles, but which may also support other radio access technologies, and which may be adapted to implement embodiments of the disclosure as described herein. Various elements of Figure 1 and certain aspects of their respective modes of operation are well-known and defined in the relevant standards administered by the 3GPP (RTM) body, and also described in many books on the subject, for example, Holma H. and Toskala A [1], It will be appreciated that operational aspects of the telecommunications networks discussed herein which are not specifically described (for example in relation to specific communication protocols and physical channels for communicating between different elements) may be implemented in accordance with any known techniques, for example according to the relevant standards and known proposed modifications and additions to the relevant standards.

[0030] The network 6 includes a plurality of base stations 1 connected to a core network 2. Each base station provides a coverage area 3 (i.e., a cell) within which data can be communicated to and from communications devices 4. Although each base station 1 is shown in Figure 1 as a single entity, the skilled person will appreciate that some of the functions of the base station may be carried out by disparate, inter-connected elements, such as antennas (or antennae), remote radio heads, amplifiers, etc. Collectively, one or more base stations may form a radio access network.

[0031] Data is transmitted from base stations 1 to communications devices 4 within their respective coverage areas 3 via a radio downlink (DL). Data is transmitted from communications devices 4 to the base stations 1 via a radio uplink (UL). The core network 2 routes data to and from the communications devices 4 via the respective base stations 1 and provides functions such as authentication, mobility management, charging and so on. Terminal devices may also be referred to as mobile stations, user equipment (UE), user terminal, mobile radio, communications device, and so forth. Services provided by the core network 2 may include connectivity to the internet or to external telephony services. The core network 2 may further track the location of the communications devices 4 so that it can efficiently contact (i.e., page) the communications devices 4 for transmitting downlink data towards the communications devices 4.

[0032] Base stations, which are an example of network infrastructure equipment, may also be referred to as transceiver stations, nodeBs, e-nodeBs, eNB, g-nodeBs, gNB and so forth. The base station may also referred to as Radio Access Network (RAN) node (e.g., EUTRAN, NG RAN). In this regard different terminology is often associated with different generations of wireless telecommunications systems for elements providing broadly comparable functionality. However, certain embodiments of the disclosure may be equally implemented in different generations of wireless telecommunications systems, and for simplicity certain terminology may be used regardless of the underlying network architecture. That is to say, the use of a specific term in relation to certain example implementations is not intended to indicate these implementations are limited to a certain generation of network that may be most associated with that particular terminology. New Radio Access Technology (5G)

[0033] An example configuration of a wireless communications network which uses some of the terminology proposed for and used in NR and 5G is shown in Figure 2. In Figure 2 a plurality of transmission and reception points (TRPs) 10 are connected to distributed control units (DUs) 41, 42 by a connection interface represented as a line 16. Each of the TRPs 10 is arranged to transmit and receive signals via a wireless access interface within a radio frequency bandwidth available to the wireless communications network. Thus, within a range for performing radio communications via the wireless access interface, each of the TRPs 10 has a coverage area as represented by a circle 12. As such, wireless communications devices 14 which are within the coverage area 12 of each of the TRPs 10 can transmit and receive signals to and from those TRPs 10 via the wireless access interface. Each of the distributed units 41, 42 are connected to a central unit (CU) 40 (which may be referred to as a controlling node) via an interface 46. The central unit 40 is then connected to the core network 20 which may contain all other functions required to transmit data for communicating to and from the wireless communications devices and the core network 20 may be connected to other networks 30.

[0034] The elements of the wireless access network shown in Figure 2 may operate in a similar way to corresponding elements of an LTE network as described with regard to the example of Figure 1. It will be appreciated that operational aspects of the telecommunications network represented in Figure 2, and of other networks discussed herein in accordance with embodiments of the disclosure, which are not specifically described (for example in relation to specific communication protocols and physical channels for communicating between different elements) may be implemented in accordance with any known techniques, for example according to currently used approaches for implementing such operational aspects of wireless telecommunications systems, e.g. in accordance with the relevant standards.

[0035] The TRPs 10 of Figure 2 may in part have a corresponding functionality to a base station, eNodeB of an LTE network, or gNB of an NR network. Similarly, the communications devices 14 may have a functionality corresponding to the UE devices 4 known for operation with an LTE network or NR network. It will be appreciated therefore that operational aspects of a new RAT network (for example in relation to specific communication protocols and physical channels for communicating between different elements) may be different to those known from LTE, NR or other known mobile telecommunications standards. However, it will also be appreciated that each of the core network component, base stations and communications devices of a new RAT network will be functionally similar to, respectively, the core network component, base stations and communications devices of an LTE or NR wireless communications network.

[0036] In terms of broad top-level functionality, the core network 20 connected to the new RAT telecommunications system represented in Figure 2 may be broadly considered to correspond with the core network 2 represented in Figure 1, and the respective central units 40 and their associated distributed units / TRPs 10 may be broadly considered to provide at least some of the functionality corresponding to the base stations 1 of Figure 1. The term network infrastructure equipment / access node may be used to encompass these elements and more conventional base station type elements of wireless telecommunications systems. Depending on the application at hand the responsibility for scheduling transmissions which are scheduled on the radio interface between the respective distributed units and the communications devices may lie with the controlling node / central unit and / or the distributed units / TRPs. A communications device 14 is represented in Figure 2 within the coverage area of the first coverage area 12. This communications device 14 may thus exchange signalling with the first central unit 40 in the first coverage area 12 via one of the distributed units / TRPs 10 associated with the first coverage area 12.

[0037] It will further be appreciated that Figure 2 represents merely one example of a proposed architecture for a new RAT based telecommunications system in which approaches in accordance with the principles described herein may be adopted, and the functionality disclosed herein may also be applied in respect of wireless telecommunications systems having different architectures.

[0038] Thus, certain embodiments of the disclosure as discussed herein may be implemented in wireless telecommunication systems / networks according to various different architectures, such as the example architectures shown in Figures 1 and 2. It will thus be appreciated the specific wireless telecommunications architecture in any given implementation is not of primary significance to the principles described herein. In this regard, certain embodiments of the disclosure may be described generally in the context of communications between network infrastructure equipment / access nodes and a communications device, wherein the specific nature of the network infrastructure equipment / access node and the communications device will depend on the network infrastructure for the implementation at hand. For example, in some scenarios the network infrastructure equipment / access node may comprise a base station, such as an LTE-type base station 1 as shown in Figure 1 which is adapted to provide functionality in accordance with the principles described herein, and in other examples the network infrastructure equipment may comprise a control unit / controlling node 40 and / or a TRP 10 of the kind shown in Figure 2 which is adapted to provide functionality in accordance with the principles described herein.

[0039] A more detailed diagram of some of the components of the network shown in Figure 2 is provided by Figure 3. In Figure 3, a TRP 10 as shown in Figure 2 comprises, as a simplified representation, a wireless transmitter 30, a wireless receiver 32 and a controller or controlling processor 34 which may operate to control the transmitter 30 and the wireless receiver 32 to transmit and receive radio signals to one or more UEs 14 within a coverage area 12 formed by the TRP 10. As shown in Figure 3, an example UE 14 is shown to include a corresponding transmitter 49, a receiver 48 and a controller 44 which is configured to control the transmitter 49 and the receiver 48 to transmit signals representing uplink data to the wireless communications network via the wireless access interface formed by the TRP 10 and to receive downlink data as signals transmitted by the transmitter 30 and received by the receiver 48 in accordance with the conventional operation.

[0040] The transmitters 30, 49 and the receivers 32, 48 (as well as other transmitters, receivers and transceivers described in relation to examples and embodiments of the present disclosure) may include radio frequency filters and amplifiers as well as signal processing components and devices in order to transmit and receive radio signals in accordance for example with the 5G / NR standard. The controllers 34, 44 (as well as other controllers described in relation to examples and embodiments of the present disclosure) may be, for example, a microprocessor, a CPU, or a dedicated chipset, etc., configured to carry out instructions which are stored on a computer readable medium, such as a non-volatile memory. The processing steps described herein may be carried out by, for example, a microprocessor in conjunction with a random access memory, operating according to instructions stored on a computer readable medium. The transmitters, the receivers and the controllers are schematically shown in Figure 3 as separate elements for ease of representation. However, it will be appreciated that the functionality of these elements can be provided in various different ways, for example using one or more suitably programmed programmable computer(s), or one or more suitably configured application-specific integrated circuit(s) / circuitry / chip(s) / chipset(s). As will be appreciated the infrastructure equipment / TRP / base station as well as the UE / communications device will in general comprise various other elements associated with its operating functionality.

[0041] As shown in Figure 3, the TRP 10 also includes a network interface 50 which connects to the DU 42 via a physical interface 16. The network interface 50 therefore provides a communication link for data and signalling traffic from the TRP 10 via the DU 42 and the CU 40 to the core network 20.

