Method performed by distributed unit of access network node, method performed by central unit of access network node, distributed unit of access network node, and central unit of access network node

Abstract

A method performed by a distributed unit of an access network node is disclosed. The method includes receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and inferring the AI / ML model.

Description

METHOD PERFORMED BY DISTRIBUTED UNIT OF ACCESS NETWORK NODE, METHOD PERFORMED BY CENTRAL UNIT OF ACCESS NETWORK NODE, DISTRIBUTED UNIT OF ACCESS NETWORK NODE, AND CENTRAL UNIT OF ACCESS NETWORK NODE

[0001] The present disclosure relates to a communication system. The disclosure has particular but not exclusive relevance to wireless communication systems and devices thereof operating according to the 3rd Generation Partnership Project (3GPP) standards or equivalents or derivatives thereof (including Long Term Evolution (LTE)-Advanced, Next Generation or 5G networks, future generations, and beyond). The disclosure has particular, although not necessarily exclusive, relevance to the implementation and co-ordination of artificial intelligence (AI) and machine learning (ML) models and AI / ML model inferences across distributed Radio Access Network (RAN) architectures (e.g., a RAN architecture comprising a central unit (CU) and one or more distributed units (DUs)).

[0002] Earlier developments of the 3GPP standards were referred to as the Long-Term Evolution (LTE) of Evolved Packet Core (EPC) network and Evolved Universal Mobile Telecommunications System (UMTS) Terrestrial Radio Access Network (E-UTRAN), also commonly referred as '4G'. More recently, the term '5G' and 'new radio' (NR) is used to refer to an evolving communication technology that supports a variety of applications and services. Various details of 5G networks are described in, for example, the 'NGMN 5G White Paper' V1.0 (NPL 1) by the Next Generation Mobile Networks (NGMN) Alliance, which document is available from https: / / www.ngmn.org / 5g-white-paper.html. 3GPP intends to support 5G by way of the so-called 3GPP Next Generation (NextGen) radio access network (RAN) and the 3GPP NextGen core network.

[0003] Under the 3GPP standards, a NodeB (or an eNB in LTE, and gNB in 5G) is the radio access network (RAN) node (or simply 'access node', 'access network node' or 'base station') via which communication devices (user equipments or 'UEs') connect to a core network and communicate with other communication devices or remote servers. For simplicity, the present application may use the term access network node, RAN node (or simply RAN) or base station to refer to any such access nodes.

[0004] For simplicity, the present application will use the term mobile device, user device, or UE, to refer to any communication device that is able to connect to the core network via one or more base stations. Although the present application may refer to mobile devices in the description, it will be appreciated that the technology described can be implemented on any communication devices (mobile and / or generally stationary) that can connect to a communication system for sending / receiving data, regardless of whether such communication devices are controlled by human input or software instructions stored in memory.

[0005] In the current 5G architecture, the RAN architecture may be distributed with the base station structure split into two or more parts. In some RAN implementations there are two parts, known as the Central Unit (CU or gNB-CU) - sometimes referred to as a 'control unit' - and the Distributed Unit (DU or gNB-DU), connected by an F1 interface. This enables the use of a 'split' architecture in which the typically 'higher' CU layers (for example, but not necessarily or exclusively, Packet Data Convergence Protocol (PDCP) and Radio Resource Control (RRC) layers) and the, 'lower' DU layers (for example, but not necessarily or exclusively, Radio Link Control (RLC), Media (sometimes referred to as 'Medium') Access Control (MAC), and Physical (PHY) layers) are separated between a particular CU, and one or more DUs that are connected to and controlled by that CU via the F1 interface. Thus, for example, the higher layer CU functionality for several base stations may be implemented centrally (for example, by a single processing unit, or in a cloud-based or virtualised system), whilst retaining the lower layer DU functionality locally separately for each base station.

[0006] In more recently proposed RAN distributed architectures, in addition to the CU and DU, the concept of a Radio Unit (RU) - sometimes referred to as a 'remote unit' - has been introduced. In this architecture the RU is responsible for handling the digital front end (DFE), digital beamforming functionality and, typically, the functionality of the lower parts of the PHY layer, whilst the DU typically handles the higher parts of the PHY layer and the RLC and MAC layers. The CU in this architecture continues to be responsible for controlling one or more DUs (each DU corresponding to a different respective gNB) and to handle higher layer signalling (typically RRC and PDCP layers).

[0007] The actual functional split between the CU and DUs (and potentially RUs where applicable) of these distributed architectures is flexible allowing the functionality to be optimised for different use cases. Effectively, the split architecture enables a 5G network to use a different distribution of protocol stacks between CU and DUs (and potentially RUs) depending on, for example, midhaul availability and network design.

[0008] In 5G, core network entities comprise logical nodes (or 'functions') including control plane functions (CPFs) and one or more user plane functions (UPFs). The CPFs include, amongst other things, one or more Access and Mobility Management Functions (AMFs), a session management function (SMF), an Authentication Server Function (AUSF), a Unified Data Management (UDM) entity for managing user specific data, a Policy Control Function (PCF), an Application Function (AF), a Security Anchor Function (SEAF), an Authentication credential Repository and Processing Function (ARPF), and / or the like. The AMF generally corresponds to the mobility management entity (MME) in 4G and performs many of the functions performed by the MME. Each UPF combines functionality of both the S-GW and P-GW - specifically user plane functionality of the S-GW (SGW-U) and user plane functionality of the P-GW (PGW-U). The SMF provides session management functionality (that formed part of MME functionality in 4G). The SMF also combines the some of the functionality provided by the S-GW and P-GW - specifically control plane functionality of the S-GW (SGW-C) and control plane functionality of the P-GW (PGW-C). The SMF also allocates IP addresses to each UE.

[0009] Some recent developments in 3GPP relate to the use of artificial intelligence (AI) and machine learning (ML), often abbreviated to AI / ML. Predictions or inferences generated using an AI / ML model can be used as part of various methods for improving the reliability or efficiency of communication in the network. For example, AI / ML models can be used to predict the path of a UE based on previous mobility of the UE, used for cell and / or beam management, or used in methods of encoding and transmitting information. An AI / ML model may be hosted at a RAN node, and the RAN node may perform control of communication resources or control related to the status of a UE (e.g., control of UE mobility, or control of a radio resource control (RRC) state of the UE) based on an inference (e.g., determination or prediction) generated using the AI / ML model. The RAN node may also transmit an inference generated using the model to another node in the network, for use at the other node.

[0010] Alternatively, an AI / ML model may be hosted at two or more nodes of the network, for example at a RAN node and at a UE. In this case, the RAN node and the UE may both make determinations or predictions using the model. For example, the UE may use the model as part of an encoding process for encoding (and / or compressing) channel state information (CSI) for transmission to the RAN node, and the RAN node may use the same model as part of a corresponding decoding (and / or decompression) process for decoding the CSI received from the UE.

[0011] It will nevertheless be appreciated that the use of AI / ML models may also be extended to other procedures and methods performed in communication system to further improve the reliability or efficiency of communication in the communication system, for example, in handover, network energy saving, network slicing, coverage and capacity optimisation, and / or beam switching procedures, to name but a few.

[0012] Current use cases of AI / ML models include procedures and methods to improve the reliability or efficiency of communication in a network. Typically, such AI / ML models are implemented in non-distributed nodes of the network (e.g., a UE and / or a non-distributed RAN node). Nevertheless, it will be appreciated that the implementation of such AI / ML models in distributed nodes (e.g., where the RAN architecture is distributed with the base station structure split into two or more parts such as a CU and DUs) may also be beneficial. Such implementation however is non-trivial. For example, where an AI / ML model is deployed at multiple different parts of the RAN architecture and / or the UE of the network, coordination between those different parts of the RAN architecture and / or the UE may be required to ensure the correct performance of the AI / ML model. Coordination may also be required to ensure that model inferences are generated and acted upon that are appropriate across the different parts of the RAN architecture and / or the UE.

[0013] NPL 1: NGMN 5G White Paper' V1.0

[0014] The disclosure aims to provide one or more apparatus and / or one or more associated methods that overcomes or at least partially ameliorates the above issues.

[0015] In one aspect there is provided a method performed by a distributed unit of an access network node, the method comprising: 　　receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and 　　inferring the AI / ML model.

[0016] In one aspect there is provided a method performed by a central unit of an access network node, the method comprising: 　　transmitting, to a distributed unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model, wherein 　　the input data is used by the distributed unit for use in inferring the AI / ML model.

[0017] In one aspect there is provided a distributed unit of an access network node, the distributed unit comprising: 　　means for receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and 　　means for inferring the AI / ML model.

[0018] In one aspect there is provided a central unit of an access network node, the central unit comprising: 　　means for transmitting, to a distributed unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model, wherein 　　the input data is used by the distributed unit for use in inferring the AI / ML model.

[0019] The various functional means described below that are part of the UE may be provided by a memory and one or more processors that execute instructions stored in the memory. Similarly, the various functional means described below that are part of the access network node may be provided by a memory and one or more processors that execute instructions stored in the memory.

[0020] Various example described below may be implemented by means of a computer program product comprising computer implementable instructions for causing a programmable computer to carry out the any of the methods described below. The computer implementable instructions may be provided as a signal or on a tangible computer readable medium.

[0021] According to the present disclosure, it is possible to provide a method performed by a distributed unit of an access network node, a method performed by a central unit of an access network node, a distributed unit of an access network node, and a central unit of an access network node.

[0022] Example embodiments of the disclosure will now be described, by way of example, with reference to the accompanying drawings in which:

[0023] Fig. 1 schematically illustrates a mobile ('cellular' or 'wireless') communication system;Fig. 2 illustrates a typical frame structure that may be used in the communication system of Fig. 1;Fig. 3 illustrates a functional framework for artificial intelligence (AI) / machine learning (ML) (AI / ML) models, and how various entities of the framework may interact with one another, that may be implemented in the communication system of Fig. 1;Fig. 4 schematically illustrates a method of training an AI / ML model, and of monitoring the performance of the AI / ML model, that may be implemented in the communication system of Fig. 1;Fig. 5 depicts a simplified sequence diagram illustrating a procedure for generating an inference using an AI / ML mode hosted at a RAN node DU that may be used in the communication system of Fig. 1;Fig. 6 depicts a simplified sequence diagram illustrating another procedure for generating an inference using an AI / ML mode hosted at a RAN node DU that may be used in the communication system of Fig. 1;Fig. 7 is a simplified sequence diagram illustrating an intra-RAN, inter-node, information exchange procedure for AI / ML model inference availability related information that may be implemented in the communication system of Fig. 1;Fig. 8 is a simplified sequence diagram illustrating another intra-RAN, inter-node, information exchange procedure for AI / ML model inference availability related information that may be implemented in the communication system of Fig. 1;Fig. 9 is simplified sequence diagram illustrating an intra-RAN, inter-node, information exchange procedure for AI / ML model inference generation disablement that may be implemented in the communication system of Fig. 1;Fig. 10 is a simplified sequence diagram illustrating another intra-RAN, inter-node, information exchange procedure for AI / ML model inference generation disablement that may be implemented in the communication system of Fig. 1;Fig. 11 is a simplified sequence diagram illustrating another intra-RAN, inter-node, information exchange procedure for AI / ML model inference generation disablement that may be implemented in the communication system of Fig. 1;Fig. 12 is a schematic block diagram illustrating the main components of a UE the communication system of Fig. 1;Fig. 13 is a schematic block diagram illustrating the main components of a non-distributed RAN node for the communication system of Fig. 1; andFig. 14 is a schematic block diagram illustrating the main components of a distributed RAN node for the communication system of Fig. 1.

[0024] < Overview > 　　An exemplary communication system will now be described in general terms, by way of example only, with reference to Figs. 1 to 4.

[0025] Fig. 1 schematically illustrates a mobile ('cellular' or 'wireless') communication system 1 to which examples of the present disclosure are applicable.

[0026] In the communication system 1, user equipments (UEs) 3 (3-1, 3-2, 3-3) (e.g., mobile telephones and / or other mobile devices) can communicate with each other via a corresponding (radio) access network ((R)AN) node 5-1, 5-2 that operates according to one or more compatible radio access technologies (RATs). In the illustrated example, each RAN node 5 (5-1, 5-2) comprises a base station 5 or 'gNB' that respectively operates one or more associated cells 9 (9-1, 9-2). In the illustrated communication system 1, the coverage provided by each RAN node 5 may be by means of a plurality of beams B (B1, B2 … Br, Br+1 … BN). It will be appreciated that while, for clarity of illustration, only a selection of possible beams B are shown for one of the RAN nodes 5, the set of beams may include any suitable number of beams and each RAN node 5 may operate a respective set of beams or may provide coverage in a non-beamformed manner.

[0027] Communication via each RAN node 5 is typically routed through a core network 7 (e.g., a 5G / 6G and / or later generations' core network or evolved packet core network (EPC)).

[0028] As those skilled in the art will appreciate, whilst three UEs 3 and two RAN nodes 5 are shown in Fig. 1 for illustration purposes, the system, when implemented, will typically include one or more other RAN nodes 5 and UEs 3.

[0029] Each RAN node 5 controls one or more associated cells 9 either directly, or indirectly via one or more other nodes (such as home base stations, relays, remote radio heads, distributed units, and / or the like). It will be appreciated that the RAN nodes 5 may be configured to support 4G, 5G, 6G and / or later generations, and / or any other 3GPP or non-3GPP communication protocols.

[0030] In this example one of the illustrated RAN nodes 5 is a RAN node 5-2 that forms part of a distributed RAN (which may be referred to as a 'distributed base station' or distributed 'RAN node'). The RAN node 5-2 of a distributed RAN comprises at least one distributed unit (DU) 5-2DU(e.g., a gNB-DU or the like), and a central unit (CU) 5-2CU(e.g., a gNB-CU or the like). The CU 5-2CUemploys a separated control plane and user plane and so is, itself, split between a control plane function (CU-CP) and a user plane function (CU-UP) which respectively communicate, with the DU 5-2DUvia a first interface (e.g., an F1-C logical interface) and a second interface (e.g., an F1-C logical interface) (the interfaces together forming a combined interface such as an F1 interface (or 'reference point')), and with one another via another interface (e.g., an E1 logical interface). It will be appreciated that while, in this example, the DU 5-2DUincludes the physical and virtual elements required to provide the functionality of the lower parts of the PHY layer and hence communicate with the UEs 3 over the air interface, the distributed RAN node 5-2 may alternatively (or additionally) include one or more separate radio units (RUs) (e.g., providing this functionality of the lower parts of the PHY layer).

