Network-side additional conditions for user-equipment-side artificial intelligence or machine learning model

Abstract

The present disclosure provides a communication apparatus and a communication method for DL Tx beam prediction for artificial intelligence (AI) / machine learning (ML) model, the communication apparatus comprising: circuitry, which in operation, determines, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (AI) or a machine learning (ML) model; and a transceiver, which in operation, transmits the one or more associated identifiers to another communication apparatus.

Description

[0001] DESCRIPTION

[0002] Title Of Invention: NETWORK-SIDE ADDITIONAL CONDITIONS FOR USER- EQUIPMENT-SIDE ARTIFICIAL INTELLIGENCE OR MACHINE LEARNING MODEL

[0003] TECHNICAL FIELD

[0004] [1] The following disclosure relates to a communication apparatus and a communication method, more particularly, for network-side additional conditions for a user-equipment-sided artificial intelligence (Al)Zmachine learning (ML) model.

[0005] BACKGROUND

[0006] [2] For Artificial Intelligence (AI)ZMachine Learning (ML) for NR air interface, 3GPP focuses on 3 main uses case, including CSI feedback enhancement, beam management (BM), and positioning accuracy enhancements. There are two representative BM use cases based on AIZML. For ease of explanation, two beam sets are defined, i.e., set A and set B, where set A is for beam prediction and set B is for beam measurement. Objectives of BM in a Rel. 19 WID on AIZML for NR air interface include downlink (DL) Tx beam prediction for both UE-sided model and NW-sided model, encompassing [RAN1ZRAN2]: ( / ) BM-Case1 : Spatial- domain DL Tx beam prediction for Set A of beams based on measurement results of Set B of beams; ( / / ) BM-Case2: Temporal DL Tx beam prediction for Set A of beams based on the historic measurement results of Set B of beams; (7 / 7) specify necessary signallingZmechanism(s) to facilitate life-cycle management (LCM) operations specific to the BM use cases, if any; and ( / ) enabling method(s) to ensure consistency between training and inference regarding network (NW)-side additional conditions (if identified) for inference at UE. It is noted that the objectives strive for common framework design to support both BM-Case1 and BM-Case2.

[0007] [3] For an AIZML-based BM, NW-side additional conditions refer to any aspects that are assumed for training of the user-equipment (UE)-sided model, but they are not a part of UE capability for the AIZML-based BM. In particular, NW-side additional conditions can be used for categorizing data, which is collected by UE, for a purpose of differentiating characteristics of the collected data. The collected data are inputted to UE-sided model training andZor UE- sided model inference. A UE-sided model training is a process to train an AIZML model by learning the inputZoutput relationship at UE side (e.g., 3rd party server or UE vendor server) in a data driven manner and obtain the trained AI / ML model for inference. A UE-sided model inference is a process of using the trained AI / ML model at UE to produce a set of outputs based on a set of inputs. NW-side additional conditions can affect accuracy of DL Tx beam prediction of UE-sided model training and UE-sided model inference.

[0008] [4] To ensure consistency of NW-side additional conditions across training and inference for a UE-sided model, associated identifiers (IDs) are introduced to represent implicitly NW- side additional conditions (or to abstract these conditions in an implicit manner) and let UE distinguish whether it collects data based on same or different NW-side additional conditions. An “associated ID” can represent implicitly a configuration of NW-side additional conditions from a gNodeB (gNB) or NW device deployed by a NW operator. If UE receives a same “associated ID”, it understands that is the same configuration of NW additional conditions from gNB or NW. When data is collected by UE with a same “associated ID” for both UE-sided model training and UE-sided model inference, accuracy of downlink (DL) transmission (Tx) beam prediction of the UE-sided model training and the UE-sided model inference can be maintained.

[0009] [5] However, different NW vendors and / or NW operators predetermine which “associated ID” to which NW-side additional conditions. This might cause the proprietary information for example, NW-side beam shaping information, also known as NW-vendor-specific implementation for beam shaping, such as DL spatial Tx filter (3dB beamwidth, codebook implementation for DLTx beam boresight direction (azimuth and elevation), etc.), RF modules, antenna patterns, etc., to be disclosed across NW vendors.