[0042] The interface 46 between the DU 42 and the CU 40 is known as the Fl interface which can be a physical or a logical interface. The Fl interface 46 between CU and DU may operate in accordance with specifications 3GPP TS 38.470, 3GPP TS 38.473 and 3GPP TS 38.401, and may be formed from a fibre optic or other wired or wireless high bandwidth connection. In one example the connection 16 from the TRP 10 to the DU 42 is via fibre optic. The connection between a TRP 10 and the core network 20 can be generally referred to as a backhaul, which comprises the interface 16 from the network interface 50 of the TRP 10 to the DU 42 and the Fl interface 46 from the DU 42 to the CU 40. The core network 20 is connected to the CU 40 via the N2 (also called NG-C) interface for carrying control data and via the N3 (also called NG-U) interface for carrying user data.

[0043] The core network 20 may comprise core network functions such as a location management function (EMF) for managing a position of communications devices in the wireless communications network. In addition, the core network 20 may comprise one or more network functions for Artificial Intelligence / Machine Learning (AI / ML) model management, training, and storage.

[0044] Although reference has been made above to 4G / LTE and 5G NR, it will be appreciated that the present disclosure is applicable to future generations of wireless communications technology including 6G. In the case of 6G, base station, which are an example of network infrastructure equipment, may also referred to as 6G NB (6G Node B), 6G RAN node, and so forth. In 6G, the core network 20 may be one or more network functions.

[0045] Artificial Intelligence (Al)

[0046] In existing techniques (also called “legacy techniques”), the position of a UE can be determined by a UE or gNB based on positioning measurements made by the gNB or UE. Positioning measurements may be performed by a UE on downlink signals such as Positioning Reference signals (PRS), or performed by a gNB on uplink signals such as Sounding Reference Signal (SRS). Positioning measurements may comprise one or more of time of arrival (TOA), angle of arrival (AOA), angle of departure (AOD), a round trip time (RTT), reference signal received power (RSRP), and any other measurement used in determining the position of a UE. A UE may determine its position for example by receiving plurality of downlink signals, each from a different gNB, measuring a time of arrival and / or angle of the downlink signals and, based on the measurements, determining the position of the UE (e.g. multilateration).

[0047] However, the precision of the position determined by existing techniques is limited. In order to address technical problems in determining a precise location / position estimation of a communications device, the use of Artificial Intelligence / Machine Learning (AI / ML) models for positioning was firstly studied in Release-18 of the 3GPP standards and further developed in later release(s).

[0048] AI / ML-based positioning tends to outperform other positioning techniques because AI / ML models are configured to collect and process large amounts of positioning measurements. By performing model training (based on input data), AI / ML models learn and establish an understanding of the environment associated with the positioning measurements. AI / ML models may then be deployed, for example at a UE or base-station (such as a gNB), in order to generate output, such as improved (i.e. more accurate) positioning measurements or a position estimate. This process / operation is also known as AI / ML model inference. For example, the output positioning measurements may be improved in the sense that they are more likely to result in a more accurate position estimate. For example, a UE may receive a DL-PRS from a gNB and make positioning measurements on it. However, the UE may not know that the DL-PRS reflected off an object before it reached the UE. The AI / ML model may be configured to correct for such reflections. The AI / ML model can produce output on refined LOS / NLOS component identification. Additional detail on how AI / ML models can be used to generate more accurate positioning measurements, or generate a position estimate, can be found in [2], the contents of which are hereby incorporated by reference in their entirety.

[0049] As examples, AI / ML models may utilise one or more of supervised learning, generative Al Autoencoding, and reinforcement learning.

[0050] Supervised learning

[0051] AI / ML models may implement a supervised machine learning model.

[0052] The supervised learning model is trained using labelled training data to learn a function that maps inputs (typically provided as feature vectors) to outputs (i.e. labels). The labelled training data comprises pairs of inputs and corresponding output labels. The output labels are typically provided by an operator to indicate the desired output for each input. The supervised learning model processes the training data to produce an inferred function that can be used to map new (i.e. unseen) inputs to a label.

[0053] The input data (during training and / or inference) may comprise various types of data, such as numerical values, images, video, text, or audio. Raw input data may be pre-processed to obtain an appropriate feature vector used as input to the model - for example, features of an image or text input may be extracted to obtain a corresponding feature vector. It will be appreciated that the type of input data and techniques for pre-processing of the data (if required) may be selected based on the specific task the supervised learning model is used for.

[0054] Once prepared, the labelled training data set is used to train the supervised learning model. During training the model adjusts its internal parameters (e.g. weights) so as to optimize (e.g. minimize) an error function, aiming to minimize the discrepancy between the model’s predicted outputs and the labels provided as part of the training data. In some cases, the error function may include a regularization penalty to reduce overfitting of the model to the training data set.

[0055] The supervised learning model may use one or more machine learning algorithms in order to learn a mapping between its inputs and outputs. Example suitable learning algorithms include linear regression, logistic regression, artificial neural networks, decision trees, support vector machines (SVM), random forests, and the K-nearest neighbour algorithm.

[0056] For positioning AI / ML models, the input to the supervised learning model may comprise positioning measurements (for example, measurements of one or more PRSs, or one or more SRSs, such as time of arrival and angle of arrival) and the output of the supervised learning model may comprise a position estimate of the communications device or improved positioning measurements.

[0057] Once trained, the supervised learning model may be used for inference - i.e. for predicting outputs for previously unseen input data. The supervised learning model may perform classification and / or regression tasks. In a classification task, the supervised learning model predicts discrete class labels for input data, and / or assigns the input data into predetermined categories. In a regression task, the supervised learning model predicts labels that are continuous values.

[0058] For positioning AI / ML models, the supervised learning model is used to predict an output comprising improved positioning measurements or a position estimate based on an input comprising positioning measurements.

[0059] In some cases, limited amounts of labelled data may be available for training of the model (e.g. because labelling of the data is expensive or impractical). In such cases, the supervised learning model may be extended to further use unlabelled data and / or to generate labelled data.

[0060] Considering using unlabelled data, the training data may comprise both labelled and unlabelled training data, and semi-supervised learning may be used to learn a mapping between the model’s inputs and outputs. For example, a graph-based method such as Laplacian regularization may be used to extend a SVM algorithm to Laplacian SVM in order to perform semi-supervised learning on the partially labelled training data.

[0061] Considering generating labelled data, an active learning model may be used in which the model actively queries an information source (such as a user, or operator) to label data points with the desired outputs. Labels are typically requested for only a subset of the training data set thus reducing the amount of labelling required as compared to fully supervised learning. The model may choose the examples for which labels are requested - for example, the model may request labels for data points that would most change the current model, or that would most reduce the model's generalization error. Semi-supervised learning algorithms may then be used to train the model based on the partially labelled data set.

[0062] Generative Al

[0063] AI / ML models may implement generative artificial intelligence (Al).

[0064] A generative Al system learns patterns and structures in its input training data, in order to then generate new output data which exhibits similar characteristics to the training data. For positioning AI / ML models, the input training data may comprise positioning measurements and the output training data may comprise a position estimate of the communications device or improved positioning measurements.

[0065] The generative Al system may generate output data based on an input prompt. The prompt may comprise various types of data, such as images, video, text, or audio. The prompt may be of the same or different data type to the model’s training and / or output data.

[0066] The generative AI / ML model may comprise a generative model trained to learn a probability distribution of the input training data, and generate new output data based on this learned distribution. For example, for a set of data instances / observable variables (X) and a set of labels / target variables (Y) in the training data set, the generative model may learn a joint probability distribution of data instances and labels p(X,Y), and / or a probability distribution of the data instances p(X) (for example where no labels are available).

[0067] Example suitable generative models for learning a probability distribution of the input training data include Variational Autoencoders (VAEs), transformer-based models, diffusion models (e.g. denoising diffusion probabilistic models (DDPMs)), Reinforcement Learning (RL), and Generative Adversarial Networks (GANs). The choice of generative model may depend on the specific task performed by the generative AI / ML.

[0068] The generative model may comprise one or more artificial neural networks. For example, a Variational Autoencoder (VAE) may comprise a pair of neural networks acting as an encoder and a decoder to and from a reduced (i.e. latent space) representation of the training data respectively, and a Generative Adversarial Network (GAN) may comprise a first ‘generator’ neural network that generates new data and a second ‘discriminator’ neural network that learns to discriminate between generated data and real data. The one or more constituent neural networks of the generative model may be trained together or separately.

[0069] During training the generative model may adjust its internal parameters (e.g. neural network weights) so as to optimize (e.g. minimize) a loss / error function, aiming to minimize discrepancy between the generated output data and desired output data. It will be appreciated that the specific loss function, and algorithm used to optimize the function may vary depending on the nature of the generative model, and its intended application. For example, a mean squared error loss function may be used for an image generation task, and a cross-entropy loss function may be used for a text generation task. These loss functions may be optimized using various existing optimization algorithms, such as gradient descent.

[0070] Once trained, the generative model may be used to generate new output data based on an input prompt. The input prompt may be provided by a user, or by an appropriate device (e.g. using an application programming interface (API)). Thus, the generative Al / ML model allows generating new based on only a prompt and without requiring detailed instructions for doing so.

[0071] Autoencoders

[0072] AI / ML models may implement Autoencoding.

[0073] An autoencoder is a type of an unsupervised machine learning model that uses one or more artificial neural networks to learn an efficient representation of unlabelled input data. The autoencoder may be used to encode various types of data, such as images, video, text, audio, or positioning measurements.