[0031] Whilst one non-distributed ('integrated') RAN node 5-1 and one distributed RAN node 5-2 are shown, it will, nevertheless, be appreciated that either (or both) of the RAN nodes 5 may be provided in a distributed or non-distributed form. It will also be appreciated that whilst the term 'RAN node' is generally used herein to refer to a whole base station, the CU 5-2CUand DU 5-2DUare also both parts of a corresponding distributed RAN and are therefore each a distinct 'RAN node' albeit that they each form part of the same RAN, and that each may have only a subset of the functionality provided by a whole base station. References to a RAN node as used herein should, therefore, be understood, accordingly.

[0032] The UEs 3 and their serving RAN node 5 are connected via an appropriate air interface (for example the so-called 'Uu' interface and / or the like). Neighbouring RAN nodes 5 may be connected to each other via an appropriate RAN node to RAN node interface (such as the so-called 'X2' interface, 'Xn' interface and / or the like).

[0033] The core network 7 includes a number of logical nodes (or 'functions') for supporting communication in the communication system 1. In this example, the core network 7 comprises control plane functions (CPFs) 10 and one or more network node entities for the communication of user data (e.g., user plane functions (UPFs) 11). The CPFs 10 include one or more network node entities for the communication of control signalling (e.g., Access and Mobility Management Functions (AMFs) 10-1), one or more network node entities for session management (e.g., Session Management Functions (SMFs) 10-2) and a number of other functions 10-n. Additional functions may include, for example: an Authentication Server Function (AUSF) which facilitates security processes; a Unified Data Management (UDM) entity for managing user specific data (e.g., for access authorization, user registration, and data network profiles); a Policy Control Function (PCF); an Application Function (AF); a Security Anchor Function (SEAF) which is in a serving network and acts as a "middleman" during an authentication process between a UE 3 and its home network; an Authentication credential Repository and Processing Function (ARPF) which maintains the authentication credentials; and / or the like. It will be appreciated that the nodes or functions may have different names in different systems.

[0034] The communication system 1 also includes an Operations, Administration and Maintenance (OAM) 14 comprising one or more OAM functions for provisioning and managing network or elements within the wider communication system 1. The OAM 14 may be responsible for the storage and analysis of some radio-related measurements and may perform some data analytics functions including some RAN analytics. The OAM 14 may, for example, communicate with one or more of the core network CPFs 10 and with a network data analytics function (NWDAF) or the like (not shown).

[0035] Each RAN node 5 is respectively connected to the core network nodes via appropriate interfaces (or 'reference points') such as an N2 reference point between the RAN node 5 and the AMF 10-1 for the communication of control signalling, and an N3 reference point between the RAN node 5 and each UPF 11 for the communication of user data. The UEs 3 are each connected to the AMF 10-1 via a logical non-access stratum (NAS) connection over an appropriate reference point (e.g., N1 reference point (analogous to the S1 reference point in LTE)). It will be appreciated, that N1 communications are routed transparently via the RAN node 5.

[0036] Each UPF 11 is connected to an external data network 20 (e.g., an IP network such as the internet) via an appropriate reference point (e.g., an N6 reference point) for communication of the user data.

[0037] The AMF 10-1 performs mobility management related functions, maintains the NAS connection with each UE 3 and manages UE registration. The AMF 10-1 is also responsible for managing paging. The AMF 10-1 receives user information sent through the network and forwards the information to the SMF 10-2.

[0038] The SMF 10-2 is connected to the AMF 10-1 via an appropriate reference point (e.g., N11 reference point). The SMF 10-2 provides session management functionality (that formed part of MME functionality in LTE) and additionally combines some control plane functions (provided by the serving gateway and packet data network gateway in LTE). The SMF 10-2 uses user information provided via the AMF 10-1 to determine what session manager would be best assigned to the user. The SMF 10-2 may be considered effectively to be a gateway from the user plane to the control plane of the network. The SMF 10-2 also allocates IP addresses to each UE 3.

[0039] Each RAN node 5 is also configured for transmission of, and the UEs 3 are configured for the reception of, control information and user data via a number of downlink (DL) physical channels and for transmission of a number of physical signals. The DL physical channels correspond to resource elements (REs) carrying information originated from a higher layer, and the DL physical signals are used in the physical layer and correspond to REs which do not carry information originated from a higher layer.

[0040] The DL physical channels may include, for example, a physical downlink shared channel (PDSCH), a physical broadcast channel (PBCH), and a physical downlink control channel (PDCCH). The PDSCH carries data sharing the PDSCH's capacity on a time and frequency basis. The PDSCH can carry a variety of items of data including, for example, user data, UE-specific higher layer control messages mapped down from higher channels, system information blocks (SIBs), and paging. The PDCCH carries downlink control information (DCI) for supporting several functions including, for example, scheduling the downlink transmissions on the PDSCH and also the uplink data transmissions on a physical uplink shared channel (PUSCH). The PBCH provides at least the UEs 3 with the Master Information Block (MIB). It also, in conjunction with the PDCCH, supports the synchronisation of time and frequency, which aids cell acquisition, selection and re-selection. Specifically, a UE 3 may receive a Synchronization Signal / Physical Broadcast Channel (PBCH) Block (SSB) (also referred to as an 'SS / PBCH block'), and the UE 3 may assume that reception occasions of a PBCH, primary synchronization signal (PSS) and secondary synchronization signal (SSS) are in consecutive symbols and form that SSB. The RAN node 5 may transmit several SSBs corresponding to different DL beams. The total number of SSBs may be confined, for example, within a 5ms duration as an SS burst.

[0041] The DL physical signals may include, for example, reference signals (RSs) and synchronization signals (SSs). A reference signal (sometimes known as a pilot signal) is a signal with a predefined special waveform known to both the UE 3 and the RAN node 5. The reference signals may include, for example, cell specific reference signals, UE-specific reference signal (UE-RS), downlink demodulation signals (DMRS), and channel state information reference signal (CSI-RS).

[0042] Similarly, the UEs 3 are configured for transmission of, and the RAN node 5 is configured for the reception of, control information and user data via several uplink (UL) physical channels corresponding to REs carrying information originated from a higher layer, and UL physical signals which are used in the physical layer and correspond to REs which do not carry information originated from a higher layer. The physical channels may include, for example, the PUSCH, a physical uplink control channel (PUCCH), and / or a physical random-access channel (PRACH). The UL physical signals may include, for example, demodulation reference signals (DMRS) for a UL control / data signal, and / or sounding reference signals (SRS) used for UL channel measurement.

[0043] When the UE 3 initially establishes a radio resource control (RRC) connection with a RAN node 5 via a cell 9 it registers with an appropriate core network node (e.g., AMF 10-1, MME). The UE 3 is in the so-called RRC connected state and an associated UE context is maintained by the network. When the UE 3 is in the so-called RRC idle state, or is in the RRC inactive state, it selects an appropriate cell for camping so that the network is aware of the approximate location of the UE 3 (although not necessarily on a cell level).

[0044] < Frame Structure > 　　Referring to Fig. 2, which illustrates a typical frame structure that may be used in the communication system 1, the RAN node 5 and UEs 3 of the communication system 1 communicate with one another using resources that are organised, in the time domain, into frames of length 10ms. Each frame comprises ten equally sized subframes of 1ms length. Each subframe is divided into one or more slots comprising 14 Orthogonal frequency-division multiplexing (OFDM) symbols of equal length.

[0045] As seen in Fig. 2, the communication system 1 supports multiple different numerologies (subcarrier spacing (SCS), slot lengths and hence OFDM symbol lengths). Specifically, each numerology is identified by a parameter, μ, where μ=0 represents 15 kHz (corresponding to the LTE SCS). Currently, the SCS for other values of μ can, in effect, be derived from μ=0 by scaling up in powers of 2 (i.e., SCS = 15 x 2μkHz). The relationship between the parameter, μ, and SCS (Δf) is as shown in Table 1.

[0046] Table 1 - Numerology

[0047] < Control Information > 　　In the communication system 1, the RAN node 5 is configured to transmit control information to the UE 3 using one or more control resource sets (CORESETs). A CORESET is a set of time-frequency resources within which the UE 3 can search for DCI transmitted by the RAN node 5 on a PDCCH. A CORESET is analogous to the control region at the start of subframes in earlier generations of communication technology. Unlike earlier generations, however, in which the frequency domain of the control region typically corresponded to the total system bandwidth, the frequency domain location for CORESET is localised to a specific region in the frequency domain and has a variable width that can be set to any suitable value (typically in multiples of six resource blocks where each resource block comprises twelve subcarriers in the frequency domain).

[0048] A number of different DCI formats can be used by the RAN node 5, depending on requirements, for transmission on a PDCCH corresponding to one of the PDCCH candidates in one of the search spaces configured for a given UE 3. For example, the RAN node 5 may be able to transmit DCI using one or more of the currently standardised DCI formats as set out in Table 2.

[0049] Table 2 - DCI Format Summary

[0050] Different DCI formats may or may not have the same DCI size. Moreover, DCI may be addressed (scrambled) using different radio network temporary identifiers (RNTIs) that a UE 3 may monitor for. Typically, the UE 3 can monitor up to three different DCI sizes for DCI formats using a cell RNTI (C-RNTI) - typically used as an identifier for scheduling purposes. Additionally, the UE 3 is typically capable of monitoring one additional DCI size using other RNTIs for specific purposes (e.g., a slot format indication RNTI (SFI-RNTI), interruption RNTI (INT-RNTI), or the like). This constraint is sometimes referred to as the "3+1" size budget and is imposed because a DCI scrambled with a C-RNTI is, generally, more time critical than a DCI scrambled with a RNTI used for another specific purpose, and so requires the UE 3 to decode it promptly to be able to process the scheduled data transmission.

[0051] To take account of the constraint imposed by the DCI size budget, the sizes of some DCI formats may be aligned by padding, truncation, and / or determining a frequency domain resource assignment field differently.

[0052] The UE 3 may monitor a set of PDCCH candidates in one or more control resource sets (CORESETs) on an active DL bandwidth part, where monitoring implies decoding each PDCCH candidate according to the monitored DCI formats. The number of blind decodes (BDs) may be restricted on a per carrier basis of a serving cell. The number of BDs may refer to the number of monitored PDCCH candidates or the number of PDCCH candidates the UE 3 is capable of decoding within a certain time frame, such as a slot or span of consecutive symbols in a slot. As an example, at a 15 kHz subcarrier spacing (SCS), the maximum number of BDs per slot per serving cell supported by the UE 3 may be 44 BDs.

[0053] < AI / ML > 　　The communication system 1 supports the use of artificial intelligence (AI) and machine learning (ML), often abbreviated to AI / ML in accordance with recent developments in cellular communication technology (e.g., as part of the work of the 3GPP) that those skilled in the art will be familiar with. These AI / ML features make use of trained AI / ML models to make one or more predictions or inferences, from a set of one or more input vectors, which can be used in the network (e.g., for improving the reliability or efficiency of communication in the network).

[0054] In respect of the communication system 1, for example, AI / ML models could potentially be trained and used for predicting the path of a UE 3 based on previous mobility of the UE 3, used for beam management, or used in methods of encoding and transmitting information. An AI / ML model may be hosted at a RAN node 5 (or any other suitable network node), and the RAN node 5 may perform control of communication resources for UEs 3 it serves, and / or perform control related to the status of the UE 3 (e.g. control of UE mobility, or control of a radio resource control, RRC, state of the UE 3) based on an inference (e.g. determination or prediction) generated using the AI / ML model. The RAN node 5 may also transmit an inference generated using the model to another node in the network, for use at the other node. An AI / ML model may also be hosted the UE 3, or at a plurality of locations within the network, for example at both the RAN node 5 and at the UE 3. For example, the RAN node 5 and the UE 3 may both make determinations and / or predictions using the same model or different models.

[0055] The support for such AI / ML features may involve different levels of collaboration between the network (RAN node 5 and / or core network 7) and the UE 3 served by the network when deploying and using such AI / ML features. For example, three possible 'network-UE collaboration levels' that may be supported are: 　　　　Level x: Involving no collaboration between the network and the UE 3. Specifically, level x is an implementation-based AI / ML operation without any dedicated AI / ML-specific enhancement. 　　　　Level y: Signalling-based collaboration without AI / ML model transfer. For example, this level is applicable when model training is performed offline, and models are registered to both the RAN node 5 and the UE 3. Here, the RAN node 5 and the UE 3 are aware of available models (before operation), and the RAN node 5 is only required to activate / deactivate the models residing at the UE 3 when needed. 　　　　Level z: Signalling-based collaboration with AI / ML model transfer (e.g., where an AI / ML model is transferred to the UE 3 when needed).

[0056] The AI / ML model types that are supported in the communication system 1 may include, for example: 　　　　Single-sided model: A single-sided AI / ML model is an AI / ML model that is deployed (hosted) only at the UE side or at the network side. For example, an AI / ML model may be hosted (stored, for generating inferences) at a UE 3, RAN node 5 (which may include either (or both) a RAN node CU 5-2CUand a RAN node DU 5-2DU), or a central entity of the communication system 1, or an operations, administration, and maintenance (OAM) / over-the-top (OTT) server. When the AI / ML model is used at the UE 3, the RAN node 5, a central entity of the communication system 1, or an OAM / OTT server only, the AI / ML model may be referred to as a 'single-sided' model. An example of this type of 'single-sided' model is an AI / ML model for beam prediction in time, which can be deployed at the UE side.

[0057] However, even when the model is a single-sided model, it will be appreciated that the model need not necessarily be trained at the node at which it is deployed (e.g., the UE 3 or the RAN node 5). For example, the model could be trained at the RAN node 5 (or at another node in the network such as a core network node / function - e.g., a central entity of the communication system 1, or an OAM / OTT server - and then transferred to the UE 3 for use at the UE 3.

[0058] Two-sided model: A 'two-sided' model is an AI / ML model (or model pair) that has one AI / ML model hosted at one node (e.g., the UE 3), and a corresponding AI / ML model hosted at another node (e.g., a RAN node 5 - which may include either (or both) a RAN node CU 5-2CUand a RAN node DU 5-2DU) - it will be appreciated that any pair of network nodes may be used. Such a two-sided model may also be referred to as a 'paired' AI / ML model. An inference using a two-sided model is performed jointly across the nodes at which the AI / ML models of the two-sided model are deployed. The joint inference may comprise, for example, a first part of the inference being performed at one node (e.g., the UE 3 or RAN node 5), and then the remaining part may be performed by the other (e.g., the RAN node 5 or UE 3). It will be appreciated that whilst the AI / ML model hosted at the different nodes may be the same AI / ML model, they need not necessarily be the same model. One example of this type of model is, for example only, channel state information (CSI) compression, where the UE 3 performs CSI compression and network performs CSI decompression. As with the single-sided model case, the two-sided model (or models) may be trained at any suitable network node, and then transmitted to the UE 3 and the RAN node 5 (or other respective node or nodes).