[0010] [6] Hence, there is a need to address one or more of the above challenges and provide communication apparatuses and communication methods for network-side additional conditions for a UE-sided AI / ML model.

[0011] [7] Furthermore, other desirable features and characteristics will become apparent from the subsequent detailed description and the appended claims, taken in conjunction with the accompanying drawings and this background of the disclosure.

[0012] SUMMARY

[0013] [8] One non-limiting and exemplary embodiment facilitates providing communication apparatuses and methods for AI / ML model. [9] In an embodiment, the techniques disclosed here provides a first communication apparatus comprising: circuitry, which in operation, determines, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or a machine learning (ML) model; and a transceiver, which in operation, transmits the one or more associated identifiers to a second communication apparatus.

[0014]

[0010] In another embodiment, the techniques disclosed here provides a second communication apparatus comprising: a transceiver, which in operation, receives, from a first communication apparatus, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model; and circuitry, which in operation, collects data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

[0015]

[0011] In an embodiment, the techniques disclosed here provides a communication method implemented by a first communication apparatus comprising: determining, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or a machine learning (ML) model; and transmitting the one or more associated identifiers to a second communication apparatus.

[0016]

[0012] In yet another embodiment, the techniques disclosed here provides a communication method implemented by a second communication apparatus comprising: receiving, from a first communication apparatus, one or more associated identifiers associated with a network vendor information identifier, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model; collecting the data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

[0017]

[0013] It should be noted that general or specific embodiments may be implemented as a system, a method, an integrated circuit, a computer program, a storage medium, or any selective combination thereof.

[0014] Additional benefits and advantages of the disclosed embodiments will become apparent from the specification and drawings. The benefits and / or advantages may be individually obtained by the various embodiments and features of the specification and drawings, which need not all be provided in order to obtain one or more of such benefits and / or advantages.

[0018] BRIEF DESCRIPTION OF THE DRAWINGS

[0019]

[0015] Embodiments of the disclosure will be better understood and readily apparent to one of ordinary skilled in the art from the following written description, by way of example only, and in conjunction with the drawings, in which:

[0020]

[0016] Figure 1 shows an exemplary 3GPP NR-RAN architecture to which exemplary embodiments of the present disclosure may be applied.

[0021]

[0017] Figure 2 shows an exemplary schematic diagram of a communication apparatus in accordance with various embodiments of the present disclosure.

[0022]

[0018] Figure 3 shows an exemplary flow diagram illustrating a communication method implemented by a communication apparatus in accordance with various embodiments of the present disclosure.

[0023]

[0019] Figure 4 shows an exemplary flow diagram illustrating another communication method implemented by another communication apparatus in accordance with various embodiments of the present disclosure.

[0024]

[0020] Figure 5 shows a block diagram illustrating an operation among NW-sided devices and user equipment (UE)-sided devices for NW-side additional conditions for a UE-sided artificial intelligence (Al)Zmachine learning (ML) model according to various embodiments of the present disclosure.

[0025]

[0021] Figure 6 shows a flow diagram illustrating a process between a gNodeB (gNB) / NW device and a UE for NW-side additional conditions for a UE-sided AI / ML model according to an embodiment of the present disclosure.

[0022] Figure 7 shows a flow chart illustrating a process carried out by a NW device for NW- side additional conditions for a UE-sided AI / ML model according to an embodiment of the present disclosure.

[0026]

[0023] Figure 8 shows a flow chart illustrating a process carried out by a UE for NW-side additional conditions for a UE-sided AI / ML model according to an embodiment of the present disclosure.

[0027]

[0024] Figure 9 shows an exemplary associated ID according to an embodiment of the present disclosure.

[0028]

[0025] Figure 10 shows exemplary functional split options in 5G O-RAN to which exemplary embodiments of the present disclosure may be applied.

[0029]

[0026] Skilled artisans will appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been depicted to scale. For example, the dimensions of some of the elements in the illustrations, block diagrams or flowcharts may be exaggerated in respect to other elements to help to improve understanding of the present embodiments.