[0074] The autoencoder may comprise an encoder neural network that encodes input data into a reduced representation (also called a “latent space”), and a decoder neural network that aims to recreate the input data from the encoded reduced representation. The latent space is typically of a lower-dimension than the input data - thus, the latent space generated by the encoder typically provides a more efficient, compressed representation of the input data that requires less memory storage than the original input data.

[0075] The encoder neural network may comprise one or more layers that transform input data into a reduced representation. The encoder neural network receives input data, and the final layer of the encoder neural network outputs a reduced representation of the input data, i.e. a latent space (also termed a “bottleneck layer”).

[0076] The decoder neural network comprises one or more layers that transform data from the latent space into output data of the same dimensionality as the data input to the encoder. The decoder aims to reconstruct the data originally input to the encoder neural network from the latent space representation of the data.

[0077] The encoder and / or decoder neural networks typically comprise a plurality of hidden layers. For example, an encoder may comprise a plurality of hidden layers that progressively extract further reduced representations of the input data. Using deeper neural networks (i.e. with a higher number of hidden layers) for the encoder and / or the decoder may improve performance of the autoencoder, and in some cases may reduce the amount of training data that is required.

[0078] The encoder and decoder neural networks are typically trained together. During training the autoencoder may adjust its internal parameters (e.g. weights and biases of the encoder and decoder neural networks) so as to optimize (e.g. minimize) a loss / error function, aiming to minimize discrepancy between the data input to the encoder and the output reconstructed data generated by the decoder. It will be appreciated that the specific loss function, and algorithm used to optimize the function may vary depending on the nature of the autoencoder model, and its intended application. In an example, a mean squared error loss function optimized using gradient descent may be used. In some cases, a sparse autoencoder may be used in order to promote sparsity of the latent representation (as compared to the input) and to prevent the autoencoder from learning the identity function - for example, a sparse autoencoder may be implemented by modifying the loss function to include a sparsity regularization penalty.

[0079] In some cases, the autoencoder may be a Variational Autoencoder (VAE). The VAE is a specific type of auto-encoder in which a probability model is imposed on the encoded representation by the training process (in that deviations from the probability model are penalised by the training process). The VAE may be used for generative artificial intelligence applications to generate new output data which exhibits similar characteristics to the input encoded data by sampling from the learned latent space.

[0080] For positioning AI / ML models, the input data may comprise positioning measurements and the output data may comprise a position estimate of the communications device or improved positioning measurements.

[0081] Reinforcement learning

[0082] AI / ML models may implement reinforcement learning (RL).

[0083] Reinforcement learning is a type of machine learning directed to training an artificial intelligence agent to take actions in an environment that maximize the notion of a cumulative reward. During reinforcement learning, the agent interacts with the environment, and learns from the results of its actions, thus allowing the agent to progressively improve its decision-making.

[0084] An RL model typically comprises an action-reward feedback loop. The feedback loop comprises: an environment, state, agent, policy, action, and reward. The environment is the system with which the agent interacts and in which the agent operates - for example, the environment may be a virtual environment of a video game. The state represents the current conditions in the environment. The agent receives the state as an input and takes an action which may affect the environment and change the state of the environment. The agent takes the action based on its policy which is a mapping from states of the environment to actions of the agent. The policy may be deterministic or stochastic. The reward represents feedback from the environment to the action taken by the agent. The reward provides an indication (typically in the form of a numerical value) of the desirability of the result of the agent’s action. The reward may comprise positive signals to reward desirable behaviour of the agent and / or negative signals to penalize undesirable behaviour of the agent.

[0085] Through multiple iterations of action-reward feedback loop, the agent aims to maximise the total cumulative reward it receives, thus learning how to take optimal actions in the environment. The reinforcement learning process thus allows the agent to learn an optimal policy that maximizes the cumulative reward. The cumulative award may be estimated using a value function which estimates the expected return starting from a given state or from a given state and action. Using the cumulative reward in the reinforcement learning process allows the agent to consider long-term effects of its policy.

[0086] A reinforcement learning algorithm may be used to refine the agent’s policy and the value function over iterations of the action-reward feedback loop. The learning algorithm may rely on a model of the environment (e.g. based on Markov Decision Processes (MDPs)) or be model-free. Example suitable model-free reinforcement learning algorithms include Q-leaming, SARSA (State-Action-Reward-State- Action), Deep Q-Networks (DQNs), or Deep Deterministic Policy Gradient (DDPG).

[0087] It will be appreciated that the agent will typically engage in both exploration and exploitation of the environment in which it operates. In exploration, the agent takes typically random actions to gather information about the environment and identify potentially desirable actions (i.e. actions that maximise cumulative reward). In exploitation, the agent takes actions that are expected to maximise reward (e.g. by selecting the action based on the agent’s latest policy). Various techniques may be used to control the proportion of explorative and exploitative actions taken by the agent - for example, a predetermined probability of taking an explorative action in a given iteration of the feedback loop may be set (and optionally reduced over time to allow the agent to shifts more towards exploitation over time to maximise cumulative reward in view of diminishing returns for further exploration).

[0088] In some cases, the RL model may be configured to learn from feedback provided by a user. Utilising user feedback in this way may allow the agent to improve its choice of actions and better align with user preferences. For example, reinforcement learning from human feedback (RLHF) techniques may be used. RLHF includes training a reward model based on user feedback and using this model for determining the reward in the reinforcement learning process described above. The user feedback may be received in various forms depending on the specific reinforcement learning problem being solved - for example, the feedback may be received in the form of a user ranking of instances of the agent’s actions. RLHF thus allows incorporating user feedback into the reinforcement learning process. RLHF approaches may be advantageous where it is easier for a user than for an algorithm to assess the quality of the machine learning model’s output (e.g. for generative artificial intelligence RL models).

[0089] For positioning AI / ML models, the input data may comprise positioning measurements and the output data may comprise a position estimate of the communications device or improved positioning measurements.

[0090] Although a number of types of AI / ML models have been described above in connection with positioning, the above types of a AI / ML can also be used in beam management or CSI tasks, or indeed other tasks known to a person skilled in the art.

[0091] For beam management AI / ML models, input data may comprise one or more of an Ll-RSRP and its associated beam / resource ID. The output data may be one or more of a predicted beam, narrow beam, improved beam prediction or a refined RSRP.

[0092] For CSI AI / ML models (e.g. CSI measurement and reporting, CSI compression or CSI prediction), input data may comprise one or more of a signal to interference and noise ratio (SINR) estimation, a Modulation and Coding (MCS) index, Channel Quality Indicator (CQI) index, or a Precoding matrix index (PMI). The output data may be one or more of a refined or predicted MCS, CQI index, PMI index, RI index. In some examples of CSI compression AI / ML models, input data may be a full CSI compression report and output data may be a reduced size CSI compression report.

[0093] Life cycle management (LCM) architecture for an AI / ML model

[0094] 3 GPP has identified a general AI / ML framework for NR air interface to facilitate different radio frequency applications including positioning applications but also for beam management and CSI feedback enhancements applications. The aim of the framework is to cover a general architecture addressing the whole Al model life cycle, Life Cycle Management, (LCM), including such as data collection, model training, etc. Here, LCM for AI / ML for NR air interface is described in the below. The following descriptions are also applied to 6G.

[0095] An example of a life cycle management (LCM) architecture for an artificial intelligence model is schematically illustrated in Figure 4 which has been reproduced from [2], the contents of which are hereby incorporated by reference in their entirety. As shown in Figure 4, LCM architecture comprises a data collection function 402, an inference function 404, a management function 406, a model training function 408 and a model storage function 410.

[0096] The data collection function 402 is configured to provide training data to the model training function 408, to provide monitoring data to the management function 406 and to provide inference data to the inference function 404. Data collection function 402 is a process / fiinction of collecting data by the network nodes, management entity, or UE for the purpose of AI / ML model training, data analytics and inference.

[0097] The inference function 404 is configured to provide an inference output to the management function 406, to receive a management instruction from the management function 406 and to receive a model from the model storage function 410. The inference function 404 is a process / fiinction of using a trained AI / ML model to produce a set of outputs based on a set of inputs.

[0098] The management function is configured to provide performance feedback and / a retraining request to the model training function 408, and to provide a model delivery request to the model storage unit 410.

[0099] The model training function 408 is configured to provide an updated model to the model storage unit 410. The model training function 408 is a process / fiinction to train an AI / ML Model (e.g., by learning the input / output relationship) in a data driven manner and obtain the trained AI / ML Model for inference.

[0100] As shown in Figure 4, the management function 406 represents the core of the LCM architecture. The function of the management function 406 is to monitor model performance at different entities (such as UEs or gNBs) and request delivery of updated models. Model management comprises two procedures - model switching and model updating.