[0059] A general discussion of how AI / ML may be implemented in the communication system 1 will now be provided, by way of example only, with reference to Figs. 3 and 4.

[0060] Fig. 3 illustrates a functional framework for AI / ML models, and how various entities of the framework may interact with one another, that may be implemented in the communication system 1.

[0061] The entities include a data collection entity 341, a model training function 343, a model inference function 345, an actor 347, a management function 349, and a model storage entity 351.

[0062] The model storage entity 351 may be a reference point for protocol terminations for model transfer and delivery. The AI / ML models could be stored at any suitable node in the network.

[0063] The data collection entity 341 provides training data to the model training function 343, inference data to the model inference function 345, and monitoring data to the management function 349. The collected data may be, for example, data regarding mobility (e.g., handover of a UE 3, or a location of the UE 3). The data may be obtained, for example, by a UE 3 or a RAN node 5 (e.g., by receiving a measurement report from the UE 3, or by receiving data from another RAN node 5 or a core network node / function) and transmitted to another RAN node 5 or core network node that generates the AI / ML model inference output (or alternatively, the same RAN node that obtains the data may generate the AI / ML model output).

[0064] The model training function 343 performs the AI / ML model training, validation, and testing, and may generate model performance metrics as part of a model testing procedure. The model training function 343 may output a trained AI / ML model to the model storage entity 351 (though it will be appreciated that the output model may be stored at locations other than model storage entity 351).

[0065] The model inference function 345 provides AI / ML model inference output (e.g., predictions or decisions), and the actor 347 is a function or node that receives the output from the model inference function 345 and triggers or performs corresponding actions (e.g., the RAN node 5 that increases / reduces its transmit power or initiates a handover procedure for the UE 3). The AI / ML model inference output may be, for example, a prediction of mobility (e.g., expected path, route or trajectory, inter-cell, or inter-beam mobility, or expected handover) of the UE 3, or one or more parameters for use in encoding or decoding transmissions between the RAN node 5 and the UE 3. The model inference function 345 may receive an AI / ML model from the model storage entity 351, and inference data from the data collection entity 341 for use with the AI / ML model. The model inference function 345 may also output monitoring data for use at the management function 349 and receive information indicating an AI / ML to activate or deactivate from the management function 349.

[0066] The management function 349 receives monitoring data from the data collection entity 341 and may also receive monitoring data from the model inference function 345. The management function 349 may transmit to the model storage entity 351, an indication of an AI / ML model to be transmitted for use at the model inference function 345. The management function 349 may also transmit to the model training function 343, performance feedback or a retraining request for the AI / ML model.

[0067] The functions illustrated in Fig. 3 may be co-located at a single node of the communication system 1 (e.g., at the RAN node 5 or a core network node / function) or may be distributed amongst a plurality of network nodes (e.g., a plurality of the RAN nodes 5).

[0068] By way of example only, terms referred to by 3GPP in the context of this framework include: 　　　　AI / ML model training: A process to train an AI / ML Model [by learning the input / output relationship] in a data driven manner and obtain the trained AI / ML Model for inference. Model training can be performed offline or online or combination of both. 　　　　AI / ML model validation: A subprocess of training, to evaluate the quality of an AI / ML model using a dataset different from one used for model training, which helps selecting model parameters that generalize beyond the dataset used for model training. 　　　　AI / ML model testing: A subprocess of training, to evaluate the performance of a final AI / ML model using a dataset different from one used for model training and validation. Differently from AI / ML model validation, testing does not assume subsequent tuning of the model. 　　　　AI / ML model Inference: A process of using a trained AI / ML model to produce a set of outputs based on a set of inputs. 　　　　Data collection: A process of collecting data by the network nodes, management entity, or UE for the purpose of AI / ML model training, data analytics and inference. 　　　　Model monitoring: A procedure that monitors the inference performance of the AI / ML model. 　　　　Model activation: Enable an AI / ML model for a specific function. 　　　　Model deactivation: Disable an AI / ML model for a specific function. 　　　　Model switching: Deactivating a currently active AI / ML model and activating a different AI / ML model for a specific function. 　　　　Supervised learning: A process of training a model from input and its corresponding labels. 　　　　Unsupervised leaning: A process of training a model without labelled data. 　　　　Semi-supervised learning: A process of training a model with a mix of labelled data and unlabelled data. 　　　　Reinforcement Learning (RL): A process of training an AI / ML model from input (also referred to as 'state') and a feedback signal (also referred to as 'reward') resulting from the model's output (also referred to as 'action') in an environment the model is interacting with.

[0069] The data collection by the data collection entity 341 may be performed at various nodes of the communication system 1 (e.g., at one or more RAN nodes 5 or UEs 3). Particularly advantageous methods of obtaining, at the UE 3, data for an AI / ML model, and transmitting the AI / ML data from the UE 3 to the RAN node 5, will be described in more detail later.

[0070] Fig. 4 schematically illustrates a method of training an AI / ML model, and of monitoring the performance of the AI / ML model. As illustrated in Fig. 4, stored data / features may first be extracted in a data extraction step. In the data validation step, a determination of whether to proceed with training or retraining the AI / ML model is made (e.g., based on the extracted data). In the data preparation stage, the data is prepared for use in training the AI / ML model. For example, the data may be cleaned (e.g., filtered), subject to a transformation, or modified in any other suitable manner. The data may also be divided in training data, validation data and test data sets in the data preparation stage.

[0071] In the model training step, the AI / ML model is trained (or retrained) using training data prepared in the data preparation step. It will be appreciated that any suitable training method can be used to train the AI / ML model (e.g., a method that comprises supervised learning or unsupervised learning). In the model evaluation step, the AI / ML model is evaluated (e.g., a prediction accuracy of the AI / ML model is evaluated) using a test data set (which may be generated in the data preparation step). In the model validation step, a determination of whether the AI / ML model is suitable for deployment in the communication system 1 is made (e.g., based on the results of the model evaluation step).

[0072] In the model serving step, the AI / ML model is deployed for use in the communication system 1. AI / ML model deployment may comprise compiling a trained AI / ML model, packaging the model into an executable format, and delivering the AI / ML model to a target device. For example, the AI / ML model may be transmitted to the RAN node 5 and / or the UE 3, for use at the RAN node 5 and / or the UE 3 to generate predictions or determinations using the AI / ML model as part of a prediction service step, as illustrated in Fig. 4. In the performance monitoring step, the performance of the deployed AI / ML model is monitored. The predictive performance of the AI / ML model may be monitored by comparing predictions generated using the model with one or more measurements. For example, when the AI / ML model is used to predict a location of the UE 3, the prediction accuracy of the AI / ML model may be assessed using a measurement of an actual location of the UE 3. If the AI / ML model is used for predicting future measurement results (e.g., the measured Reference Signal Received Power (RSRP) of reference signals) at some point in time, the prediction accuracy of the AI / ML model may be assessed using actual measurement results acquired by the UE 3 when that point in time is reached. If the AI / ML model is used for determining parameters for use in encoding and decoding data transmitted between the RAN node 5 and the UE 3, the model may be assessed based on the performance of the encoding and / or decoding processes. In the retraining trigger step, retraining of the AI / ML model is triggered (e.g., because the prediction accuracy of the AI / ML model has fallen below an acceptable threshold accuracy, or because a performance of a method that uses inferences from the AI / ML model has fallen below an acceptable threshold performance), and the method returns to the data extraction step.

[0073] Each step of the method of Fig. 3 may be executed at a single node of the communication system 1 (including at the RAN node 5-1, the RAN node CU 5-2CU, and / or the RAN node DU 5-2DU(or at the UE 3)), or alternatively steps of the method may be distributed between a plurality of different nodes (or indeed one or more of these steps may be performed online or offline).

[0074] As discussed above with reference to Figs. 3 and 4, information collected by nodes / functions in the communication system 1 (e.g., at a UE 3 and / or RAN nodes 5) can be used as training data for an AI / ML model and used as inference data for use in generating one or more model inferences using the AI / ML model. The information used as training data, monitoring data, and / or to generate the one or more model inferences may be referred to as 'AI / ML information' or 'AI / ML data.'

[0075] < Configuration information for AI / ML > 　　Configuration information for an AI / ML model (which may be referred to as "AI / ML configuration information") may be exchanged between nodes in the communication network. For example, a core network node may transmit AI / ML configuration information to a RAN node 5 (or any other entity of the network that supports an AI / ML-based prediction functionality) that hosts an AI / ML model. The AI / ML configuration information may include a list of supported use cases for the AI / ML model (the AI / ML model need not necessarily be for predicting UE mobility). The supported use cases may include, for example: energy saving; traffic steering; anomaly detection; quality of experience (QoE) optimisation; mobility robustness optimisation (MRO); RAN slice service level agreement (SLA) assurance; massive multiple-input multiple-output (MIMO) beamforming optimisation; network slice subnet instance (NSSI) resource allocation; optimisation coverage and capacity optimisation (CCO); mobility load balancing (MLB); RACH optimisation; or UE transmission power optimisation. The AI / ML configuration information may include an indication of a particular AI / ML model to use for a particular use case. The AI / ML configuration information may also include an indication of whether feedback is required (e.g., from another network node). The feedback may include, for example, communication performance feedback (e.g., indicating a communication performance for communication between a UE 3 and a RAN node 5).

[0076] < AI / ML-implementation in Distributed RAN nodes > 　　While procedures and mechanisms have been developed for the implementation of single-sided and two-sided AI / ML models at UEs 3 and / or non-distributed RAN nodes 5-1 (e.g., RAN node 5-1) as described above, with the increased use of distributed RAN architectures there is a need to develop appropriate procedures and mechanisms to allow implementation of AI / ML models at distributed RAN nodes 5-2 (e.g., RAN node 5-2).

[0077] By way of example only, to allow implementation of AI / ML models at distributed RAN nodes 5-2 (e.g., RAN node 5-2), new appropriate procedures and mechanisms are required that facilitate: 　　-　　Real-time load management and performance optimization in the communication system 1 (e.g., to facilitate network slicing management, Coverage and Capacity Optimization (CCO), network energy savings (NES), beam management, and the like); and / or; 　　-　　AI / ML model inference function coordination between distributed components of the RAN architecture (e.g., between the RAN node CU 5-2CUand RAN node DU 5-2DU).

[0078] Therefore, as will be described in more detail later, the communication system 1 supports one or more enhancements to: allow for the implementation of AI / ML models at distributed RAN nodes 5-2; and / or facilitate real-time load management and performance optimization in the communication system 1, and AI / ML model inference function coordination between distributed components of the RAN architecture.

[0079] Specifically, the various communication entities of the communication system 1 are configured to support a number of different procedures for supporting AI / ML models in distributed RAN nodes 5-2 of a communication system 1, as will now be described with reference to Figs. 5 to 11.

[0080] It will be appreciated that whilst the UEs 3 and RAN nodes 5 of the described communication system 1 support a number of different procedures they could be configured to support only one, or a subset of, the procedures.

[0081] < AI / ML model inference generated at RAN Node DU: Model Training at an Operations, Administration and Maintenance (OAM) > 　　As mentioned above, the communication system 1 supports a number of different procedures for supporting the use of AI / ML models at distributed RAN nodes 5-2.

[0082] One such procedure will now be described with reference to Fig. 5, which depicts a simplified sequence diagram illustrating a procedure for generating an inference using an AI / ML mode hosted at a RAN node DU 5-2DUthat may be used in the communication system 1 of Fig. 1.

[0083] As shown in Fig. 5 there is provided a UE 3, a RAN node DU 5-2DU, a RAN node CU 5-2CU, and an OAM 14 deployed in the communication system 1 of Fig. 1.

[0084] It will be appreciated that while the procedure of Fig. 5 involves only a single RAN node DU 5-2DU, any number of RAN node DUs 5-2DUmay form part of the distributed RAN architecture. In that case the following procedure may be appropriately adapted to accommodate multiple RAN node DUs 5-2DUincluding, by way of example only, the deployment / updating of AI models hosted at a subset of one or more of the multiple RAN node DUs 5-2DU.

[0085] At step S502, the RAN node CU 5-2CUsends an appropriate measurement configuration message, to the UE 3, to configure the UE 3 with a specific measurement configuration. For example, the RAN node CU 5-2CUmay determine appropriate measurement information that the UE 3 should measure for reporting to the RAN node CU 5-2CUand may prepare (configure) an appropriate measurement configuration message for sending to the UE 3 to configure the UE 3 to measure that appropriate measurement information. By way of example only, appropriate measurement information may include cell / beam / slice-level UE measurements related to RSRP, Reference Signal Received Quality (RSRQ), and / or Signal to Interference and Noise Ratio (SINR) of a serving cell and / or one or more neighboring cells, or the like.

[0086] At step S504, having received the measurement configuration message at step S502, the UE 3 may perform the appropriate measurements indicated in the measurement configuration message. For example, the UE 3 may begin to perform cell / beam / slice-level UE measurements related to RSRP, RSRQ, and / or SINR of a serving cell and / or one or more neighboring cells. Having made those measurements, the UE 3, at step S506 reports those measurements to the RAN node CU 5-2CUvia one or more appropriate measurement reports (e.g., one or more layer-3 (L3) measurement reports of the like).

[0087] At step S508, the RAN node CU 5-2CUmay send input data to the OAM 14 to allow the OAM 14 to train an AI / ML model hosted (stored) at the OAM 14 using the input data. For example, the input data sent to the OAM 14 by the RAN node CU 5-2CUmay include one or more UE measurement reports that the RAN node CU 5-2CUreceived from the UE 3 at step S506. Alternatively, the RAN node CU 5-2CUmay send appropriate information, extracted from, or based on, the contents of one or more such UE measurement reports to the OAM 14 in another measurement-type report, or the like.

[0088] Additionally (or alternatively), the input data sent to the OAM 14 by the RAN node CU 5-2CUmay include other appropriate information needed to facilitate training of the AI / ML model hosted (stored) at the OAM 14.

[0089] While Fig. 5 shows input data being received by the OAM 14 from the RAN node CU 5-2CUonly, it will nevertheless be appreciated that the OAM 14 may receive data input from any number of other RAN nodes 5 in the communication system 1 that may be leveraged to provide information for training the AI / ML model at the OAM 14.

[0090] At step S510, the OAM 14, having received input data at step S508, trains the AI / ML model that it hosts in accordance with the procedures described above with reference to Figs. 3 and 4 by way of example.

[0091] At step S512, having trained the AI / ML model, the OAM 14 deploys the AI / ML model into the RAN node DU 5-2DU. For example, the OAM 14 may send an appropriate message or configuration to the RAN node DU 5-2DUthat includes the trained AI / ML model for storage at the RAN node DU 5-2DU. Alternatively, where an AI / ML model is already stored at the RAN node DU 5-2DU, the OAM 14 may send an appropriate message or configuration to the RAN node DU 5-2DUto update the AI / ML model stored at the RAN node DU 5-2DU.