[0030] DETAILED DESCRIPTION

[0031]

[0027] Some embodiments of the present disclosure will be described, by way of example only, with reference to the drawings. Like reference numerals and characters in the drawings refer to like elements or equivalents.

[0032]

[0028] 3GPP has been working at the next release for the 5thgeneration cellular technology, simply called 5G, including the development of a new radio access technology (NR) operating in frequencies ranging up to 100 GHz. The first version of the 5G standard was completed at the end of 2017, which allows proceeding to 5G NR standard-compliant trials and commercial deployments of smartphones. The second version of the 5G standard was completed in June 2020, which further expand the reach of 5G to new services, spectrum and deployment such as unlicensed spectrum (NR-U), non-public network (NPN), time sensitive networking (TSN) and cellular-V2X.

[0033]

[0029] 5G NR system architecture assumes an NG-RAN (Next Generation - Radio Access Network) that comprises gNBs, providing the NG-radio access user plane (i.e., SDAP / PDCP / RLC / MAC / PHY) and control plane (i.e., RRC) protocol terminations towards the UE. The gNBs are interconnected with each other by means of the Xn interface. The gNBs are also connected by means of the Next Generation (NG) interface to the NGC (Next Generation Core), more specifically to the AMF (Access and Mobility Management Function) (e.g., a particular core entity performing the AMF) by means of the NG-C interface and to the UPF (User Plane Function) (e.g., a particular core entity performing the UPF) by means of the NG-U interface. The NG-RAN architecture is illustrated in Fig. 1 (see e.g., 3GPP TS 38.300 v15.6.0, section 4).

[0034]

[0030] The user plane protocol stack for NR (see e.g., 3GPP TS 38.300, section 4.4.1 ) comprises the PDCP (Packet Data Convergence Protocol, see section 6.4 of 3GPP TS

[0035] 38.300), RLC (Radio Link Control, see section 6.3 of 3GPP TS 38.300) and MAC (Medium Access Control, see section 6.2 of 3GPP TS 38.300) sublayers, which are terminated in the gNB on the network side. Additionally, a new access stratum (AS) sublayer (SDAP, Service Data Adaptation Protocol) is introduced above PDCP (see e.g., sub-clause 6.5 of 3GPP TS

[0036] 38.300). A control plane protocol stack is also defined for NR (see for instance 3GPP TS 38.300, section 4.4.2). An overview of the Layer 2 functions is given in sub-clause 6 of 3GPP TS 38.300. The functions of the PDCP, RLC and MAC sublayers are listed respectively in sections 6.4, 6.3, and 6.2 of 3GPP TS 38.300. The functions of the radio resource control (RRC) layer are listed in sub-clause 7 of 3GPP TS 38.300.

[0037]

[0031] For instance, the Medium-Access-Control layer handles logical-channel multiplexing, and scheduling and scheduling-related functions, including handling of different numerologies.

[0038]

[0032] The physical layer (PHY) is for example responsible for coding, PHY hybrid automatic repeat request (HARQ) processing, modulation, multi-antenna processing, and mapping of the signal to the appropriate physical time-frequency resources. It also handles mapping of transport channels to physical channels. The physical layer provides services to the MAC layer in the form of transport channels. A physical channel corresponds to the set of time-frequency resources used for transmission of a particular transport channel, and each transport channel is mapped to a corresponding physical channel. For instance, the physical channels are PRACH (Physical Random Access Channel), PUSCH (Physical Uplink Shared Channel) and PUCCH (Physical Uplink Control Channel) for uplink, PDSCH (Physical Downlink Shared Channel), PDCCH (Physical Downlink Control Channel) and PBCH (Physical Broadcast Channel) for downlink, and PSSCH (Physical Sidelink Shared Channel), PSCCH (Physical Sidelink Control Channel) and Physical Sidelink Feedback Channel (PSFCH) for sidelink (SL).