[0101] Existing positioning procedures in wireless communications networks, such as 5G NR networks, involve communications between a UE, a gNB and a LMF. The AI / ML model may be deployed on the UE, gNB or LMF. An AI / ML model for positioning may be a “direct AI / ML positioning model” or an “AI / ML assisted positioning model”. Both direct AI / ML positioning models and AI / ML assisted positioning models use positioning measurements (such as measurements based on received PRS or SRS) as input data. However, the output data for direct AI / ML positioning models is an estimate of a position of the UE (such as a co-ordinate) whereas the output data for AI / ML assisted positioning models is a more accurate positioning measurement or a new set of positioning measurements. Examples of existing positioning techniques using a direct and assisted AI / ML positioning models are illustrated in Figure 5. Segment (A) of Figure 5 illustrates an example of UE-based positioning with a UE-side direct AI / ML model. In this example, one or more gNB transmits PRS to the UE. The UE performs positioning measurements on the PRS and uses a direct AI / ML model to generate an estimate of a position of the UE based on the positioning measurements. The UE forwards the position estimate to the LMF.

[0102] Segment (B) of Figure 5 illustrates an example of UE-assisted / LMF -based positioning with a UE-side AI / ML assisted positioning model. In this example, one or more gNB transmits PRS to the UE. The UE performs positioning measurements on the PRS and uses an AI / ML assisted positioning model to generate improved positioning measurements of the PRS. The UE forwards the improved measurements of the PRS to the LMF. The LMF generates position estimate of the UE based on the improved measurements of the PRS.

[0103] Segment (C) of Figure 5 illustrates an example of LMF-based positioning with an LMF-side direct AI / ML model. In this example, one or more gNB transmits PRS to the UE. The UE performs positioning measurements on the PRS and transmits the positioning measurements of the PRS to the LMF. The LMF then uses the direct AI / ML model generate a position estimate for the UE based on the positioning measurements of the PRS.

[0104] Segment (D) of Figure 5 illustrates an example of NG-RAN node assisted positioning with a gNB-side AI / ML assisted positioning model. In this example, the UE transmits an SRS to one or more gNB. The gNB performs positioning measurements on the DRS and uses the AI / ML assisted positioning model to generate more improved positioning measurements of the SRS. The gNB then transmits the improved positioning measurements of the SRS to the LMF. The LMF generates a position estimate of the UE based on the improved positioning measurements of the SRS.

[0105] Segment (E) of Figure 5 illustrates NG-RAN node assisted positioning with an LMF-side direct AI / ML model. In this example, the UE transmits an SRS to one or more gNB. The gNB performs positioning measurements on the SRS and transmits the positioning measurements of the SRS to the LMF. The LMF uses the direct AI / ML model to generate a position estimate of the UE based on the received positioning measurements of the SRS.

[0106] An AI / ML positioning model may be deployed for a particular area, while another AI / ML positioning model may be deployed for another area. In this way, AI / ML positioning models can be adapted to be effective in the areas in which they are deployed. For example, a particular AI / ML model may be trained under particular environmental conditions because these environmental conditions will be similar to the environmental conditions in which the AI / ML model will be deployed. Having multiple AI / ML models, each adapted for different areas, also facilitates model switching and selection, thus enabling the provision of high accuracy positioning across a diverse range of areas.

[0107] However, the environmental conditions of an area in which an AI / ML positioning model is deployed may change overtime. For example, the position (e.g. layout or constellation) of communications devices relative to passive environmental objects (such as a table or a wall) in the area may change over time. Since the passive environmental objects may reflect / scatter signals transmitted by the communications devices, the change in position of the communications devices relative to passive environmental objects means that different signals will be reflected / scattered in different ways over time. This may affect channel conditions and propagation characteristics such as line-of-sight (LOS) in the area. Accordingly, an AI / ML model adapted for a particular area may become outdated over time. If the AI / ML model becomes outdated, then the accuracy of positioning measurements or positioning estimates generated by the AI / ML model decreases to unacceptable levels. The decrease in performance of an AI / ML model over time may be due to data drift and / or model drift, data drift and / or model drift.

[0108] Data drift occurs when the distribution of input data (for example, positioning measurements) to the AI / ML model change overtime. This happens, for example, when the distribution of the positioning measurements obtained by a UE deviates from the distribution of the positioning measurements used to train the AI / ML model. For example, when the AI / ML model is trained, it may be assumed that the UEs may evenly spread out in the room. However, in practice or in a reality at a given period of time, UEs tend to stay in sub-areas of the room (e.g. most UEs may be near the centre of the room). This deviation between the assumed UE statistic (distribution of location and positioning measurements) in training phase and the realistic UE statistic might harm the overall performance.

[0109] Model drift occurs when a relationship between the input data and output data of the AI / ML model changes. This relationship may change due to a change in environmental conditions. For example, a room may have different reflectors such as tables and chairs. The location of these reflectors may not be fixed all the time. Considering a case where the configuration of the tables and chairs in the room is changed but the UE is still using the same AI / ML model for positioning, then the issue of model drift occurs, leading to compromised positioning performance.

[0110] Therefore, the decrease in performance of AI / ML problems over time represents a technical problem.

[0111] There is therefore a need for improved methods, communications devices, infrastructure equipment and information processing servers which address at least some of the above issues.

[0112] Figure 6A shows a part schematic, part message flow diagram representation of a wireless communications system in accordance with example embodiments. The wireless communications system comprises a communications device 601 (e.g. a UE 14), infrastructure equipment 602 (e.g. a gNB) of a radio access network and an information processing server 603 (such as an LMF server, a dedicated Al server or another server configured to communicate with an LMF server). In some embodiments, the information processing server 603 is a server in a core network and the infrastructure equipment 602 forms part of a radio access network through which communications devices can access the core network.

[0113] The communications device 601 comprises a transceiver 601.1 (or transceiver circuitry) and a controller 601.2 (or controller circuitry). The infrastructure equipment 602 comprises a transceiver 602.1 (or transceiver circuitry) and a controller 602.2 (or controller circuitry). The information processing server 603 comprises a transceiver 603.1 (or transceiver circuitry) and a controller 603 (or controller circuitry). The transceivers 601.1, 602.1, 603.1 are configured to transmit and receive signals. The transceivers 601.1, 602. 1 , 603.1 (or transceiver circuitry) may each comprise a separate transmitter or receiver (or separate transmitter and receiver circuitry), or the transceivers 601.1, 602. 1 , 603.1 (or transceiver circuitry) may each comprise a device (or circuitry) configured to perform both transmission and reception. Each of the controllers 601.2, 602.2, 603.2 may be, for example, a microprocessor, a CPU, or a dedicated chipset (System on Chip, SoC), etc.

[0114] The communications device 601 is configured to communicate with the infrastructure equipment 602 using wireless communications signals over a radio access interface. The infrastructure equipment 602 is configured to communicate with the information processing server 603 using signals which may be transmitted wirelessly, or transmitted via wired connections, or a combination of both. The communications device 601 is configured to communicate with the information processing server 603 via the infrastructure equipment 602 of the radio access network.

[0115] As shown in Figure 6A, the controller 603.2 of the information processing server 603 is configured to control the transceiver 603.1 to transmit 604, to the communications device 601 via the infrastructure equipment 602 of the radio access network, trigger configuration information. Accordingly, the controller 601.2 of the communications device 601 is configured to control the transceiver 601.1 of the communications device 601 to receive the trigger configuration information.

[0116] The trigger configuration information is for triggering an update of an artificial intelligence (Al) model used by the communications device 601 to perform a task. The task may be a positioning task for positioning the communications device 601, a beam management task, a channel state information (CSI) measurement and reporting task, a CSI compression task, or a CSI prediction task, for example. Alternatively, or additionally, the task may be another task performed by the AI / ML models.

[0117] In some embodiments, the Al model may be a machine learning model. Machine learning Al models will be referred to herein as “AI / ML models”. Although “AI / ML models” will be referred to in connection with example embodiments, the present disclosure is not so limited and Al models not implementing machine learning may be used.

[0118] The trigger configuration information comprises one or more trigger conditions for triggering the communications device 601 to transmit a request to update the Al model. Each of the one or more trigger conditions may be based on one or more performance metrics indicative of a suitability of the Al model to perform the task.

[0119] The trigger conditions may comprise one or more thresholds which, if met, trigger the communications device 601 to transmit the request to update the Al model. For example, the threshold may be confidence level (such as a percentage confidence) of a positioning accuracy. The calculated performance metric may provide an indication of the positioning accuracy achieved by the communications device 601. In another example, the threshold may be a statistic representative of a maximum acceptable deviation in positioning measurements in a time period.

[0120] The controller 601.2 of the communications device 601 is configured to control 605 the communications device 601 to calculate the one or more performance metrics for each of the one or more trigger conditions which may be configured by the trigger configuration information.

[0121] The controller 601.2 of the communications device 601 is configured to control 606 the communications device 601 to determine, based on the calculated one or more performance metrics for each of the one or more trigger conditions which may be configured by the trigger configuration information, that one or more of the trigger conditions are met.