[0092] At step S514, the UE 3 may send a latest set of measurement reports to the RAN node CU 5-2CU.For example, the measurement configuration message received at step S502, may indicate that the UE 3 is to perform appropriate measurements desired by the RAN node CU 5-2CU, and that those measurements are to be performed periodically or upon the UE 3 experiencing a measurement trigger. For example, the UE 3 may perform measurements related to the RSRP, the RSRQ, and / or the SINR of a serving cell and / or one or more neighboring cells periodically, or each time the UE 3 experiences a measurement trigger. Having made those measurements, the UE 3 reports those measurements to the RAN node CU 5-2CUvia one or more appropriate measurement reports.

[0093] At step S516, RAN node CU 5-2CUsends input data to the RAN node DU 5-2DUfor input to, and use with, the AI / ML model at the RAN node DU 5-2DUto generate model inferences. For example, the RAN node CU 5-2CUmay send, to the RAN node DU 5-2DU, the UE measurement reports that the RAN node CU 5-2CUreceived from the UE 3 at step S514. It will be appreciated that alternatively (or additionally), the RAN node CU 5-2CUmay send, to the RAN node DU 5-2DUusing one or more appropriate messages, appropriate information extracted from, or based on, the contents of one or more UE (e.g., L3) measurement reports received from the UE 3 at S514. Additionally (or alternatively), the input data sent to the RAN node DU 5-2DUby the RAN node CU 5-2CUmay include other appropriate information needed to facilitate the generation of a model inference using the AI / ML model.

[0094] At step S518, the UE 3 may send one or more other measurement reports to the RAN node DU 5-2DUas appropriate. For example, the UE 3 may send one or more Layer-1 (L1) and / or Layer-2 (L2) type measurement reports to the RAN node DU 5-2DUsuch as CSI reports and the like. Each measurement report may be sent to the RAN node DU 5-2DUin an appropriate control element (CE) such as a MAC CE.

[0095] At step S520, the RAN node DU 5-2DU, uses the locally deployed AI / ML model to generate a model inference based on local inputs and / or received inputs from the RAN node CU 5-2CU. The inference may, for example, be generated based on: local inputs in the form of information extracted from one or more UE measurement reports received at step S518; and / or other inputs (e.g., based on the input data received from the RAN node CU 5-2CUat S516).

[0096] By way of example only, the AI / ML model may, based on the nature of the inputs and the type of AI / ML model, output model inferences such as slice-level resource modifications, slice-level resource allocation strategies, mobility assistance information, cell coverage adjustments, future cell planning information, cell DTX / DRX activation / deactivation commands / triggers, cell / beam on / off time period indications, predicted beam patterns, beam switching commands / triggers, and / or the like.

[0097] Examples of the input data, and the outputs that may be generated by the AI / ML model are described in more detail below. In summary however, in one example, where the AI / ML model is a model used for network-slice resource allocation purposes such as the optimisation of slice-level physical resource block (PRB) usage and / or allocation of neighbor cells, the input data sent to the OAM 14 at step S508 (e.g., slice level PRB usage data, neighbor cell allocations data, and the like) may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as slice-level resource modifications, slice-level resource allocation strategies and / or mobility assistance information.

[0098] In another example, where the AI / ML model is a model used for coverage and capacity optimisation (CCO) to ensure appropriate cell coverage, cell load, and the like, the input data sent to the OAM 14 at step S508 (e.g., neighbor cell loads data, and the like) may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as cell coverage configurations, cell coverage adjustments, future cell planning information, and the like.

[0099] In yet another example, where the AI / ML model is a model used for achieving network energy savings (NES), the input data sent to the OAM 14 at step S508 (e.g., cell DTX / DRX configurations of neighbor cells, cell loads of neighbor cells) may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as cell DTX / DRX activation / deactivation, cell / beam on / off time period indications, and the like.

[0100] In yet another example, where the AI / ML model is a model used beam management procedures, and the like, the input data sent to the OAM 14 at step S508 (e.g., beam-level UE measurements, UE trajectory information, and the like) may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as predicted beam patterns, beam switching occasions, target beams, and the like.

[0101] At step S522, the RAN node DU 5-2DUmay (optionally) send an appropriate model performance feedback message or indication to the OAM 14 if applicable. For example, the RAN node DU 5-2DUmay send a feedback message or indication to the OAM 14 to indicate appropriate status updates to the OAM 14 and / or key performance indicators (KPIs) of the communication system 1 following the generation of the model inference to indicate the appropriateness of the generated model inference (i.e., an indication of the performance of the AI / ML).

[0102] Having generated the model inference, the RAN node DU 5-2DUuses that model inference to execute an appropriate action at step S524. The appropriate action may be an action performed at the RAN node DU 5-2DU. Additionally (or alternatively), the action may include sending appropriate messages and / or commands to the UE 3 and / or the RAN node CU 5-2CUfor the UE 3 and / or the RAN node CU 5-2CUto perform an appropriate action based on the model inference output by the AI / ML model.

[0103] For example, where the AI / ML model is used to enhance mobility strategies, the RAN node DU 5-2DUmay send appropriate messages / commands to the RAN node CU 5-2CUto assist the RAN node CU 5-2CUto select the most appropriate target cell for the UE 3 before it performs a handover procedure.

[0104] In other examples, based on the output of the AI / ML model, the RAN node DU 5-2DUmay perform any necessary action (including sending appropriate messages and / or commands the to the UE 3 and / or the RAN node CU 5-2CU) to e.g., modify a network-slice, modify RF parameters, modify cell coverage configurations, activate / deactivate cell DTX / DRX modes, perform beam refinement procedures, performing beam switching procedures, performing beam selection procedures, and the like.

[0105] At step S526, the RAN node DU 5-2DUsends an appropriate feedback message or indication to the OAM 14. For example, the RAN node DU 5-2DUmay send a feedback message or indication to the OAM 14 including model performance feedback (e.g., if not provided earlier) and / or other feedback related to an action performed by the RAN node DU 5-2DU(or an action performed by another node that the RAN node DU 5-2DUis made aware of).

[0106] < AI / ML model inference generated at RAN Node DU: Model Training at RAN Node CU > 　　As mentioned above, the communication system 1 supports a number of different procedures for supporting the use of AI / ML models at distributed RAN nodes 5-2.

[0107] One such procedure will now be described with reference to Fig. 6, which depicts simplified sequence diagram illustrating another procedure for generating an inference using an AI / ML mode hosted at a RAN node DU that may be used in the communication system 1 of Fig. 1.

[0108] As shown in Fig. 6 there is provided a UE 3, a RAN node DU 5-2DU, a RAN node CU 5-2CU, and another RAN node 5-1 deployed in the communication system 1 of Fig. 1. Optionally, the RAN node 5-1 may have a copy of an AI / ML model stored in its memory.

[0109] It will be appreciated that while the procedure of Fig. 6 involves only a single RAN node DU 5-2DU, any number of RAN node DUs 5-2DUmay form part of the RAN architecture. In that case procedure may be appropriately adapted to accommodate multiple RAN node DUs 5-2DUincluding, by way of example only, the deployment / updating of AI models hosted at a subset of one or more RAN node DUs 5-2DU.

[0110] At step S602, the RAN node CU 5-2CUsends an appropriate measurement configuration message to the UE 3 to configure the UE 3 with a specific measurement configuration. For example, the RAN node CU 5-2CUmay determine appropriate measurement information that the UE 3 should measure for reporting to the RAN node CU 5-2CUand may prepare (configure) an appropriate measurement configuration message for sending to the UE 3 to configure the UE 3 to measure that appropriate measurement information. By way of example only, appropriate measurement information that the RAN node CU 5-2CUmay determine the UE 3 should measure may include cell / beam / slice-level UE measurements related to RSRP, RSRQ, and / or SINR of a serving cell and / or one or more neighboring cells, or the like.

[0111] At step S604, having received the measurement configuration message at step S602, the UE 3 may perform the appropriate measurements indicated in the measurement configuration message. For example, the UE 3 may begin to perform cell / beam / slice-level UE measurements related to RSRP, RSRQ, and / or SINR of a serving cell and / or one or more neighboring cells. Having made those measurements, the UE 3, at step S606 reports those measurements to the RAN node CU 5-2CUvia one or more appropriate measurement reports (e.g., one or more layer-3 (L3) measurement reports of the like).

[0112] At step S608, the RAN node 5-1 may send RAN node-related input data to the RAN node CU 5-2CUfor use by the RAN node CU 5-2CUto train an AI / ML model hosted (stored) at the RAN node CU 5-2CU. For example, the input data sent to the RAN node CU 5-2CUby the RAN node 5-1 may include: cell / beam / slice-level UE measurements related to RSRP, RSRQ, and / or SINR of one or more neighboring cells provided by the RAN node 5-1. Additionally (or alternatively), the input data sent to the RAN node CU 5-2CUby the RAN node 5-1 may include other appropriate information needed to facilitate training of the AI / ML model hosted (stored) at the RAN node CU 5-2CU.

[0113] While Fig. 6 shows input data being received by the RAN node CU 5-2CUfrom one further RAN node 5-1, it will nevertheless be appreciated that the RAN node CU 5-2CUmay receive data input from any number of RAN nodes 5 in the communication system 1 that may be leveraged to provide information for training the AI / ML model at the RAN node CU 5-2CU.

[0114] At step S610, the RAN node CU 5-2CU, having received the UE measurement reports at step S606 and input data at step S608, trains the AI / ML model hosted at the RAN node CU 5-2CU. For example, it will be appreciated that the RAN node CU 5-2CUmay use either (or both) the input data received from other RAN nodes 5 (e.g., RAN node 5-1), and / or the information extracted from, or based on, the UE measurement reports the RAN node CU 5-2CUreceived from the UE 3, to train the AI / ML model.

[0115] At step S612, having trained the AI / ML model, the RAN node CU 5-2CUdeploys the AI / ML model into the RAN node DU 5-2DU. For example, the RAN node CU 5-2CUmay send an appropriate message or configuration to the RAN node DU 5-2DUthat includes the trained AI / ML model for storage at the RAN node DU 5-2DU. Alternatively, where an AI / ML model is already stored at the RAN node DU 5-2DU, the RAN node CU 5-2CUmay send an appropriate message or configuration to the RAN node DU 5-2DUto update the AI / ML model stored at the RAN node DU 5-2DU.

[0116] At step S613, the RAN node 5-1 may send a set of input data to the RAN node CU 5-2CUfor use by the AI / ML model for generating an inference. For example, the input data sent to the RAN node CU 5-2CUby the other RAN node 5-1 may include: cell / beam / slice-level UE measurements related to RSRP, RSRQ, and / or SINR of one or more neighboring cells provided by the RAN node 5-1.

[0117] At step S614, the UE 3 may send a latest set of measurement reports to the RAN node CU 5-2CU.For example, the measurement configuration message received at step S602, may indicate that the UE 3 is to perform appropriate measurements periodically or upon the UE 3 experiencing a measurement trigger. For example, the UE 3 may perform cell / beam / slice-level UE measurements related to RSRP, RSRQ, and / or SINR of a serving cell and / or one or more neighboring cells periodically or each time the UE 3 experiences a measurement trigger. Having made those measurements, the UE 3 reports those measurements to the RAN node CU 5-2CUvia one or more appropriate measurement reports.

[0118] At step S616, the RAN node CU 5-2CUsends input data to the RAN node DU 5-2DUfor input to, and use with, the AI / ML model at the RAN node DU 5-2DUto generate model inferences. For example, the RAN node CU 5-2CUmay send, to the RAN node DU 5-2DU, the UE measurement reports that the RAN node CU 5-2CUreceived from the UE 3 at step S614. Additionally (or alternatively) the RAN node CU 5-2CUmay forward, to the RAN node DU 5-2DU,the set of input data that it received from the RAN node 5-2 at step S613 for input to the AI / ML model hosted at RAN node DU 5-2DU, which may in turn be used by the AI / ML model to generate an AI / ML model inference. It will be appreciated that alternatively (or additionally), the RAN node CU 5-2CUmay send, to the RAN node DU 5-2DUusing one or more appropriate messages, appropriate information extracted from, or based on: the contents of one or more UE (e.g., L3) measurement reports received from the UE 3 at S614; and / or the set of input data received from the RAN node 5-2 at step S613. Additionally (or alternatively), the input data sent to the RAN node DU 5-2DUby the RAN node CU 5-2CUmay include other appropriate information needed to facilitate the generation of a model inference using the AI / ML model.

[0119] At step S618, the UE 3 may send one or more other measurement reports to the RAN node DU 5-2DUas appropriate. For example, the UE 3 may send one or more L1 and / or L2 type measurement reports to the RAN node DU 5-2DUsuch as CSI reports and the like. Each measurement report may be sent to the RAN node DU 5-2DUin an appropriate control element (CE) such as a MAC CE.

[0120] At step S620, the RAN node DU 5-2DU, uses the locally deployed AI / ML model to generate a model inference based on local inputs and / or received inputs from the RAN node CU 5-2CU. The inference may, for example, be generated based on: local inputs in the form of information extracted from one or more UE measurement reports received at step S618; and / or other inputs (e.g., based on the input data received from the RAN node CU 5-2CUat S616).

[0121] By way of example only, the AI / ML model may, based on the nature of the inputs and the type of AI / ML model, output model inferences such as slice-level resource modifications, slice-level resource allocation strategies, mobility assistance information, cell coverage adjustments, future cell planning information, cell DTX / DRX activation / deactivation commands / triggers, cell / beam on / off time period indications, predicted beam patterns, beam switching commands / triggers, and / or the like.

[0122] Examples of the nature of that input data, and the outputs that may be generated by the AI / ML model is described in more detail below. In summary however, in one example, where the AI / ML model is a model used for network-slice resource allocation purposes such as the optimisation of slice-level PRB usage and / or allocation of neighbor cells, the input data (e.g., input data sent at step S608 and / or (information from) one or more UE measurement reports sent at step S606) sent to the RAN node CU 5-2CU, and which may include, by way of example, slice level PRB usage data, neighbor cell allocations data, or the like, may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as slice-level resource modifications, slice-level resource allocation strategies and / or mobility assistance information.

[0123] In another example, where the AI / ML model is a model used for CCO to ensure appropriate cell coverage, cell load, and the like, the input data (e.g., input data sent at step S608 and / or (information from) one or more UE measurement reports sent at step S606) sent to the RAN node CU 5-2CU, and which may include, by way of example, neighbor cell load information, and the like, may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as cell coverage configurations, cell coverage adjustments, future cell planning information, and the like.