[0033] SL supports UE-to-UE direct communication using the SL resource allocation modes, physical layer signals / channels, and physical layer procedures. Two SL resource allocation mode are supported: (a) mode 1 , where the SL resource allocation is provided by the network; and (b) mode 2, where UE decides SL transmission resource in the resource pool(s).

[0039]

[0034] PSCCH indicates resource and other transmission parameters used by a UE for PSSCH. PSCCH transmission is associated with a demodulation reference signal (DM-RS). PSSCH transmits the transport blocks (TBs) of data themselves, and control information for HARQ procedure and channel state information (CSI) feedback triggers, etc. At least 6 Orthogonal Frequency Division Multiplex (OFDM) symbols within a slot are used for PSSCH transmission. PSSCH transmission is associated with a DM-RS and may be associated with a phase-tracking reference signal (PT-RS).

[0040]

[0035] PSFCH carries HARQ feedback over the SL from a UE which is an intended recipient of a PSSCH transmission to the UE which performed the transmission. PSFCH sequence is transmitted in one PRB repeated over two OFDM symbols near the end of the SL resource in a slot.

[0041]

[0036] The SL synchronization signal consists of SL primary and SL secondary synchronization signals (S-PSS, S-SSS), each occupying 2 symbols and 127 subcarriers. Physical Sidelink Broadcast Channel (PSBCH) occupies 9 and 5 symbols for normal and extended cyclic prefix cases respectively, including the associated demodulation reference signal (DM-RS).

[0042]

[0037] Regarding physical layer procedure for HARQ feedback for sidelink, SL HARQ feedback uses PSFCH and can be operated in one of two options. In one option, which can be configured for unicast and groupcast, PSFCH transmits either ACK or NACK using a resource dedicated to a single PSFCH transmitting UE. In another option, which can be configured for groupcast, PSFCH transmits NACK, or no PSFCH signal is transmitted, on a resource that can be shared by multiple PSFCH transmitting UEs.

[0043]

[0038] In SL resource allocation mode 1 , a UE which received PSFCH can report SL HARQ feedback to gNB via PUCCH or PUSCH.

[0039] Regarding physical layer procedure for power control for sidelink, for in-coverage operation, the power spectral density of the SL transmissions can be adjusted based on the pathloss from the gNB; whereas for unicast, the power spectral density of some SL transmissions can be adjusted based on the pathloss between the two communicating UEs.

[0044]

[0040] Regarding physical layer procedure for CSI report, for unicast, channel state information reference signal (CSI-RS) is supported for CSI measurement and reporting in sidelink. A CSI report is carried in a SL MAC CE (Control Element).

[0045]

[0041] For measurement on the sidelink, the following UE measurement quantities are supported:

[0046] • PSBCH reference signal received power (PSBCH RSRP);

[0047] • PSSCH reference signal received power (PSSCH-RSRP);

[0048] • PSCCH reference signal received power (PSCCH-RSRP);

[0049] • Sidelink received signal strength indicator (SL RSSI);

[0050] • Sidelink channel occupancy ratio (SL CR);

[0051] • Sidelink channel busy ratio (SL CBR).

[0052]

[0042] Use cases / deployment scenarios for NR could include enhanced mobile broadband (eMBB), ultra-reliable low-latency communications (URLLC), massive machine type communication (mMTC), which have diverse requirements in terms of data rates, latency, and coverage. For example, eMBB is expected to support peak data rates (20Gbps for downlink and 10Gbps for uplink) and user-experienced data rates in the order of three times what is offered by IMT-Advanced. On the other hand, in case of URLLC, the tighter requirements are put on ultra-low latency (0.5ms for UL and DL each for user plane latency) and high reliability (1-1 O'5within 1ms). Finally, mMTC may preferably require high connection density (1 ,000,000 devices / km2in an urban environment), large coverage in harsh environments, and extremely long-life battery for low-cost devices (15 years).