[0122] In response to determining the one or more of the trigger conditions are met, the controller 601.2 of the communications device 601 is configured to control the transceiver 601. 1 of the communications device 601 to transmit 607, to the information processing server 603 via the infrastructure equipment 602 of the radio access network, the request to update the Al model. Alternatively, from the communications device 601 (e.g., UE) perspective, the destination where the communications device 601 transmits the request may be the infrastructure equipment 602. In this case, the infrastructure equipment 602 may transfer, to the information processing server 603, the request or information corresponding to the request. In accordance with example embodiments, after the information processing server 603 receives the request to update the Al model, the controller 603.2 of the information processing server 603 may control the information processing server 604 to update the Al model. The updating of the Al model may comprise updating one or more parameters of the Al model and / or updating a structure of the Al model. The updating of the Al model may comprise re-training the Al model and / or fine tuning the Al model. For example, the Al model the information processing server 603 may obtain updated training data and use the updated training data to retrain the Al model. The information processing server 603 may determine how to obtain the updated training data. For example, the information processing sever 603 may transmit, to other communications devices (such as PRUs) in an area to which the AI / ML model is applicable, a request to collect updated training data (such as new positioning measurements). The communications devices may report the updated training data to the information processing server 603 for the information processing server 603 to use to update the AI / ML model.

[0123] In accordance with example embodiments, the controller 603.2 of the information processing server 603 may control the transceiver 603. 1 of the information processing server 603 to transmit, to the communications device 601 an indication of the updated Al model. The communications device 601 may then use the updated Al model to perform the task (e.g. the positioning task, the beam management task, the CSI task or another task performed by the AI / ML model.).

[0124] Example embodiments can therefore trigger a communications device to transmit a request to update an Al model based on one or more performance metrics indicative of a suitability of the Al model for performing a task. Accordingly, example embodiments can update an Al model before it becomes unsuitable for performing the task due to deterioration over time due to, for example, model drift or data drift. Therefore, the performance of Al models in wireless communications networks is improved.

[0125] Figure 6B shows a part schematic, part message flow diagram representation of a wireless communications system in accordance with example embodiments. The wireless communications system comprises the infrastructure equipment 602 (e.g. a gNB) of the radio access network and the information processing server 603 (such as an LMF server, a dedicated Al server or another server configured to communicate with an LMF server). The configuration of the infrastructure equipment 602 and information processing server 603 has already been described with reference to Figure 6A and will not be repeated here for brevity.

[0126] As shown in Figure 6B, the controller 603.2 of the information processing server 603 is configured to control the transceiver 603.1 to transmit 608, to the infrastructure equipment 602 of the radio access network, trigger configuration information. Accordingly, the controller 602.2 of the infrastructure equipment is configured to control the transceiver 602.1 of the infrastructure equipment 602 to receive the trigger configuration information.

[0127] The trigger configuration information is for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment 602 to perform a task. The task may be a positioning task for positioning the communications device 601, a beam management task or a channel state information (CSI) task, for example.

[0128] The trigger configuration information comprises one or more trigger conditions for triggering the infrastructure equipment 602 to transmit a request to update the Al model. Each of the one or more trigger conditions are based on one or more performance metrics indicative of a suitability of the Al model to perform the task.

[0129] The controller 602.2 of the infrastructure equipment 602 is configured to control 609 the infrastructure equipment 602 to calculate the one or more performance metrics for each of the one or more trigger conditions which may be configured by the trigger configuration information.

[0130] The controller 602.2 of the infrastructure equipment is configured to control 610 the infrastructure equipment 602 to determine, based on the calculated one or more performance metrics for each of the one or more trigger conditions which may be configured by the trigger configuration information, that one or more of the trigger conditions are met.

[0131] In response to determining the one or more of the trigger conditions are met, the controller 602.2 of the infrastructure equipment 602 is configured to control the transceiver 602. 1 of the infrastructure equipment 602 to transmit 611, to the information processing server 603, the request to update the Al model.

[0132] In accordance with example embodiments, after the information processing server 604 receives the request to update the Al model, the controller 603.2 of the information processing server 604 may control the information processing server 604 to update the Al model. The updating of the Al model may comprise updating one or more parameters of the Al model and / or updating a structure of the Al model. The updating of the Al model may comprise re-training the Al model and / or fine tuning the Al model. For example, the Al model the information processing server may obtain updated training data and use the updated training data to retrain the Al model.

[0133] In accordance with example embodiments, the controller 603.2 of the information processing server 604 may control the transceiver 603.1 of the information processing server 604 to transmit, to the communications device 601 an indication of the updated Al model. The infrastructure equipment 602 may then use the updated Al model to perform the task (e.g. the positioning task, the beam management task, the CSI task or another task performed by the AI / ML models).

[0134] Example embodiments can therefore trigger infrastructure equipment of a radio access network to transmit a request to update an Al model based on one or more performance metrics indicative of a suitability of the Al model for performing a task. Accordingly, example embodiments can update an Al model before it becomes unsuitable for performing the task due to deterioration over time due to, for example, model drift or data drift. Therefore, the performance of Al models in wireless communications networks is improved.

[0135] Example embodiments below will refer to UEs, gNBs and an LMF server for explanation only. Unless otherwise stated, example embodiments explained with reference to UEs are generally applicable to communications devices, example embodiments explained with reference to gNBs are generally applicable to infrastructure equipment of a radio access network and example embodiments explained with reference to LMF servers are generally applicable to information processing servers.

[0136] Example embodiments are applicable to scenarios where a direct, or assisted, AI / ML model is deployed on a UE or a gNB. Example scenarios to which example embodiments are applicable include:

[0137] — UE-based positioning with a UE-side direct AI / ML model (e.g. segment A of Figure 5)

[0138] — UE-assisted / LMF -based positioning with UE-side AI / ML assisted positioning model (e.g. segment B of Figure 5) — NG-RAN node assisted positioning with gNB-side AI / ML assisted positioning model (e.g. segment D of Figure 5)

[0139] According to example embodiments, methods for updating an AI / ML model in radio nodes (communications devices such as a UEs, or infrastructure equipment of a radio access network such as gNBs) for performing a positioning task are provided. As explained above, an information processing server transmits trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the radio node to perform a task. In some embodiments, the trigger configuration information comprises an indication of a format for calculating the one or more performance metrics for each of the one or more trigger conditions. In some embodiments, one or more of the trigger conditions are based on comparing a performance metric for the trigger condition with a threshold. In such embodiments, the threshold is in the same format as the format for calculating the performance metric which was indicated in the trigger configuration information.

[0140] In some embodiments, one or more of the trigger conditions comprise a condition on an expiry time for the AI / ML model. For example, the information processing server may determine, based on previous experience, for example, that after an expiry time the AI / ML model is no longer suitable to perform the task (e.g. the positioning task). This may be due to model drift or data drift overtime for example. In such embodiments, the calculation of the performance metric may comprise calculating a time for which the AI / ML model has been in use, and the trigger condition is that the time for which the AI / ML model has been in use is greater than or equal to the expiry time. In cases where the information processing server expects the performance of the AI / ML model to degrade over time, the time for which the AI / ML model is in use is an indication of the suitability of the AI / ML model to perform the task.

[0141] In some embodiments the trigger configuration information comprises an indication of a calculation trigger time for the AI / ML model. In such embodiments, the one or more performance metrics for each of the one or more trigger conditions are calculated in response to the communications device or infrastructure equipment determining that the calculation trigger time has been reached. In such embodiments, power can be saved because the communications device or infrastructure equipment only needs to calculate the performance metrics when the calculation time is reached.

[0142] In some embodiments, after the radio node determines that a trigger condition has been met, the radio node may transmit the update request immediately, after a pre-defined interval or periodically. In some embodiments, the trigger configuration information comprises an indication of a time interval during which the request to update the AI / ML model should be transmitted. The time interval may be referred to as AI / ML update request computation time, AI / ML update request processing time, or so forth.

[0143] In some embodiments, the trigger configuration information may comprise an instruction to one or more radio nodes to calculate one or more performance metrics. For example, the trigger configuration information may comprise an instruction to a first radio node that the first radio node should calculate a performance metric indicative of data drift. Trigger configuration information transmitted from the information processing server to another radio node may comprise an instruction to calculate a performance metric indicative of model drift. In such embodiments the second radio node may be a positioning reference unit (PRU).

[0144] In some embodiments, the one or more performance metrics for one or more of the trigger conditions comprise a performance metric indicative of a data drift of input data to the AI / ML model. Data drift may be due to UE behaviour. For example, UEs may all prefer staying in the center of the room, rather than evenly spread out in the room. Performance metrics indicative of data drift may measure changes in the statistical properties of the model input (e.g. positioning measurements). Correcting for data drift does not require a Ground Truth (GT) label. Since mobile UEs may not be able to acquire their location from other source accurately, correcting for data drift is easier than correcting for model drift (which requires a GT label). Therefore, it is particularly advantageous for mobile UEs if one or more performance metrics indicative of a data drift are used for one or more of the trigger conditions. Examples of performance metrics indicative of data drift comprise: i. The changes of mean, variance, deviation or other relevant statistics of committed information rates (CIRs), Time of Arrivals (TOAs), Reference Signal Received Powers (RSRPs), captured at different periods of time. ii. Statistical distance metrics, such as Kolmogorov-Smirnov (KS) statistic, to quantify the difference between the distribution of positioning measurements at different periods of time.