[0124] In yet another example, where the AI / ML model is a model used for achieving NES, the input data (e.g., input data sent at step S608 and / or (information from) one or more UE measurement reports sent at step S606) sent to the RAN node CU 5-2CU, and which may include, by way of example, cell DTX / DRX configurations of neighbor cells, cell loads of neighbor cells, and the like, may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as cell DTX / DRX activation / deactivation, cell / beam on / off time period indications, and the like.

[0125] In yet another example, where the AI / ML model is a model used beam management procedures, and the like, the input data (e.g., input data sent at step S608 and / or (information from) one or more UE measurement reports sent at step S606) sent to the RAN node CU 5-2CU, and which may include, by way of example, beam-level UE measurements, UE trajectory information, and the like, may be used to train the AI / ML model so that it may be used to generate appropriate outputs such as predicted beam patterns, beam switching occasions, target beams, and the like.

[0126] At step S622, the RAN node DU 5-2DUmay (optionally) send an appropriate model performance feedback message or indication to the RAN node CU 5-2CUif applicable. For example, the RAN node DU 5-2DUmay send a feedback message or indication to the RAN node CU 5-2CUto indicate appropriate status updates to the RAN node CU 5-2CUand / or KPIs of the communication system 1 following the generation of the model inference to indicate the appropriateness of the generated model inference (i.e., an indication of the performance of the AI / ML).

[0127] Having generated the model inference, the RAN node DU 5-2DUuses that model inference to execute an appropriate action at step S624. The appropriate action may be an action performed at the RAN node DU 5-2DU. Additionally (or alternatively), the action may include sending appropriate messages and / or commands to the UE 3 and / or the RAN node CU 5-2CU, and / or the RAN node 5-1 for the UE 3 and / or the RAN node CU 5-2CUand / or the RAN node 5-1 to perform an appropriate action based on the model inference output by the AI / ML model.

[0128] For example, where the AI / ML model is used to enhance mobility strategies, the RAN node DU 5-2DUmay send appropriate messages / commands to the RAN node CU 5-2CUto assist the RAN node CU 5-2CUto select the most appropriate target cell for the UE 3 before it performs a handover procedure.

[0129] In other examples, based on the output of the AI / ML model, the RAN node DU 5-2DUmay perform any necessary action (including sending appropriate messages and / or commands the to the UE 3 and / or the RAN node CU 5-2CU, and / or the RAN node 5-1) to e.g., modify a network-slice, modify RF parameters, modify cell coverage configurations, activate / deactivate cell DTX / DRX modes, perform beam refinement procedures, performing beam switching procedures, performing beam selection procedures, and the like.

[0130] At step S626, the RAN node DU 5-2DUsends an appropriate feedback message or indication to the RAN node CU 5-2CU. For example, the RAN node DU 5-2DUmay send a feedback message or indication to the RAN node CU 5-2CUincluding model performance feedback (e.g., if not provided earlier) and / or other feedback related to an action performed by the RAN node DU 5-2DU(or an action performed by another node that the RAN node DU 5-2DUis made aware of).

[0131] Furthermore, at step S628, the RAN node 5-1 may also send an appropriate feedback message or indication to the RAN node CU 5-2CU. For example, the RAN node 5-1 may send a feedback message or indication to the RAN node CU 5-2CUincluding model performance feedback (from the perspective of the RAN node 5-1) and / or other feedback related to an action performed by the RAN node 5-1 (or an action performed by another node that the RAN node 5-1is made aware of).

[0132] < Example AI / ML-based use cases > 　　It will be appreciated that the AI / ML models trained and used to generate associated inferences, in the procedures of Figs. 5 and 6, may be trained and used for many different use cases. The AI / ML models may, by way of example only, be trained and used for any of the following use cases: 　　　　AI / ML-based Network Slicing: In the use case of network slicing, the input data that is sent to the RAN node DU 5-2DUby the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface), and which is input to the AI / ML model may, by way of example, include slice level PRB usage information, information pertaining to the allocation of neighbour cells, and the like.

[0133] Based on those inputs (and any other inputs as described with reference to Figs. 5 and 6), the RAN node DU 5-2DUmay generate an AI / ML model inference, which may, by way of example, include messages, indications and / or commands relating to slice-level resource modifications, slice-level resource allocation strategies, slice-level resource status predications for a target cell, and the like. Additionally (or alternatively) the AI / ML model inference may include, by way of example, mobility assistance information.

[0134] In response to the inference, the RAN node DU 5-2DUmay execute any appropriate corresponding action. For example, the RAN node DU 5-2DUmay execute a network-slice modification such as a modification to a network slice's resource allocation. In another example, the RAN node DU 5-2DUmay execute a specific handover based on mobility assistance information inferred by the AI / ML model.

[0135] Where appropriate, an AI / ML model inference generated by the AI / ML model (and / or a command or message based on the AI / ML model inference) may also be sent (forwarded) to other nodes of the communication system 1. For example, where appropriate, the AI / ML model inference (and / or an associated command / message) may be sent, by the RAN node DU 5-2DU, to the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface) so that the RAN node CU 5-2CUmay execute an appropriate action / response to the inference. The RAN node CU 5-2CUmay, alternatively or additionally, send (forward) the AI / ML model inference (and / or an associated command / message) to another RAN node 5 (such as RAN node 5-1 in Fig. 6) to allow that RAN node 5 to perform an appropriate action / response. For example, the RAN node CU 5-2CU(and / or another RAN node 5 (such as RAN node 5-1 in Fig. 6)) may execute a network-slice modification such as a modification to a network slice's resource allocation, or the like.

[0136] Having received the inference from the AI / ML model and / or after having executed an appropriate corresponding action, the RAN node DU 5-2DUmay send appropriate AI / ML model performance feedback information to the OAM 14 (procedure of Fig. 5) or the RAN node CU 5-2CU(procedure of Fig. 6) as appropriate. For example, the RAN node DU 5-2DUmay send an appropriate message or indication to the OAM 14 or the RAN node CU 5-2CUto report slice-level resource status information and system KPIs to the RAN node CU 5-2CU. Based on that slice-level resource status information and system KPIs, the OAM 14 or the RAN node CU 5-2CUmay deduce the success or otherwise of the AI / ML model in generating an appropriate model inference and may thus make appropriate adjustments to its training of the AI / ML model.

[0137] AI / ML-based CCO: In the use case of CCO, the input data that is sent to the RAN node DU 5-2DUby the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface), and which is input to the AI / ML model may, by way of example, include neighbour cell load information, and the like.

[0138] Based on those inputs (and any other inputs as described with reference to Figs. 5 and 6), the RAN node DU 5-2DUmay generate an AI / ML model inference, which may, by way of example, include messages, indications and / or commands relating to cell coverage adjustments, future cell planning, and / or mobility assistance information, e.g., the number of UEs 3 that need to be offloaded from a cell 9.

[0139] In response to the inference, the RAN node DU 5-2DUmay execute any appropriate corresponding action. For example, the RAN node DU 5-2DUmay execute cell coverage modifications such as a modification RF parameters, and the like.

[0140] Where appropriate, an AI / ML model inference generated by the AI / ML model (and / or a command or message based on the AI / ML model inference) may also be sent (forwarded) to other nodes of the communication system 1. For example, where appropriate, the AI / ML model inference (and / or an associated command / message) may be sent, by the RAN node DU 5-2DU, to the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface) so that the RAN node CU 5-2CUmay execute an appropriate action / response to the inference. The RAN node CU 5-2CUmay, alternatively or additionally, send (forward) the AI / ML model inference (and / or an associated command / message) to another RAN node 5 (such as RAN node 5-1 in Fig. 6) to allow that RAN node 5 to perform an appropriate action / response. For example, the RAN node CU 5-2CU(and / or another RAN node 5 (such as RAN node 5-1 in Fig. 6)) may execute cell coverage modifications such as a modification RF parameters, and the like.

[0141] Having received the inference from the AI / ML model and / or after having executed an appropriate corresponding action, the RAN node DU 5-2DUmay send appropriate AI / ML model performance feedback information to the OAM 14 (procedure of Fig. 5) or the RAN node CU 5-2CU(procedure of Fig. 6) as appropriate. For example, the RAN node DU 5-2DUmay send an appropriate message or indication to the OAM 14 or the RAN node CU 5-2CUto report system KPIs to the RAN node CU 5-2CU. Based on the system KPIs, the OAM 14 or the RAN node CU 5-2CUmay deduce the success or otherwise of the AI / ML model in generating an appropriate model inference and may thus make appropriate adjustments to its training of the AI / ML model.

[0142] AI / ML-based NES: In the use case of NES, the input data that is sent to the RAN node DU 5-2DUby the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface), and which is input to the AI / ML model may, by way of example, include cell DTX / DRX configurations of neighbour cells, cell loads of neighbour cells, and the like.

[0143] Based on those inputs (and any other inputs as described with reference to Figs. 5 and 6), the RAN node DU 5-2DUmay generate an AI / ML model inference, which may, by way of example, include messages, indications and / or commands relating to cell DTX / DRX activation / deactivation, cell / beam on / off time period indications, and the like.

[0144] In response to the inference, the RAN node DU 5-2DUmay execute any appropriate corresponding action. For example, the RAN node DU 5-2DUmay execute the activation or deactivation of a cell DTX / DRX mode, the switching on and off of cells / beams, and the like.

[0145] Where appropriate, an AI / ML model inference generated by the AI / ML model and / or a command or message based on the AI / ML model inference) may also be sent (forwarded) to other nodes of the communication system 1. For example, where appropriate, the AI / ML model inference (and / or an associated command / message) may be sent, by the RAN node DU 5-2DU, to the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface) so that the RAN node CU 5-2CUmay execute an appropriate action / response to the inference. The RAN node CU 5-2CUmay, alternatively or additionally, send (forward) the AI / ML model inference (and / or an associated command / message) to another RAN node 5 (such as RAN node 5-1 in Fig. 6) to allow that RAN node 5 to perform an appropriate action / response. For example, the RAN node CU 5-2CU(and / or another RAN node 5 (such as RAN node 5-1 in Fig. 6)) may execute the activation or deactivation of a cell DTX / DRX mode, the switching on and off of cells / beams, and the like.

[0146] Having received the inference from the AI / ML model and / or after having executed an appropriate corresponding action, the RAN node DU 5-2DUmay send appropriate AI / ML model performance feedback information to the OAM 14 (procedure of Fig. 5) or the RAN node CU 5-2CU(procedure of Fig. 6) as appropriate. For example, the RAN node DU 5-2DUmay send an appropriate message or indication to the OAM 14 or the RAN node CU 5-2CUto report system KPIs to the RAN node CU 5-2CU. Based on the system KPIs, the OAM 14 or the RAN node CU 5-2CUmay deduce the success or otherwise of the AI / ML model in generating an appropriate model inference and may thus make appropriate adjustments to its training of the AI / ML model.

[0147] AI / ML-based Beam Management: In the use case of beam management, the input data that is sent to the RAN node DU 5-2DUby the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface), and which is input to the AI / ML model may, by way of example, include beam level UE measurements, UE trajectory information (e.g., UE location and UE speed), and the like.

[0148] Based on those inputs (and any other inputs as described with reference to Figs. 5 and 6), the RAN node DU 5-2DUmay generate an AI / ML model inference, which may, by way of example, include messages, indications and / or commands relating to predicted beam patterns, beam switching, mobility assistance information, and the like.

[0149] In response to those outputs, the RAN node DU 5-2DUmay execute any appropriate corresponding action. For example, the RAN node DU 5-2DUmay execute beam refinement procedures, beam selection procedures, beam switching procedures, and the like.

[0150] Where appropriate, an AI / ML model inference generated by the AI / ML model and / or a command or message based on the AI / ML model inference) may also be sent (forwarded) to other nodes of the communication system 1. For example, where appropriate, the AI / ML model inference (and / or an associated command / message) may be sent, by the RAN node DU 5-2DU, to the RAN node CU 5-2CUover an appropriate interface (e.g., an F1 interface) so that the RAN node CU 5-2CUmay execute an appropriate action / response to the inference. The RAN node CU 5-2CUmay, alternatively or additionally, send (forward) the AI / ML model inference (and / or an associated command / message) to another RAN node 5 (such as RAN node 5-1 in Fig. 6) to allow that RAN node 5 to perform an appropriate action / response. For example, the RAN node CU 5-2CU(and / or another RAN node 5 (such as RAN node 5-1 in Fig. 6))may execute beam refinement procedures, beam selection procedures, beam switching procedures, and the like.

[0151] Having received the inference from the AI / ML model and / or after having executed an appropriate corresponding action, the RAN node DU 5-2DUmay send appropriate AI / ML model performance feedback information to the OAM 14 (procedure of Fig. 5) or the RAN node CU 5-2CU(procedure of Fig. 6) as appropriate. For example, the RAN node DU 5-2DUmay send an appropriate message or indication to the OAM 14 or the RAN node CU 5-2CUto report system KPIs to the RAN node CU 5-2CU. Based on the system KPIs, the OAM 14 or the RAN node CU 5-2CUmay deduce the success or otherwise of the AI / ML model in generating an appropriate model inference and may thus make appropriate adjustments to its training of the AI / ML model.

[0152] Other AI / ML-based use cases: It will be appreciated that the use cases given herein are by way of example only and that the AI / ML model trained at the OAM 14 (Fig. 5) or the RAN node 5-1 (Fig. 6) and deployed to the RAN node DU 5-2DUmay be adapted for any number of different use cases. It will also be appreciated that depending on how the AI / ML model is adapted for different use cases will change the data inputs that need to be fed into the AI / ML model and the subsequent inferences output by the AI / ML model. Nevertheless, the skilled person will appreciate that the procedures of Figs. 5 and 6 are applicable irrespective of how the AI / ML model is adapted.

[0153] < AI / ML model inference generated at RAN Node CU (and / or some RAN Node DUs) > 　　In the procedures described above with reference to Figs. 5 and 6, the AI / ML model is always hosted at the RAN node DU 5-2DU, and thus AI / ML model inferences are generated by the RAN node DU 5-2DUusing the AI / ML model stored locally. However, it will nevertheless be appreciated that the procedures of Figs. 5 and 6 may be adapted such that the AI / ML model is hosted at the RAN node CU 5-2CUin addition to one or more of a set of RAN node DUs 5-2DU(e.g., RAN node DU 5-2DU).

[0154] It will also be appreciated that in a scenario where the AI / ML model is hosted at the RAN node CU 5-2CUand one or more of a set of RAN node DUs 5-2DU(e.g., RAN node DU 5-2DU), appropriate procedures and mechanisms may need to be implemented to facilitate co-ordination of the AI / ML model inferences generated by the RAN node CU 5-2CUand the set of RAN node DUs 5-2DU.