[0053]

[0043] Therefore, the OFDM numerology (e.g. subcarrier spacing, OFDM symbol duration, cyclic prefix (CP) duration, number of symbols per scheduling interval) that is suitable for one use case might not work well for another. For example, low-latency services may preferably require a shorter symbol duration (and thus larger subcarrier spacing) and / or fewer symbols per scheduling interval (aka, TTI) than an mMTC service. Furthermore, deployment scenarios with large channel delay spreads may preferably require a longer CP duration than scenarios with short delay spreads. The subcarrier spacing should be optimized accordingly to retain the similar CP overhead. NR may support more than one value of subcarrier spacing. Correspondingly, subcarrier spacing of 15kHz, 30kHz, 60 kHz... are being considered at the moment. The symbol duration Tuand the subcarrier spacing Af are directly related through the formula Af = 1 / Tu. In a similar manner as in LTE systems, the term “resource element” can be used to denote a minimum resource unit being composed of one subcarrier for the length of one OFDM / SC-FDMA symbol.

[0054]

[0044] In the new radio system 5G-NR for each numerology and carrier a resource grid of subcarriers and OFDM symbols is defined respectively for uplink and downlink. Each element in the resource grid is called a resource element and is identified based on the frequency index in the frequency domain and the symbol position in the time domain (see 3GPP TS 38.211 V16.3.0).

[0055]

[0045] As mentioned earlier, for artificial intelligence (Al) / machine learning (ML) for NR air interface, 3GPP focuses on 3 main uses case, including CSI feedback enhancement, beam management (BM), and positioning accuracy enhancements.

[0056]

[0046] Objectives of BM in a Release 19 working item description (WID) on AI / ML for NR air interface are shown in the following [RP-234039]:

[0057] • Downlink (DL) Tx beam prediction for both user equipment (UE)-sided model and NW- sided model, encompassing [Radio Layer 1 (RAN1 ) / RAN2]: o BM-Case1 : Spatial-domain downlink (DL) transmission (Tx) beam prediction for Set A of beams based on measurement results of Set B of beams; o BM-Case2: Temporal DL Tx beam prediction for Set A of beams based on the historic measurement results of Set B of beams; o Specify necessary signalling / mechanism(s) to facilitate life cycle management (LCM) operations specific to the BM use cases, if any; o Enabling method(s) to ensure consistency between training and inference regarding NW-sided additional conditions (if identified) for inference at UE. o Note: Strive for common framework design to support both BM-Case1 and BM- Case2.

[0058]

[0047] For an AI / ML-based BM, NW-sided additional conditions refer to any aspects that are assumed for training of the UE-sided model, but they are not a part of UE capability for the AI / ML-based BM. In particular, NW-side additional conditions can be used for categorizing data, which is collected by UE, for a purpose of differentiating characteristics of the collected data. The collected data are inputted to UE-sided model training and / or UE-sided model inference. NW-side additional conditions can affect accuracy of DL Tx beam prediction of UE- sided model training and UE-sided model inference.

[0059]

[0048] NW-side additional conditions might include two types of information:

[0060] (i) Non-proprietary information: Configuration of RS resources of Set A and / or Set B, QCL configurations for Set A and / or Set B, configuration of association between Set A and Set B (if any), number of historic measurement results of Set B, etc.; and

[0061] (ii) Proprietary information: NW-side beam shaping information (also known as (aka) NW- vendor-specific implementation for beam shaping), such as DL spatial Tx filter (3dB beamwidth, codebook implementation for DLTx beam boresight direction (azimuth and elevation), etc.), radio frequency (RF) modules, antenna patterns, etc.

[0062]

[0049] To ensure consistency of NW-side additional conditions across training and inference for a UE-sided model, associated identifiers (IDs) are introduced to represent implicitly NW- side additional conditions (or to abstract these conditions in an implicit manner) and let UE distinguish whether it collects data based on same or different NW-side additional conditions. An “associated ID” can represent implicitly a configuration of NW-side additional conditions from a gNodeB (gNB) or NW device deployed by a NW operator. If UE receives a same “associated ID”, it understands that is the same configuration of NW additional conditions from gNB or NW. When data is collected by UE with a same “associated ID” for both UE-sided model training and UE-sided model inference, accuracy of downlink (DL) transmission (Tx) beam prediction of the UE-sided model training and the UE-sided model inference can be maintained.