[0145] To calculate a performance metric indicative of data drift, the radio node may perform a plurality positioning measurements over a plurality of time periods. For example, a UE may perform 100 positioning measurements in one day and calculate a reference statistic S 1 based on the positioning measurements. In each subsequent day, the UE may perform another 100 positioning measurements per day and measure a statistic Sn, where n is the day in which the positioning measurements are measured. The performance metric may be a difference in the statistics calculated from the positioning measurements in each day, for example |Sn-Sl|. In one example, a trigger condition may be that the UE transmits a request to update the AI / ML model when |Sn-Sl| is first determined to exceed a threshold, X. For example, if on day four, the statistic S4 is such that | S4-S 11 > X, then the UE transmits a request to update the AI / ML model. In another example, the statistic | Sn-S 11 may be averaged over a period of time and when the average exceeds the threshold X, then the UE transmits the request to update the Al mode.

[0146] In some embodiments, the one or more performance metrics for one or more of the trigger conditions comprise a performance metric indicative of a model drift of input data to the AI / ML model. Typically, model drift occurs due to changes the environment in the area in which the AI / ML model is applicable. Performance metrics indicative of model drift may measure changes in the relationship between the positioning measurements and output data of the AI / ML model such as improved positioning measurements or a position estimate. Correction for model drift requires a GT label. Correction of model drift is therefore easier for stationary UE which has a fixed, known location (e.g. a positioning reference unit (PRU) or a UE which can obtain an accurate position estimate from other sources).

[0147] Examples of performance metrics indicative of model drift comprise: i. A statistic indicative of a difference between model output (prediction) and the GT label (in other words, the positioning accuracy). ii. A statistic indicating whether there are sudden drops or fluctuations of positioning accuracy. iii. A / B testing indicator. The A / B indicator may be one or more of a positioning accuracy, an Fl score, precision and recall of the AI / ML model.

[0148] For a direct-AI / ML positioning model, the model output and GT label, are position estimates of the UE (such as a co-ordinate of the UE). For an assisted- AI / ML positioning model, the model output and GT label, may be accurate positioning measurements (such as LOS / NLOS indicators, TOA, or RSRP). When there a plurality of performance metrics to be calculated, the radio node may calculate the performance metrics at the same time, or may calculate each of the performance metrics at different times. Furthermore, the computation of the performance metrics may be performed by a specific radio node. For example, performance metrics indicative of model drift may be calculated by a UE such as a PRU.

[0149] In some embodiments, the request to update the AI / ML model comprises an identification of the AI / ML model currently used by the communications device. The ID may be a global ID of the AI / ML model, for example. In some embodiments, the request to update the AI / ML model comprises an indication of the calculated one or more performance metrics for the trigger condition that was met. In some embodiments, where the trigger condition which was met was based on a plurality of performance metrics calculated by the radio node, the update request may comprise an indication of the plurality of performance metrics calculated by the node. The indication may be calculated values of the performance metrics. Alternatively, or additionally, the indication may be one or more indices indicating or associated with the calculated performance metrics. In such embodiments, the indication of each performance metric may be transmitted at the same time or at different times.

[0150] In response to receiving the update request, the information processing server updates the AI / ML model. For example, the information processing server (e.g. a model training unit of the information processing server) may initiate a data collection event and re-train or fine-tune the AI / ML model based on the collected data. The updated model may be delivered to the radio node and a model storage unit of the information processing server to replace the outdated AI / ML model.

[0151] Figure 7 is a flow diagram illustrating a method of model management in accordance with example embodiments. The method is performed by a radio node (such as a UE or gNB). The method starts in step S702.

[0152] In step S704, the radio node receives the trigger configuration information from an information processing server (e.g. an LMF server). In the example shown in Figure 7, the trigger configuration information is for an update of an AI / ML model used by the radio node to perform a positioning task. The trigger configuration information comprises one or more trigger conditions for triggering the communications device to transmit a request to update the AI / ML model. Each of the one or more trigger conditions is based on one or more performance metrics indicative of a suitability of the AI / ML model to perform the positioning task.

[0153] In step S706, the radio node calculates the one or more performance metrics for each of the one or more trigger conditions. In the example shown in Figure 7, the radio node calculates a performance metric for a trigger condition based on channel measurements and positioning measurements. For example, the performance metric may be a statistic indicative of a deviation of the channel and positioning measurements overtime.

[0154] In steps S708, the radio node compares the performance metric with a second threshold (Threshold2). The second threshold represents a threshold deviation of channel and positioning measurements which is higher than a threshold deviation of channel and positioning measurements represented by a first threshold (Thresholdl). If the radio node determines, that the performance metric is greater than the second threshold, then the method proceeds to step S710. In step S170, the radio node uses legacy positioning techniques (i.e. positioning techniques not using an AI / ML model) to determine the position of the UE. The high deviation represented by the performance metric is an indication that legacy techniques would provide a more accurate position estimate compared to techniques using an AI / ML model, because it is an indication that the AI / ML currently being used is not working effectively.

[0155] If the radio node determines, that the performance metric is less than or equal to the second threshold, then the method proceeds to step S712. In step S712, the radio node compares the performance metric with the first threshold. If the radio node determines that the performance metric is greater than the first threshold, then then the method proceeds to step S714. In other words, the radio node determines that a trigger condition for transmitting an update request to the information processing server has been met.

[0156] In step S714, the radio node transmits, to the information processing server, the request to update the AI / ML model. In the example shown in Figure 7, the radio node also transmits an indication of the calculated performance metric to the information processing server.

[0157] If the radio node determines that the performance metric is less than or equal to the first threshold in step S712, the method proceeds to step S716 where the method ends. In other words, the radio node determines that the AI / ML model is currently sufficiently suitable to perform the positioning task that no update is required.

[0158] Figure 8 is a signaling diagram illustrating model management in accordance with example embodiments. In the example of Figure 8, the AI / ML model is a UE-side model.

[0159] In step S802, an information processing server transmits an AI / ML model for performing a positioning task to a UE via a gNB. In other words, the gNB transparently transfer the received AI / ML model to the UE. In step S804, the server transmits trigger configuration information to the UE via the gNB. In other words, the gNB transparently transfer the received trigger configuration information to the UE. In the example shown in Figure 8, the trigger configuration information is for triggering an update of the AI / ML model transmitted to the UE in step S802. The trigger configuration information comprises one or more trigger conditions for triggering the UE to transmit a request to update the AI / ML model. Each of the one or more trigger conditions is based on one or more performance metrics indicative of a suitability of the AI / ML model to perform the positioning task.

[0160] In step S806, the gNB transmits one or more downlink position reference signals (e.g., DL-PRSs) to the UE. In some embodiments, one or more other gNBs also transmit one or more DL-PRSs to the UE. In step S808, the UE performs positioning measurements on the received DL-PRSs. The positioning measurements may comprise, for example, TOA measurements, and / or AOA measurements, for example. The UE calculates a performance metric based on the positioning measurements (for example, a statistic indicative of a deviation in the positioning measurements over time). In one example, the UE may obtain one hundred positioning measurements in a day and computes a reference distribution represented by a statistic SI based on the measurements. In the second day, the UE may perform another one hundred positioning measurements and computes a distribution represented by a statistic S2. Then the UE may calculate the performance metric based on the difference between S2 and S 1. Since the performance metric is based on a deviation of input data to the AI / ML model, this performance metric is indicative of data drift.

[0161] In step S810, the UE determines, based on the calculated performance metric, whether or not a trigger condition has been met. If the trigger condition is met (for example, the performance metric exceeds a threshold indicated in the trigger condition) the UE subsequently transmits a request, in step S812, to the information processing server via the gNB, to update the AI / ML model. In step S814, the information processing server updates the AI / ML model by, for example, re-training or fine tuning the model.

[0162] In step S816, the information processing server transmits the updated AI / ML model to the UE via the gNB.

[0163] In step S818, the UE deploys the updated AI / ML model and uses it to perform future positioning tasks.

[0164] Figure 9 is a signaling diagram illustrating model management in accordance with example embodiments. In the example of Figure 9, the AI / ML model is a UE-side model. In Figure 9, the information processing server comprises a model management unit, a model training unit and a model storage unit. The model management unit, model training unit and model storage unit be located in the same or different apparatus. One or more of the management unit, model training unit and model storage unit May be located in an LMF server, for example.

[0165] In step S902, the UE transmits, to the model management unit, a request for an AI / ML model for a positioning task.

[0166] In step S904, the model management forwards the request for the AI / ML model to the model storage unit.

[0167] In step S906, the model storage unit transmits the requested AI / ML model to the UE via a gNB. In other words, the gNB transparently transfer the received AI / ML model to the UE.

[0168] In step S908, the model management unit transmits trigger configuration information to the UE via the gNB. In other words, the gNB transparently transfer the received trigger configuration information to the UE. In the example shown in Figure 9, the trigger configuration information is for an update of the AI / ML model transmitted to the UE in step S906. The trigger configuration information comprises one or more trigger conditions for triggering the UE to transmit a request to update the AI / ML model. Each of the one or more trigger conditions is based on one or more performance metrics indicative of a suitability of the AI / ML model to perform the positioning task.