[0155] In the case that model inference is performed only at the RAN node DU 5-2DU, but not RAN node CU 5-2CU, the procedures described with reference to Figs. 5 and 6 are sufficient. In the case that model inference is performed at the RAN node CU 5-2CUand one or more RAN node DUs 5-2DU(e.g., RAN node DU 5-2DU), it is beneficial for the RAN node DU 5-2DUto indicate to the RAN node CU 5-2CUwhether it has model inference functionality. This model inference availability / capability may be indicated during a DU-CU interface setup (e.g., F1 setup) procedure and / or updated by a DU configuration update procedure (e.g., a gNB-DU configuration update procedure or the like).

[0156] There now follows a description of several different procedures and mechanisms for the coordination of AI / ML models at distributed RAN nodes 5-2 with reference to Figs. 7 and 8. It will be appreciated that these procedures are neither mutually exclusive, nor dependent upon one another - all, or a subset of one or more of the procedures may be implemented in the communication system 1.

[0157] < DU-CU Interface Setup and DU Configuration Update > 　　Fig. 7 is a simplified sequence diagram illustrating an intra-RAN, inter-node, information exchange procedure for AI / ML model inference availability related information that may be implemented in the communication system 1 of Fig. 1.

[0158] As seen in Fig. 7, the RAN node DU 5-2DUand RAN node CU 5-2CUare configured to engage in a dedicated setup procedure for exchanging application level data needed for the RAN node DU 5-2DUand the RAN node CU 5-2CUto correctly interoperate over the DU-CU ('F1') interface (end hence create a logical (e.g., 'F1') connection between the control planes of the RAN node CU 5-2CUand the RAN node DU 5-2DU).

[0159] As part of the dedicated setup procedure the RAN node DU 5-2DU, sends, at step S702, an appropriate setup request message (e.g., F1 Setup Request) for initiating the procedure. The setup request message may, by way of example only, include any appropriate information necessary for establishing communication via the (e.g., F1) interface between the RAN node DU 5-2DUand RAN node CU 5-2CU.

[0160] In addition, the setup request message may include an appropriate indication of whether the RAN node DU 5-2DUhas the capability to generate AI / ML model inferences (i.e., whether the RAN node DU 5-2DUhas an AI / ML model inference function) and / or whether the RAN node DU 5-2DUhas AI / ML model inferences available for sending to the RAN node CU 5-2CU. In other words, the setup request message may include an appropriate indication of whether the RAN node DU 5-2DUhas the capability to store and use an AI / ML model and / or whether the RAN node DU 5-2DUhas any AI / ML model inferences already stored at the RAN node DU 5-2DUthat it can share with the RAN node CU 5-2CU.

[0161] The RAN node CU 5-2CUmay acknowledge receipt of the setup request message, at S704, by sending an appropriate dedicated setup response message (e.g., a F1 Setup Response). Following completion of the setup procedure, communication via the DU-CU interface (e.g., F1 interface) may be established between the RAN node DU 5-2DUand the RAN node CU 5-2CU.

[0162] Following completion of the intra-RAN, inter-node, information exchange procedure of Fig. 7, the communication system 1 may perform either of the procedures described above with reference to Fig. 5 or Fig. 6 (or an adapted version thereof).

[0163] The procedure of Fig. 7 described above involves exchanging application level data needed for the RAN node DU 5-2DUand the RAN node CU 5-2CUto correctly interoperate over the DU-CU (e.g., F1) interface between a RAN node DU 5-2DUand a RAN node CU 5-2CU(e.g., an F1 interface), however it will also be appreciated that upon establishment of communication via that interface, procedures and mechanisms may also be required to update the RAN node CU 5-2CUif (and when) the capability of the RAN node DU 5-2DUto generate AI / ML model inferences (i.e., whether the RAN node DU 5-2DUhas an AI / ML model inference function) and / or whether the RAN node DU 5-2DUhas AI / ML model inferences available for sending to the RAN node CU 5-2CUchanges.

[0164] Such changes in the capability of the RAN node DU 5-2DUto generate AI / ML model inferences and / or changes in the availability of AI / ML model inferences at the RAN node DU 5-2DUmay, by way of example only, be communicated to the RAN node CU 5-2CUvia appropriate DU configuration update messages using another intra-RAN, inter-node, information exchange procedure comprising a DU configuration update procedure.

[0165] Fig. 8 is a simplified sequence diagram illustrating another intra-RAN, inter-node, information exchange procedure for AI / ML model inference availability related information that may be implemented in the communication system 1 of Fig. 1.

[0166] As seen in Fig. 8, the intra-RAN, inter-node, information exchange procedure comprises a DU configuration update procedure. This DU configuration update procedure may, for example, be a modified form of a conventional F1 application protocol (F1AP) based RAN node DU configuration update procedure. As those skilled in the art will appreciate, the DU configuration update configuration procedure is specified for communication system 1, for updating application-level configuration data needed for a RAN node DU 5-2DUand a RAN node CU 5-2CUto interoperate correctly on the appropriate established interface (e.g. the F1 interface).

[0167] In the DU configuration update procedure, the RAN node DU 5-2DUand RAN node CU 5-2CUare configured to communicate over an appropriate established interface (e.g., an F1 interface) between the RAN node DU 5-2DUand RAN node CU 5-2CU. It will be appreciated that the established interface may, for example, be established via the procedure described above with reference to Fig. 7.

[0168] In this example the DU configuration update procedure is configured to update the RAN node CU 5-2CUwith an AI / ML model inference generation capability / availability update. The DU configuration update procedure is initiated by the RAN node DU 5-2DU, sending, at step S802, an appropriate DU configuration update message (e.g., a gNB-DU configuration update message or the like) to update the RAN node CU 5-2CUwith associate information about the configuration of the RAN node DU 5-2DU. For example, the DU configuration update message sent to the RAN node CU 5-2CUmay include an appropriate indication of whether the RAN node DU 5-2DUhas the capability to generate AI / ML model inferences (i.e., whether the RAN node DU 5-2DUhas an AI / ML model inference function) and / or whether the RAN node DU 5-2DUhas AI / ML model inferences available for sending to the RAN node CU 5-2CU. In other words, DU configuration update message may include an appropriate indication of whether the RAN node DU 5-2DUhas the capability to store and use an AI / ML model and / or whether the RAN node DU 5-2DUhas an AI / ML model inferences already stored at the RAN node DU 5-2DUthat it can share with the RAN node CU 5-2CU.

[0169] The RAN node CU 5-2CUmay acknowledge receipt of the DU configuration message, at S804, by sending an appropriate dedicated DU configuration update acknowledgement message (e.g., a gNB-DU configuration acknowledge message or the like).

[0170] Following completion of the intra-RAN, inter-node, information exchange procedure of Fig. 8, the communication system 1 may perform either of the procedures described above with reference to Fig. 5 or Fig. 6 (or an adapted version thereof as described in more detail below).

[0171] It will be appreciated that if, during either the DU-CU interface setup procedure (and / or during a DU configuration update procedure) model inference capability / availability is indicated to be available at the RAN node DU 5-2DU, after an AI / ML model at the RAN node CU 5-2CU / OAM 14 is trained, the trained AI / ML model can be deployed / updated to the RAN node DU 5-2DU.

[0172] It will also be appreciated that, in a case where model inference capability is indicated to be available at the RAN node DU 5-2DU, to support model inference at the RAN node DU 5-2DU, the RAN node CU 5-2CUmay send corresponding input data to the RAN node DU 5-2DUfor the model inference, e.g., slice level resource status, beam status, etc.

[0173] Nevertheless, even when model inference capability is indicated to be available at the RAN node DU 5-2DU, the RAN node CU 5-2CUmay execute mobility optimisation-based actions related to model inferences by itself, albeit taking into account appropriate assistance information provided by the RAN node DU 5-2DU(e.g., the model inference output of the RAN node DU 5-2DU, or a message / command generated based on that model inference).

[0174] Moreover, as those skilled in the art will understand, in the event that the RAN node DU 5-2DUand RAN node CU 5-2CUeach generate a respective AI / ML model inference output (and / or an associated message and / or command based on such a AI / ML model inference), and the resulting AI / ML model inference outputs / messages / commands contradict one another (e.g. for radio resource allocation, cell on / off decision), then the RAN node CU 5-2CUdecision takes precedence (e.g., it is treated as having a higher priority).

[0175] < Disablement of RAN Node DU AI / ML Model Inference Forwarding: CU Disables > 　　Where the procedures of Figs. 7 and 8 are used to indicate to the RAN node CU 5-2CUthat the RAN node DU 5-2DUis capable of generating an AI / ML model inference or has an AI / ML model inference available for sending to the RAN node CU 5-2CU, the RAN node CU 5-2CUmay nevertheless decide it does not want to receive or use any AI / ML model inferences generated by the RAN node DU 5-2DU.

[0176] For example, where network energy savings wish to be achieved, or where there are power consumption issues associated with the communication system 1 (and more particularly the RAN node 5-2), the RAN node CU 5-2DUmay decide it does not want to receive or use any AI / ML model inferences generated by the RAN node DU 5-2DU.

[0177] Fig. 9 is simplified sequence diagram illustrating an intra-RAN, inter-node, information exchange procedure for AI / ML model inference generation disablement that may be implemented in the communication system 1 of Fig. 1.

[0178] It will be appreciated that the procedure of Fig. 9 may be performed following an indication that model inference capability / availability is available (e.g., using the procedure of Fig. 7 and / or Fig. 8 described).

[0179] As seen in Fig. 9, the intra-RAN, inter-node, information exchange procedure comprises a CU configuration update procedure. This CU configuration update procedure may, for example, be a modified form of a conventional F1 application protocol (F1AP) based RAN node CU configuration update procedure. As those skilled in the art will appreciate, the CU configuration update configuration procedure is specified for ,communication system 1, for updating application-level configuration data needed for a RAN node DU 5-2DUand a RAN node CU 5-2CUto interoperate correctly on an established appropriate interface (e.g. the F1 interface).

[0180] In the CU configuration update procedure, the RAN node DU 5-2DUand RAN node CU 5-2CUare configured to facilitate AI / ML model inference generation disablement at the RAN node DU 5-2DU.

[0181] For example, as part of the procedure, the RAN node CU 5-2CU, sends, at step S902, an appropriate CU configuration update message (e.g., a gNB-CU configuration update message or the like) to update the RAN node DU 5-2DUwith a configuration from the RAN node CU 5-2CU. For example, the CU configuration update message sent to the RAN node DU 5-2DUmay include an appropriate indication or command (e.g., a 'ModelInferenceDisbled' IE, or the like) to inform the RAN node DU 5-2DUto disable the generation of AI / ML model inferences.

[0182] Additionally, the DU-CU configuration update message sent to the RAN node DU 5-2DUmay (optionally) include an appropriate indication of the cause / reason for disabling the generation of AI / ML model inferences by the RAN node DU 5-2DU(e.g., a 'ModelInferenceDisbledCause' IE, or the like).

[0183] The RAN node DU 5-2DUmay acknowledge receipt of the DU-CU configuration message, at S904, by sending an appropriate dedicated CU configuration update acknowledgement message (e.g., a gNB-CU configuration acknowledge message or the like). Having sent the acknowledgement, the RAN node DU 5-2DUdisables generation of AI / ML model inferences using the AI / ML model it stores in its memory.

[0184] < Disablement of RAN Node DU AI / ML Model Inference Forwarding: DU Disables > 　　Where the procedures of Figs. 7 and 8 are used to indicate to the RAN node CU 5-2CUthat the RAN node DU 5-2DUis capable of generating an AI / ML model inference or has an AI / ML model inference available for sending to the RAN node CU 5-2CU, the RAN node DU 5-2DUmay nevertheless decide it does not want to generate or send any AI / ML model inferences to the RAN node CU 5-2CU.

[0185] For example, where network energy savings wish to be achieved, or where there are power consumption issues associated with the communication system 1 (and more particularly the RAN node 5-2), the RAN node DU 5-2CUmay decide it does not want to generate or send any AI / ML model inferences to the RAN node CU 5-2CU.

[0186] Fig. 10 is a simplified sequence diagram illustrating another intra-RAN, inter-node, information exchange procedure for AI / ML model inference generation disablement that may be implemented in the communication system 1 of Fig. 1.

[0187] It will be appreciated that the procedure of Fig. 10 may be performed following an indication that model inference capability / availability is available (e.g., using the procedure of Fig. 7 and / or Fig. 8 described).

[0188] As seen in Fig. 10, the intra-RAN, inter-node, information exchange procedure comprises a DU configuration update procedure. This DU configuration update procedure may, for example, be a modified form of a conventional F1 application protocol (F1AP) based RAN node DU configuration update procedure. As those skilled in the art will appreciate, the DU configuration update configuration procedure is specified for communication system 1, for updating application-level configuration data needed for a RAN node DU 5-2DUand a RAN node CU 5-2CUto interoperate correctly on an established appropriate interface (e.g. the F1 interface).

[0189] In the CU configuration update procedure, the RAN node DU 5-2DUand RAN node CU 5-2CUare configured facilitate AI / ML model inference generation disablement at the RAN node DU 5-2DU.

[0190] For example, as part of the procedure, the RAN node DU 5-2DU, sends, at step S1002, an appropriate DU configuration update message (e.g., a gNB-DU configuration update message or the like) to update a configuration of the RAN node CU 5-2CUwith a configuration from the RAN node DU 5-2DU. For example, the DU configuration update message sent to the RAN node CU 5-2CUmay include an appropriate indication or command (e.g., a 'ModelInferenceDisbled' IE, or the like) to inform the RAN node CU 5-2CUthat the RAN node DU 5-2DUhas, or wishes to, disable the generation of AI / ML model inferences at the RAN node DU 5-2DU.

[0191] Additionally, the DU configuration update message sent to the RAN node CU 5-2CUmay (optionally) include an appropriate indication of the cause / reason for the disabling the generation of AI / ML model inferences by the RAN node DU 5-2DU(e.g., a 'ModelInferenceDisbledCause' IE, or the like).

[0192] The RAN node CU 5-2CUmay acknowledge receipt of the DU-CU configuration message, at S1004, by sending an appropriate dedicated DU configuration update acknowledgement message (e.g., a gNB-DU configuration acknowledge message or the like). Having received an acknowledgement, the RAN node DU 5-2DUmay disable the generation of AI / ML model inferences using the AI / ML model it stores in its memory if it has not already done so.

[0193] Fig. 11 is a simplified sequence diagram illustrating another intra-RAN, inter-node, information exchange procedure for AI / ML model inference generation disablement that may be implemented in the communication system 1 of Fig. 1.

[0194] It will be appreciated that the procedure of Fig. 11 may be performed following an indication that model inference capability / availability is available (e.g., using the procedure of Fig. 7 and / or Fig. 8 described).

[0195] As seen in Fig. 11, the intra-RAN, inter-node, information exchange procedure comprises a DU status indication procedure. This DU status indication procedure may, for example, be a modified form of a conventional F1 application protocol (F1AP) based RAN node DU status indication procedure. As those skilled in the art will appreciate, the RAN node DU status indication procedure is specified for communication system 1, for informing a RAN node CU 5-2CUwhen one or more of its RAN node DUs 5-2DUis overloaded so that appropriate overload reduction mechanisms can be applied.

[0196] In the DU status indication procedure, the RAN node DU 5-2DUand RAN node CU 5-2CUare configured to facilitate AI / ML model inference generation disablement at the RAN node DU 5-2DU.

[0197] For example, as part of an AI / ML model inference generation disablement procedure, the RAN node DU 5-2DU, sends, at step S1102, an appropriate RAN node DU status indication (e.g., a gNB-DU status indication or the like), which may form part of e.g., a RAN node DU status indication message, or the like, to indicate a 'current' status of the RAN node DU 5-2DUto the RAN node CU 5-2CU.

[0198] For example, the RAN node DU status indication sent to the RAN node CU 5-2CU(sent in e.g., (e.g., a gNB-DU status indication or the like) may include, for example, an indication that the RAN node DU 5-2DUis overloaded and / or that the RAN node DU 5-2DUhas, or wishes to, disable the generation of AI / ML model inferences at the RAN node DU 5-2DU. It will be appreciated that where the RAN node DU status indication indicates that the RAN node DU 5-2DUis overloaded, the RAN node CU 5-2CUmay deduced from that indication that the RAN node DU 5-2DUhas, or wishes to, disable the generation of AI / ML model inferences at the RAN node DU 5-2DU.

[0199] Additionally, the RAN node DU status indication message sent to the RAN node CU 5-2CUmay (optionally) include an appropriate indication of the cause / reason for the disabling the generation of AI / ML model inferences by the RAN node DU 5-2DU(e.g., a 'ModelInferenceDisbledCause' IE, or the like).

[0200] < Devices of the Communication System > 　　< User Equipment > 　　Fig. 12 is a schematic block diagram illustrating the main components of a UE 3 as shown in Fig. 1.

[0201] As shown, the UE 3 has a transceiver circuit 31 that is operable to transmit signals to and to receive signals from a RAN node 5 via one or more antennas 33 (e.g., comprising one or more antenna elements). The UE 3 has a controller 37 to control the operation of the UE 3. The controller 37 is associated with a memory 39 and is coupled to the transceiver circuit 31. Although not necessarily required for its operation, the UE 3 might, of course, have all the usual functionality of a conventional UE 3 (e.g., a user interface 35, such as a touch screen / keypad / microphone / speaker and / or the like for, allowing direct control by and interaction with a user) and this may be provided by any one or any combination of hardware, software, and firmware, as appropriate. Software may be pre-installed in the memory 39 and / or may be downloaded via the communication system 1 or from a removable data storage device (RMD), for example.

[0202] The controller 37 is configured to control overall operation of the UE 3 by, in this example, program instructions or software instructions stored within memory 39. As shown, these software instructions include, among other things, an operating system 41, and a communication control module 43.

[0203] The communication control module 43 is operable to control the communication between the UE 3 and its serving RAN node or RAN nodes 5 (and other communication devices connected to the RAN node 5, such as further UEs and / or core network nodes). The communication control module 43 is configured for the overall handling of uplink communications via associated uplink channels (e.g., via a physical uplink control channel (PUCCH), random access channel (RACH), and / or a physical uplink shared channel (PUSCH)) including both dynamic and semi-static signalling (e.g., SRS). The communication control module 43 is also configured for the overall handling of receipt of downlink communications via associated downlink channels (e.g., of DCI via a physical downlink control channel (PDCCH) and / or a physical downlink shared channel (PDSCH)) including both dynamic and semi-persistent scheduling (e.g., SPS). The communication control module 43 is responsible, for example: for determining where to monitor for downlink control information; for determining the resources to be used by the UE 3 for transmission / reception of UL / DL communications (including interleaved resources and resources subject to frequency hopping); for managing frequency hopping at the UE side; for determining how slots / symbols are configured (e.g., for UL, DL or full duplex communication, or the like); for determining which bandwidth parts are configured for the UE 3; for determining how uplink transmissions should be encoded and the like.

[0204] It will be appreciated that the communication control module 43 may include a number of sub-modules ('layers' or 'entities') to support specific functionalities. For example, the communication control module 43 may include a PHY sub-module, a MAC sub-module, an RLC sub-module, a PDCP sub-module, an RRC sub-module, etc.

[0205] The communication control module 43 is configured, in particular, to control the UE's communications, where applicable, in accordance with any of the methods described herein.

[0206] < RAN node (non-distributed) > 　　Fig. 13 is a schematic block diagram illustrating the main components of a RAN node 5-1 for the communication system 1 shown in Fig. 1. As shown, the RAN node 5-1 has a transceiver circuit 51 for transmitting signals to and for receiving signals from the communication devices (such as UEs 3) via one or more antennas 53 (e.g., a single or multi-panel antenna array / massive antenna), and a core network interface 55 (e.g., comprising the N2, N3 and other reference points / interfaces) for transmitting signals to and for receiving signals from network nodes in the core network 7. Although not shown, the RAN node 5-1 may also be coupled to other RAN nodes via an appropriate interface (e.g., the so-called 'Xn' interface in NR). The RAN node 5-1 has a controller 57 to control the operation of the RAN node 5-1. The controller 57 is associated with a memory 59. Software may be pre-installed in the memory 59 and / or may be downloaded via the communication system 1 or from a removable data storage device (RMD), for example. The controller 57 is configured to control the overall operation of the RAN node 5-1 by, in this example, program instructions or software instructions stored within memory 59.

[0207] As shown, these software instructions include, among other things, an operating system 61, and a communication control module 63.

[0208] The communication control module 63 is operable to control the communication between the RAN node 5-1 and UEs 3 and other network entities that are connected to the RAN node 5-1. The communication control module 63 is configured for the overall control of the reception and decoding of uplink communications, via associated uplink channels (e.g., via a physical uplink control channel (PUCCH), a random-access channel (RACH), and / or a physical uplink shared channel (PUSCH)) including both dynamic and semi-static signalling (e.g., SRS). The communication control module 63 is also configured for the overall handling the transmission of downlink communications via associated downlink channels (e.g., via a physical downlink control channel (PDCCH) and / or a physical downlink shared channel (PDSCH)) including both dynamic and semi-static signalling (e.g., CSI-RS, SSBs etc.). The communication control module 63 is also responsible, for example, for determining and scheduling the resources to be used by the UE 3 for receiving in DL / transmitting in UL, for configuring slots / symbols appropriately (e.g., for UL, DL, flexible, full duplex communication, or the like), for configuring one or more bandwidth parts for the UE 3, and for providing related configuration signalling to the UE 3.

[0209] It will be appreciated that the communication control module 63 may include a number of sub-modules (or 'layers') to support specific functionalities. For example, the communication control module 63 may include a PHY sub-module, a MAC sub-module, an RLC sub-module, a PDCP sub-module, an SDAP sub-module, an IP sub-module, an RRC sub-module, etc.

[0210] The communication control module 63 is configured, in particular, to control the RAN Node's communications, where applicable, in accordance with any of the methods described herein.

[0211] < RAN node (distributed) > 　　Fig. 14 is a simplified block schematic illustrating the main components of a distributed RAN node 5-2 comprising a distributed type of base station for implementation in the system of Fig. 1. As shown, the RAN node 5-2 includes a central unit (RAN node CU) 5-2CUand a distributed unit (RAN node DU) 5-2DU(although it may include other RAN node DUs 5-2DUas described above). Each unit 5-2CU, 5-2DUincludes respective transceiver circuitry 51c, 51d.

[0212] The transceiver circuitry 51d of the distributed unit 5-2DUis operable to transmit signals to and to receive signals from UEs 3 via an air interface 53d and one or more antennas and is also operable to transmit signals to and to receive signals from the central unit 5-2CUvia an interface, for example the distributed unit side of an F1 interface (which may be provided over a satellite radio interface).

[0213] The transceiver circuitry 51c of the central unit 5-2CUis operable to transmit signals to and to receive signals from functions of the core network 7 and / or other RAN nodes via a network interface 55c. The network interface 55c typically includes an N2 and / or N3 interfaces for communicating with the core network 7 and a RAN node to RAN node (e.g., Xn) interface for communicating with other RAN nodes 5. The transceiver circuitry 51c of the central unit 5-2CUis also operable to transmit signals to and to receive signals from one or more distributed units 5-2DU, for example the central unit side of the F1 interface provided.

[0214] Each unit 5-2CU, 5-2DUincludes a respective controller 57c, 57d which controls the operation of the corresponding transceiver circuitry 51c, 51d in accordance with software stored in the respective memories 59c and 59d of the central unit 5-2CUand the distributed unit 5-2CU. The software of each unit may be pre-installed in the memory 59c, 59d and / or may be downloaded via the communication system 1 or from a removable data storage device (RMD), for example. The software of each unit includes, among other things, a respective operating system 61c, 61d, and a respective communication control module 63c, 63d.

[0215] Each communication control module 63c, 63d is operable to control the communication of its corresponding unit 5-2CU, 5-2DUincluding the communication from one unit to the other. The communication control module 63d of the distributed unit 5-2DUcontrols communication between the distributed unit 5-2DUand the UEs 3, and the communication control module 63c of the central unit 5-2CUcontrols communication between the central unit 5-2CUand other network entities that are connected to the distributed RAN node 5-2.

[0216] The communication control modules 63c, 63d also respectively control the part played by the central unit 5-2CUand distributed unit 5-2DUin the flow of uplink and downlink user traffic and control data to be received from and transmitted to the communications devices served by the RAN node 5-2 including, for example, control data for managing operation of the UEs 3. Each communication control module 63c, 63d is responsible, for example, for controlling the respective part played by the central unit 5-2CUand distributed unit 5-2DUin the reception and decoding of uplink communications, via associated uplink channels (e.g., via a physical uplink control channel (PUCCH), a random-access channel (RACH), and / or a physical uplink shared channel (PUSCH)) including both dynamic and semi-static signalling (e.g., SRS). Each communication control module 63c, 63d is responsible, for example, for controlling the respective part played by the central unit 5-2CUand distributed unit 5-2DUin the overall handling the transmission of downlink communications via associated downlink channels (e.g., via a physical downlink control channel (PDCCH) and / or a physical downlink shared channel (PDSCH)) including both dynamic and semi-static signalling (e.g., CSI-RS, SSBs etc.). Each communication control module 63c, 63d is responsible, for example, for controlling the respective part played by the central unit 5-2CUand distributed unit 5-2DUin determining and scheduling the resources to be used by the UE 3 for receiving in DL / transmitting in UL, for configuring slots / symbols appropriately (e.g., for UL, DL, flexible, full duplex communication, or the like), for configuring one or more bandwidth parts for the UE 3, and for providing related configuration signalling to the UE 3.

[0217] It will be appreciated that each communication control module 63c, 63d may include a number of sub-modules (or 'layers') to support specific functionalities supported by the by the central unit 5-2CUand distributed unit 5-2DU. For example, a communications PHY sub-module, a MAC sub-module, an RLC sub-module, a PDCP sub-module, an SDAP sub-module, an IP sub-module, an RRC sub-module, etc may be distributed between the central unit 5-2CUand distributed unit 5-2DUappropriately depending on where the functional split is configured between the central unit 5-2CUand distributed unit 5-2DU.

[0218] Each communication control module 63c, 63d is configured, in particular, to control the respective communications of the central unit 5-2CUand distributed unit 5-2DU, where applicable, in accordance with any of the methods described herein.

[0219] < Modifications and Alternatives > 　　Detailed examples been described above. As those skilled in the art will appreciate, a number of modifications and alternatives can be made to the above examples whilst still benefiting from the concepts embodied therein.

[0220] It will be appreciated that description of features of and actions performed by a RAN node (base station), apply equally to distributed type base stations as to non-distributed type base stations.

[0221] It will also be appreciated that whilst information elements having specific names have been described differently named information elements but having a similar purpose may be used.

[0222] In the above description the UE and the base station are described for ease of understanding as having a number of discrete functional components or modules. Whilst these modules may be provided in this way for certain applications, for example where an existing system has been modified to implement the disclosed enhancements, in other applications, for example in systems designed with the inventive features in mind from the outset, these modules may be built into the overall operating system or code and so these modules may not be discernible as discrete entities.

[0223] In the above examples, a number of software modules were described. As those skilled in the art will appreciate, the software modules may be provided in compiled or un-compiled form and may be supplied to the UE or base station as a signal over a computer network, or on a recording medium. Further, the functionality performed by part, or all, of this software may be performed using one or more dedicated hardware circuits. However, the use of software modules is preferred as it facilitates the updating of the UE or the base station in order to update their functionalities.

[0224] Each controller may comprise any suitable form of processing circuitry including (but not limited to), for example: one or more hardware implemented computer processors; microprocessors; central processing units (CPUs); arithmetic logic units (ALUs); input / output (IO) circuits; internal memories / caches (program and / or data); processing registers; communication buses (e.g. control, data and / or address buses); direct memory access (DMA) functions; hardware or software implemented counters, pointers and / or timers; and / or the like. Various other modifications will be apparent to those skilled in the art and will not be described in further detail here.

[0225] The User Equipment (or "UE," "mobile station," "mobile device" or "wireless device") in the present disclosure is an entity connected to a network via a wireless interface. It should be noted that the present disclosure is not limited to a dedicated communication device and can be applied to any device having a communication function as explained in the following paragraphs.

[0226] The terms "User Equipment" or "UE" (as the term is used by 3GPP), "mobile station", "mobile device", and "wireless device" are generally intended to be synonymous with one another, and include standalone mobile stations, such as terminals, cell phones, smart phones, tablets, cellular IoT devices, IoT devices, and machinery. It will be appreciated that the terms "mobile station" and "mobile device" also encompass devices that remain stationary for an extended period of time.

[0227] A UE may, for example, be an item of equipment for production or manufacture and / or an item of energy related machinery (for example equipment or machinery such as: boilers; engines; turbines; solar panels; wind turbines; hydroelectric generators; thermal power generators; nuclear electricity generators; batteries; nuclear systems and / or associated equipment; heavy electrical machinery; pumps including vacuum pumps; compressors; fans; blowers; oil hydraulic equipment; pneumatic equipment; metal working machinery; manipulators; robots and / or their application systems; tools; moulds or dies; rolls; conveying equipment; elevating equipment; materials handling equipment; textile machinery; sewing machines; printing and / or related machinery; paper converting machinery; chemical machinery; mining and / or construction machinery and / or related equipment; machinery and / or implements for agriculture, forestry and / or fisheries; safety and / or environment preservation equipment; tractors; precision bearings; chains; gears; power transmission equipment; lubricating equipment; valves; pipe fittings; and / or application systems for any of the previously mentioned equipment or machinery etc.).