[0169] In step S910, the gNB transmits one or more downlink position reference signals (e.g., DL-PRSs) to the UE. In some embodiments, one or more other gNBs also transmit one or more DL-PRSs to the UE. In step S912, the UE performs positioning measurements on the received DL-PRSs. The UE calculates a performance metric based on the positioning measurements (for example, a statistic indicative of a deviation in the positioning measurements over time). Since the performance metric is based on a deviation of input data to the AI / ML model, this performance metric is indicative of data drift.

[0170] In step S912, the UE determines, based on the calculated performance metric, whether or not a trigger condition has been met. If the trigger condition is met (for example, the performance metric exceeds a threshold indicated in the trigger condition) the UE subsequently transmits, in step S916 a request, to the model management unit via the gNB, to update the AI / ML model.

[0171] In step S918, the model management unit forwards the update request to the model training unit.

[0172] In step S920, the model training unit updates the AI / ML model (for example, by retraining and / or fine tuning the AI / ML model). In step S922, the model training unit transmits the updated AI / ML model to the UE via the gNB.

[0173] In step S924, the model training unit transmits the updated AI / ML model to the model storage unit for storage.

[0174] In step S926, the UE deploys the updated AI / ML model and uses it to perform future positioning tasks.

[0175] In step S928, the model storage unit updates the AI / ML model in the storage unit to the updated AI / ML model. Bor example, the model storage unit may delete the previously stored AI / ML model and store the updated AI / ML model.

[0176] Although Figures 8 and 9 describe a UE-side AI / ML model with the UE calculating a performance metric based on positioning measurements of DL-PRSs, example embodiments are equally applicable to cases where there is a gNB-side AI / ML model with the gNB calculating a performance metric based on positioning measurements of uplink signals (such as SRSs).

[0177] In some embodiments, the information processing server may determine whether one or more trigger conditions for updating the AI / ML model are met based on calculated performance metrics received from the radio node (such as a communications device or infrastructure equipment of a radio access network). In such embodiments, the radio node periodically transmits an indication of one or more performance metrics calculated by the radio node to the information processing sever. The indication of the one or more performance metrics may be transmitted periodically without receiving any trigger, or trigger condition, for transmitting the one or more performance metrics to the radio node. For example, the radio node does not receive an instruction to transmit the performance metrics and does not receive a trigger condition indicating to transmit the performance metrics if a condition is met. In embodiments, where the information processing server determines whether one or more trigger conditions for updating the AI / ML model is sent, the information processing server does not transmit, to the radio node, the trigger configuration information described with reference to Figure 6A and 6B.

[0178] Those skilled in the art would appreciate that the method shown by Figures 6A-9 may be adapted in accordance with embodiments of the present technique. For example, other intermediate steps may be included in such a method, or the steps may be performed in any logical order. Though embodiments of the present technique have been described by way of the example systems and methods shown in Figures 6A-9, it would be clear to those skilled in the art that they could be equally applied to other systems to those described herein.

[0179] Those skilled in the art would further appreciate that such infrastructure equipment and / or communications devices as herein defined may be further defined in accordance with the various arrangements and embodiments discussed in the preceding paragraphs. It would be further appreciated by those skilled in the art that such infrastructure equipment and communications devices as herein defined and described may form part of communications systems other than those defined by the present disclosure.

[0180] The following numbered paragraphs provide further example aspects and features of the present technique:

[0181] Paragraph 1. A method of operating a communications device, the method comprising receiving, from an information processing server via a radio access network, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, calculating the one or more performance metrics for each of the one or more trigger conditions, determining, based on the calculated one or more performance metrics for each of the one or more trigger conditions, that one or more of the trigger conditions are met, and in response, transmitting, to the information processing server via the radio access network, the request to update the Al model.

[0182] Paragraph 2. A method according to paragraph 1, wherein the information processing server comprises one or more of: a location management function (LMF) server, an Al server, or a server configured to communicate with an LMF server.

[0183] Paragraph 3. A method according to paragraph 1 or paragraph 2, wherein the one or more performance metrics for one or more of the trigger conditions comprise a performance metric indicative of data drift of input data to the Al model.

[0184] Paragraph 4. A method according to any of paragraphs 1 to 3, wherein the one or more performance metrics for one or more of the trigger conditions comprise performance a metric indicative of a model drift of the Al model.

[0185] Paragraph 5. A method according to any of paragraphs 1 to 4, wherein the trigger configuration information comprises an indication of a format for calculating the one or more performance metrics for each of the one or more trigger conditions.

[0186] Paragraph 6. A method according to paragraph 5, wherein one or more of the trigger conditions are based on comparing a performance metric for the trigger condition with a threshold.

[0187] Paragraph 7. A method according to paragraph 6, wherein an indication of the threshold for one or more of the trigger conditions is comprised in the trigger configuration information.

[0188] Paragraph 8 A method according to paragraph 6 or paragraph 7, wherein the threshold is in the same format as the format for calculating the performance metric.

[0189] Paragraph 9. A method according to any of paragraphs 1 to 8, wherein one or more of the trigger conditions comprise a condition on an expiry time for the Al model.

[0190] Paragraph 10. A method according to any of paragraphs 1 to 9, wherein the trigger configuration information comprises an indication of a time interval during which the request to update the Al model should be transmitted.

[0191] Paragraph 11. A method according to any of paragraphs 1 to 10, wherein the request to update the Al model comprises an identification of the Al model currently used by the communications device

[0192] Paragraph 12. A method according to any of paragraphs 1 of 11, wherein the request to update the Al model comprises an indication of the calculated one or more performance metrics for the trigger condition that was met. Paragraph 13. A method according to paragraph 12, wherein the trigger condition which was met was based on a plurality of performance metrics calculated by the communications device, and the method comprises transmitting an indication of the plurality of performance metrics calculated by the communications device to the information processing server via the radio access network, wherein each of the plurality of performance metrics calculated by the communications device are transmitted at different times.

[0193] Paragraph 14. A method according to any of paragraphs 1 to 13, comprising receiving, from the information processing server via the radio access network, the updated Al model

[0194] Paragraph 15. A method according to any of paragraphs 1 to 14, wherein the task performed by the Al model is a positioning task for determining a position of the communications device, or a beam management task or a channel state information (CSI) measurement and reporting task, a CSI prediction task, or a CSI compression task.

[0195] Paragraph 16. A method according to paragraph 15, wherein the task performed by the Al model is a positioning task and the input data for the Al model comprises positioning signal measurements and the Al model is configured to generate, based on the input data, an estimate of more accurate positioning signal measurements or to generate, based on the input data, an estimate of a position of the communications device.

[0196] Paragraph 17. A method according to any of paragraphs 1 to 16, wherein the calculating the one or more performance metrics for each of the one or more trigger conditions comprises using the Al model to calculate the one or more performance metrics.

[0197] Paragraph 18. A method of operating infrastructure equipment of a radio access network, the method comprising receiving, from an information processing server, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and, transmitting, to the information processing server, a request to update the Al model, wherein the request to update the Al model is transmitted in response to a calculation of the one or more performance metrics for the one or more of the trigger conditions and a determination, based on the calculated one or more performance metrics for the one or more trigger conditions, that one or more of the trigger conditions are met.

[0198] Paragraph 19. A method of operating an information processing server, the method comprising transmitting, to infrastructure equipment of a radio access network or to a communications device via the infrastructure equipment, configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment or the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment or the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and receiving, from the infrastructure equipment or from the communications device via the infrastructure equipment, a request to update the Al model.

[0199] Paragraph 20. A method according to paragraph 19, wherein the information processing server comprises one or more of: a location management server (LMF), an Al server, a server configured to communicate with the LMF.

[0200] Paragraph 21. A method according to paragraph 20, wherein the information processing sever comprises a model management unit configured to receive the request to update the Al model from the infrastructure equipment, a model training unit configured to receive the request to update the Al model from the model management unit and to update the Al model based on the request, and a model storage unit configured to receive the updated Al model from the model training unit and to store the updated Al model.

[0201] Paragraph 22. A method of operating a communications device, the method comprising calculating one or more performance metrics indicative of a suitability of an artificial intelligence (Al) model used by the communications device to perform a task, and periodically transmitting, to an information processing server via a radio access network, an indication of the calculated one or more performance metrics.

[0202] Paragraph 23. A method of operating infrastructure equipment of a radio access network, the method comprises periodically transmitting, to an information processing server, an indication of one or more performance metrics calculated by the infrastructure equipment, the one or more performance metrics being indicative of a suitability of an artificial intelligence (Al) model used by the infrastructure equipment to perform a task.

[0203] Paragraph 24. A method of operating an information processing server, the method comprising periodically receiving, from infrastructure equipment of a radio access network or from a communications device via the infrastructure equipment, an indication of one or more performance metrics calculated by the infrastructure equipment or the communications device, the one or more performance metrics being indicative of a suitability an artificial intelligence (Al) model used by the infrastructure equipment or the communications device to perform a task, determining, based on the received performance metrics, that one or more trigger conditions for updating the Al model, are met updating the Al model, and transmitting an indication of the updated Al model to the infrastructure equipment or to the communications device via the infrastructure equipment.

[0204] Paragraph 25. A communications device, the communications device comprising transceiver circuitry configured to transmit and receive signals, controller circuitry configured in combination with the transceiver circuitry to receive, from an information processing server via a radio access network, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, calculate the one or more performance metrics for each of the one or more trigger conditions, determine, based on the calculated one or more performance metrics for each of the one or more trigger conditions, that one or more of the trigger conditions are met, and, in response, transmit, to the information processing server via the radio access network, the request to update the Al model.

[0205] Paragraph 26. Infrastructure equipment for a radio access network, the infrastructure equipment comprising transceiver circuitry configured to transmit and to receive signals, controller circuitry configured in combination with the transceiver circuitry to receive, from an information processing server, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and transmit, to the information processing server, a request to update the Al model, wherein the request to update the Al model is transmitted in response to a calculation of the one or more performance metrics for the one or more of the trigger conditions and a determination, based on the calculated one or more performance metrics for the one or more trigger conditions, that one or more of the trigger conditions are met.

[0206] Paragraph 27. An information processing server, the information processing server comprising transceiver circuitry configured to transmit and to receive signals, controller circuitry configured in combination with the transceiver circuitry to transmit, to infrastructure equipment of a radio access network or to a communications device via the infrastructure equipment, configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment or the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment or the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and receive, from the infrastructure equipment or from the communications device via the infrastructure equipment, a request to update the Al model.

[0207] It will be appreciated that the above description for clarity has described embodiments with reference to different functional units, circuitry and / or processors. However, it will be apparent that any suitable distribution of functionality between different functional units, circuitry and / or processors may be used without detracting from the embodiments.

[0208] Described embodiments may be implemented in any suitable form including hardware, software, firmware or any combination of these. Described embodiments may optionally be implemented at least partly as computer software running on one or more data processors and / or digital signal processors. The elements and components of any embodiment may be physically, functionally and logically implemented in any suitable way. Indeed, the functionality may be implemented in a single unit, in a plurality of units or as part of other functional units. As such, the disclosed embodiments may be implemented in a single unit or may be physically and functionally distributed between different units, circuitry and / or processors.

[0209] Although the present disclosure has been described in connection with some embodiments, it is not intended to be limited to the specific form set forth herein. Additionally, although a feature may appear to be described in connection with particular embodiments, one skilled in the art would recognise that various features of the described embodiments may be combined in any manner suitable to implement the technique.

[0210] References

[0211] [1] Holma H. and Toskala A, “LTE for UMTS OFDMA and SC-FDMA based radio access”, John Wiley and Sons, 2009.

[0212] [2] Technical Report (TR) 38.843, vl8.0.0, December 2023, 3rdGeneration Partnership Project (3GPP).

Claims

CLAIMSWhat is claimed is:

1. A method of operating a communications device, the method comprising receiving, from an information processing server via a radio access network, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, calculating the one or more performance metrics for each of the one or more trigger conditions, determining, based on the calculated one or more performance metrics for each of the one or more trigger conditions, that one or more of the trigger conditions are met, and in response, transmitting, to the information processing server via the radio access network, the request to update the Al model.

2. A method according to claim 1, wherein the information processing server comprises one or more of: a location management function (LMF) server, an Al server, or a server configured to communicate with an LMF server.

3. A method according to claim 1, wherein the one or more performance metrics for one or more of the trigger conditions comprise a performance metric indicative of data drift of input data to the Al model.

4. A method according to claim 1, wherein the one or more performance metrics for one or more of the trigger conditions comprise a performance metric indicative of a model drift of the Al model.

5. A method according to claim 1, wherein the trigger configuration information comprises an indication of a format for calculating the one or more performance metrics for each of the one or more trigger conditions.

6. A method according to claim 5, wherein one or more of the trigger conditions are based on comparing a performance metric for the trigger condition with a threshold.

7. A method according to claim 6, wherein an indication of the threshold for one or more of the trigger conditions is comprised in the trigger configuration information.8 A method according to claim 6, wherein the threshold is in the same format as the format for calculating the performance metric.

9. A method according to claim 1, wherein the request to update the Al model comprises an identification of the Al model currently used by the communications device10. A method according to claim 1, wherein the request to update the Al model comprises an indication of the calculated one or more performance metrics for the trigger condition that was met.

11. A method according to claim 10, wherein the trigger condition which was met was based on a plurality of performance metrics calculated by the communications device, and the method comprises transmitting an indication of the plurality of performance metrics calculated by the communications device to the information processing server via the radio access network, wherein each of the plurality of performance metrics calculated by the communications device are transmitted at different times.

12. A method according to claim 1, comprising receiving, from the information processing server via the radio access network, the updated Al model13. A method according to claim 1, wherein the task performed by the Al model is a positioning task for determining a position of the communications device, or a beam management task or a channel state information (CSI) measurement and reporting task, a CSI prediction task, or a CSI compression task.

14. A method according to claim 13, wherein the task performed by the Al model is a positioning task and the input data for the Al model comprises positioning signal measurements and the Al model is configured to generate, based on the input data, an estimate of more accurate positioning signal measurements or to generate, based on the input data, an estimate of a position of the communications device.

15. A method according to claim 1, wherein the calculating the one or more performance metrics for each of the one or more trigger conditions comprises using the Al model to calculate the one or more performance metrics.

16. A method of operating infrastructure equipment of a radio access network, the method comprising receiving, from an information processing server, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment to transmit a request to update the Al model, each of the one or more triggerconditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and transmitting, to the information processing server, a request to update the Al model, wherein the request to update the Al model is transmitted in response to a calculation of the one or more performance metrics for the one or more of the trigger conditions and a determination, based on the calculated one or more performance metrics for the one or more trigger conditions, that one or more of the trigger conditions are met.

17. A method of operating an information processing server, the method comprising transmitting, to infrastructure equipment of a radio access network or to a communications device via the infrastructure equipment, configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment or the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment or the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and receiving, from the infrastructure equipment or from the communications device via the infrastructure equipment, a request to update the Al model.

18. A method according to claim 17, wherein the information processing server comprises one or more of: a location management server (LMF), an Al server, a server configured to communicate with the LMF.

19. A method according to claim 18, wherein the information processing sever comprises a model management unit configured to receive the request to update the Al model from the infrastructure equipment, a model training unit configured to receive the request to update the Al model from the model management unit and to update the Al model based on the request, and a model storage unit configured to receive the updated Al model from the model training unit and to store the updated Al model.

20. A method of operating a communications device, the method comprising calculating one or more performance metrics indicative of a suitability of an artificial intelligence (Al) model used by the communications device to perform a task, and periodically transmitting, to an information processing server via a radio access network, an indication of the calculated one or more performance metrics.

21. A method of operating infrastructure equipment of a radio access network, the method comprises periodically transmitting, to an information processing server, an indication of one or more performance metrics calculated by the infrastructure equipment, the one or more performance metrics being indicative of a suitability of an artificial intelligence (Al) model used by the infrastructure equipment to perform a task.

22. A method of operating an information processing server, the method comprising periodically receiving, from infrastructure equipment of a radio access network or from a communications device via the infrastructure equipment, an indication of one or more performance metrics calculated by the infrastructure equipment or the communications device, the one or more performance metrics being indicative of a suitability an artificial intelligence (Al) model used by the infrastructure equipment or the communications device to perform a task, determining, based on the received performance metrics, that one or more trigger conditions for updating the Al model, are met updating the Al model, and transmitting an indication of the updated Al model to the infrastructure equipment or to the communications device via the infrastructure equipment.

23. A communications device, the communications device comprising transceiver circuitry configured to transmit and receive signals, controller circuitry configured in combination with the transceiver circuitry to receive, from an information processing server via a radio access network, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, calculate the one or more performance metrics for each of the one or more trigger conditions, determine, based on the calculated one or more performance metrics for each of the one or more trigger conditions, that one or more of the trigger conditions are met, and, in response, transmit, to the information processing server via the radio access network, the request to update the Al model.

24. Infrastructure equipment for a radio access network, the infrastructure equipment comprising transceiver circuitry configured to transmit and to receive signals, controller circuitry configured in combination with the transceiver circuitry to receive, from an information processing server, trigger configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment to transmit a request to update the Al model, each of the one or more triggerconditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and transmit, to the information processing server, a request to update the Al model, wherein the request to update the Al model is transmitted in response to a calculation of the one or more performance metrics for the one or more of the trigger conditions and a determination, based on the calculated one or more performance metrics for the one or more trigger conditions, that one or more of the trigger conditions are met.

25. An information processing server, the information processing server comprising transceiver circuitry configured to transmit and to receive signals, controller circuitry configured in combination with the transceiver circuitry to transmit, to infrastructure equipment of a radio access network or to a communications device via the infrastructure equipment, configuration information for triggering an update of an artificial intelligence (Al) model used by the infrastructure equipment or the communications device to perform a task, the trigger configuration information comprising one or more trigger conditions for triggering the infrastructure equipment or the communications device to transmit a request to update the Al model, each of the one or more trigger conditions being based on one or more performance metrics indicative of a suitability of the Al model to perform the task, and receive, from the infrastructure equipment or from the communications device via the infrastructure equipment, a request to update the Al model.

Images