[0228] A UE may, for example, be an item of transport equipment (for example transport equipment such as: rolling stocks; motor vehicles; motorcycles; bicycles; trains; buses; carts; rickshaws; ships and other watercraft; aircraft; rockets; satellites; drones; balloons etc.).

[0229] A UE may, for example, be an item of information and communication equipment (for example information and communication equipment such as: electronic computer and related equipment; communication and related equipment; electronic components etc.).

[0230] A UE may, for example, be a refrigerating machine, a refrigerating machine applied product, an item of trade and / or service industry equipment, a vending machine, an automatic service machine, an office machine or equipment, a consumer electronic and electronic appliance (for example a consumer electronic appliance such as: audio equipment; video equipment; a loud speaker; a radio; a television; a microwave oven; a rice cooker; a coffee machine; a dishwasher; a washing machine; a dryer; an electronic fan or related appliance; a cleaner etc.).

[0231] A UE may, for example, be an electrical application system or equipment (for example an electrical application system or equipment such as: an x-ray system; a particle accelerator; radio isotope equipment; sonic equipment; electromagnetic application equipment; electronic power application equipment etc.).

[0232] A UE may, for example, be an electronic lamp, a luminaire, a measuring instrument, an analyser, a tester, or a surveying or sensing instrument (for example a surveying or sensing instrument such as: a smoke alarm; a human alarm sensor; a motion sensor; a wireless tag etc.), a watch or clock, a laboratory instrument, optical apparatus, medical equipment and / or system, a weapon, an item of cutlery, a hand tool, or the like.

[0233] A UE may, for example, be a wireless-equipped personal digital assistant or related equipment (such as a wireless card or module designed for attachment to or for insertion into another electronic device (for example a personal computer, electrical measuring machine)).

[0234] A UE may be a device or a part of a system that provides applications, services, and solutions described below, as to "internet of things (IoT)," using a variety of wired and / or wireless communication technologies.

[0235] Internet of Things devices (or "things") may be equipped with appropriate electronics, software, sensors, network connectivity, and / or the like, which enable these devices to collect and exchange data with each other and with other communication devices. IoT devices may comprise automated equipment that follow software instructions stored in an internal memory. IoT devices may operate without requiring human supervision or interaction. IoT devices might also remain stationary and / or inactive for an extended period of time. IoT devices may be implemented as a part of a (generally) stationary apparatus. IoT devices may also be embedded in non-stationary apparatus (e.g., vehicles) or attached to animals or persons to be monitored / tracked.

[0236] It will be appreciated that IoT technology can be implemented on any communication devices that can connect to a communication system for sending / receiving data, regardless of whether such communication devices are controlled by human input or software instructions stored in memory.

[0237] It will be appreciated that IoT devices are sometimes also referred to as Machine-Type Communication (MTC) devices or Machine-to-Machine (M2M) communication devices. It will be appreciated that a UE may support one or more IoT or MTC applications. Some examples of MTC applications are listed in the following table. This list is not exhaustive and is intended to be indicative of some examples of machine type communication applications.

[0238] Table 3

[0239] Further, the above-described UE categories are merely examples of applications of the technical ideas and exemplary examples described in the present document. Needless to say, these technical ideas and examples are not limited to the above-described UE and various modifications can be made thereto.

[0240] Various other modifications will be apparent to those skilled in the art and will not be described in further detail here.

[0241] Although the present disclosure has been described with reference to the example embodiments, the present disclosure is not limited to the above. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the disclosure.

[0242] This application is based upon and claims the benefit of priority from UK patent application No. 2409200.9, filed on June 26, 2024, the disclosure of which is incorporated herein in its entirety by reference.

[0243] The program can be stored and provided to the computer device using any type of non-transitory computer readable media. Non-transitory computer readable media include any type of tangible storage media. Examples of non-transitory computer readable media include magnetic storage media (such as floppy disks, magnetic tapes, hard disk drives, etc.), optical magnetic storage media (e.g. magneto-optical disks), CD-ROM (Read Only Memory), CD-R, CD-R / W, and semiconductor memories (such as mask ROM, PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, RAM (Random Access Memory), etc.). The program may be provided to the computer device using any type of transitory computer readable media. Examples of transitory computer readable media include electric signals, optical signals, and electromagnetic waves. Transitory computer readable media can provide the program to the computer device via a wired communication line, such as electric wires and optical fibers, or a wireless communication line.

[0244] For example, the whole or part of the example embodiments disclosed above can be described as, but not limited to, the following supplementary notes. (Supplementary note 1) 　　A method performed by a distributed unit of an access network node, the method comprising: 　　receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and 　　inferring the AI / ML model. (Supplementary note 2) 　　The method according to supplementary note 1, further comprising: 　　receiving, from a mobile device, measurement results, and 　　wherein the inferring the AI / ML model is performed based on the measurement results. (Supplementary note 3) 　　The method according to supplementary note 1 or 2, further comprising: 　　receiving, from the central unit or an operations, administration and maintenance (OAM) function, updated AI / ML model, and 　　wherein the inferring is performed based on the updated AI / ML model. (Supplementary note 4) 　　The method according to supplementary note 3, wherein 　　the updated AI / ML model is trained by the OAM function, based on input data from the central unit. (Supplementary note 5) 　　The method according to supplementary note 3, wherein 　　the updated AI / ML model is trained by the central unit, based on input data from another access network node. (Supplementary note 6) 　　The method according to any one of supplementary notes 3 to 5, further comprising: 　　transmitting, to the central unit or the OAM function, feedback of performance of the AI / ML model. (Supplementary note 7) 　　The method according to supplementary note 6, wherein 　　the feedback includes at least one of: 　　　　user equipment (UE) performance; 　　　　information indicating system key performance indicators (KPIs); or 　　　　information indicating resource status. (Supplementary note 8) 　　The method according to any one of supplementary notes 1 to 7, further comprising; 　　performing an action corresponding to output of the inferring the AI / ML model. (Supplementary note 9) 　　The method according to supplementary note 8, wherein 　　the output of the inferring the AI / ML model includes at least one of: 　　　　information indicating slice-level resource modification / allocation; 　　　　information for mobility assistance of a mobile device; 　　　　information indicating cell coverage configuration; 　　　　information indicating coverage adjustment of cells; 　　　　information indicating future cell planning; 　　　　information indicating cell discontinuous transmission (DTX) / discontinuous reception (DRX) activation / deactivation; 　　　　information indicating cell / beam on / off period; or 　　　　information indicating predicted beam pattern or beam switching. (Supplementary note 10) 　　The method according to supplementary note 8 or 9, wherein 　　the action includes at least one of: 　　　　network slice modification; 　　　　cell coverage adjustment; 　　　　activation / deactivation of discontinuous transmission (DTX) / discontinuous reception (DRX); 　　　　cell / beam on / off; 　　　　beam refinement; 　　　　beam selection; or 　　　　beam switching. (Supplementary note 11) 　　The method according to any one of supplementary notes 8 to 10, further comprising: 　　transmitting feedback of the action. (Supplementary note 12) 　　The method according to supplementary note 11, wherein 　　the feedback includes at least one of: 　　　　user equipment (UE) performance; 　　　　information indicating slice-level resource status; 　　　　information indicating system key performance indicators (KPIs); or 　　　　information indicating resource status. (Supplementary note 13) 　　The method according to any one of supplementary notes 1 to 12, wherein 　　the input data for inferring the AI / ML model includes at least one of: 　　　　information indicating slice level resource usage or allocation of neighbour cells; 　　　　information indicating load of neighbour cells; 　　　　information indicating configuration of cell discontinuous transmission (DTX) / discontinuous reception (DRX); or 　　　　information indicating at least one of: 　　　　　　beam level measurements of a mobile device, 　　　　　　trajectory of the mobile device, or 　　　　　　location and speed of the mobile device. (Supplementary note 14) 　　The method according to any one of supplementary notes 1 to 13, further comprising: 　　transmitting, to the central unit, information indicating whether the distributed unit is able to infer the AI / ML model, and 　　the inferring is performed in a case where the distributed unit is able to infer the AI / ML model. (Supplementary note 15) 　　The method according to supplementary note 14, wherein 　　the information is transmitted in a central unit-distributed unit interface setup procedure or a distributed unit configuration update procedure. (Supplementary note 16) 　　The method according to any one of supplementary notes 1 to 15, further comprising: 　　receiving, from the central unit, information for requesting the central unit to disable to infer the AI / ML model; and 　　determining to disable to infer the AI / ML model. (Supplementary note 17) 　　The method according to any one of supplementary notes 1 to 16, further comprising: 　　determining to disable to infer the AI / ML model; and 　　transmitting, to the central unit, information indicating that the distributed unit disables to infer the AI / ML model. (Supplementary note 18) 　　A method performed by a central unit of an access network node, the method comprising: 　　transmitting, to a distributed unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model, wherein 　　the input data is used by the distributed unit for use in inferring the AI / ML model. (Supplementary note 19) 　　A distributed unit of an access network node, the distributed unit comprising: 　　means for receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and 　　means for inferring the AI / ML model. (Supplementary note 20) 　　A central unit of an access network node, the central unit comprising: 　　means for transmitting, to a distributed unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model, wherein 　　the input data is used by the distributed unit for use in inferring the AI / ML model.

[0245] 1　　communication system 3, 3-1, 3-2, 3-3　　UEs 5, 5-1, 5-2　　radio access network (RAN) node 5-2CU　　RAN node CU 5-2DU　　RAN node DU 7　　core network 9, 9-1, 9-2　　cells 10　　control plane functions (CPFs) 10-1　　access and mobility management Functions (AMFs) 10-2　　session management function (SMF) 11　　user plane functions (UPFs) 14　　Operations, Administration and Maintenance (OAM) 20　　external data network 31, 51, 51c, 51d　　transceiver circuit 33, 53, 53d　　antenna 35　　user interface 37, 57　　controller 57c　　CU controller 57d　　DU controller 39, 59　　memory 59c　　CU memory 59d　　DU memory 41, 61　　operating system 61c　　CU operating system 61d　　DU operating system 43, 63　　communications control module 63c　　CU communications control module 63d　　DU communications control module 55　　core network interface 55c　　network interface

Claims

1. A method performed by a distributed unit of an access network node, the method comprising: 　　receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and 　　inferring the AI / ML model.

2. The method according to claim 1, further comprising: 　　receiving, from a mobile device, measurement results, and 　　wherein the inferring the AI / ML model is performed based on the measurement results.

3. The method according to claim 1 or 2, further comprising: 　　receiving, from the central unit or an operations, administration and maintenance (OAM) function, updated AI / ML model, and 　　wherein the inferring is performed based on the updated AI / ML model.

4. .　　The method according to claim 3, wherein 　　the updated AI / ML model is trained by the OAM function, based on input data from the central unit.

5. The method according to claim 3, wherein 　　the updated AI / ML model is trained by the central unit, based on input data from another access network node.

6. The method according to any one of claims 3 to 5, further comprising: 　　transmitting, to the central unit or the OAM function, feedback of performance of the AI / ML model.

7. The method according to claim 6, wherein 　　the feedback includes at least one of: 　　　　user equipment (UE) performance; 　　　　information indicating system key performance indicators (KPIs); or 　　　　information indicating resource status.

8. The method according to any one of claims 1 to 7, further comprising; 　　performing an action corresponding to output of the inferring the AI / ML model.

9. The method according to claim 8, wherein 　　the output of the inferring the AI / ML model includes at least one of: 　　　　information indicating slice-level resource modification / allocation; 　　　　information for mobility assistance of a mobile device; 　　　　information indicating cell coverage configuration; 　　　　information indicating coverage adjustment of cells; 　　　　information indicating future cell planning; 　　　　information indicating cell discontinuous transmission (DTX) / discontinuous reception (DRX) activation / deactivation; 　　　　information indicating cell / beam on / off period; or 　　　　information indicating predicted beam pattern or beam switching.

10. The method according to claim 8 or 9, wherein 　　the action includes at least one of: 　　　　network slice modification; 　　　　cell coverage adjustment; 　　　　activation / deactivation of discontinuous transmission (DTX) / discontinuous reception (DRX); 　　　　cell / beam on / off; 　　　　beam refinement; 　　　　beam selection; or 　　　　beam switching.

11. The method according to any one of claims 8 to 10, further comprising: 　　transmitting feedback of the action.

12. The method according to claim 11, wherein 　　the feedback includes at least one of: 　　　　user equipment (UE) performance; 　　　　information indicating slice-level resource status; 　　　　information indicating system key performance indicators (KPIs); or 　　　　information indicating resource status.

13. The method according to any one of claims 1 to 12, wherein 　　the input data for inferring the AI / ML model includes at least one of: 　　　　information indicating slice level resource usage or allocation of neighbour cells; 　　　　information indicating load of neighbour cells; 　　　　information indicating configuration of cell discontinuous transmission (DTX) / discontinuous reception (DRX); or 　　　　information indicating at least one of: 　　　　　　beam level measurements of a mobile device, 　　　　　　trajectory of the mobile device, or 　　　　　　location and speed of the mobile device.

14. The method according to any one of claims 1 to 13, further comprising: 　　transmitting, to the central unit, information indicating whether the distributed unit is able to infer the AI / ML model, and 　　the inferring is performed in a case where the distributed unit is able to infer the AI / ML model.

15. The method according to claim 14, wherein 　　the information is transmitted in a central unit-distributed unit interface setup procedure or a distributed unit configuration update procedure.

16. The method according to any one of claims 1 to 15, further comprising: 　　receiving, from the central unit, information for requesting the central unit to disable to infer the AI / ML model; and 　　determining to disable to infer the AI / ML model.

17. The method according to any one of claims 1 to 16, further comprising: 　　determining to disable to infer the AI / ML model; and 　　transmitting, to the central unit, information indicating that the distributed unit disables to infer the AI / ML model.

18. A method performed by a central unit of an access network node, the method comprising: 　　transmitting, to a distributed unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model, wherein 　　the input data is used by the distributed unit for use in inferring the AI / ML model.

19. A distributed unit of an access network node, the distributed unit comprising: 　　means for receiving, from a central unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model; and 　　means for inferring the AI / ML model.

20. A central unit of an access network node, the central unit comprising: 　　means for transmitting, to a distributed unit of the access network node, input data for inferring an artificial intelligence / machine learning (AI / ML) model, wherein 　　the input data is used by the distributed unit for use in inferring the AI / ML model.

Images