[0063]

[0050] According to an agreement (RAN1#116-bis), further study is needed, for the consistency of NW-side additional condition across training and inference for a UE-sided model for BM-Case 1 and BM Case 2, where the NW-side additional condition may at least impact UE assumption on beams of Set A / Set B:

[0064] • Option 1 : Based on associated ID (Referring to agenda item 9.1.3.3) o Further studies are required on what can be assumed by UE with the same associated ID across training and inference o Further studies are required on how associated ID is introduced, e.g., within CSI framework, or outside of CSI framework

[0065] • Option 2: Performance monitoring based o Further studies on the details are required

[0066] • Other options are not precluded.

[0067]

[0051] For definition of “associated ID”, it might cause a potential risk of disclosing proprietary information for NW vendor and / or NW operator [Huawei (R1-2403929), CATT(R1 -2404388), LG(R1 -2404546), Fujitsu (R1-2404586), vivo (R1-2404165)]. In particular, ( / ) different NW vendors and / or NW operators (or NW devices deployed by them) need to coordinate to assign which “associated ID” to which NW-side additional conditions. That may cause the proprietary information to be disclosed across NW vendors (e.g., when “associated ID” is defined as a global ID), and ( / / ) NW operator also does not want to disclose information of which NW vendor(s) are deploying in which area / site.

[0068]

[0052] Therefore, in RAN1#117 meeting, RAN1 assumed to restrict applicability of “associated ID” at least within a cell-level. Particularly, it has assumed that NW-side additional conditions with the same associated ID are consistent at least within a cell. It implies that a same “associated ID” in different cells may represent different NW-side additional conditions. Hence, it causes a limitation of usage of UE-sided model training and model inference within a cell. It cannot handle a case that UE-sided model training and model inference are performed at different cells. It is further study on whether / how to apply “associated ID” for multiple cells. According to RAN1#117, further studies are required on whether / how UE assumption can be applicable for multiple cells (including the feasibility study).

[0069]

[0053] RAN1 has discussed how to apply “associated ID” for different cells because it is more useful for a UE-sided model perspective. In particular, from RAN1 perspective, for a UE-sided model(s) developed (e.g., trained, updated) at UE side, the following procedure is an example (noted as Al-Example 1 ) of model identification (Ml)-Option 1 for further study (including the feasibility / necessity):

[0070] A: For data collection, NW device signals the data collection related configuration(s) and associated ID(s) for each sub use case in relation with NW-sided additional conditions B: UE(s) collects the data corresponding to the associated I D(s)

[0071] C: AI / ML models are developed (e.g., trained, updated) at UE side based on the collected data corresponding to the associated ID(s).

[0072] D: UE reports information of its AI / ML models corresponding to associated IDs to the NW device. Model ID is determined / assigned for each AI / ML model relationship between model ID(s) and the associated I D(s).

[0054] It is noted that step D is carried out to facilitate AI / ML model inference while step A / B / C and additional interaction of associated IDs between UE and NW device can be considered as a different solution for resolving the consistency without model identification.

[0073]

[0055] The model ID(s) is determined / assigned, for example through four different implementations below:

[0074] 1 : NW device assigns Model ID

[0075] 2: UE assigns / reports Model ID

[0076] 3: Associated ID(s) is assumed as model ID(s). In this way, “Model ID is determined / assigned for each AI / ML model” in D is not needed

[0077] 4: Model ID is determined by pre-defined rule(s) in the specification

[0078]

[0056] However, further studies are required on how the UE reports the information of AI / ML models.

[0079]

[0057] The applicant of “associated ID’ helps to simplify dataset categorization at UE-sided where data is collected from UEs in different cells. UE-sided model training and inference can be used in different cells. However, as mentioned earlier, this might still cause the potential risk of disclosing proprietary information. This is not desirable by NW vendors and / or NW operators. For example, different NW vendors and / or NW operators predetermine which “associated ID” to which NW-side additional conditions. This might cause the proprietary information to be disclosed across NW vendors.

[0080]

[0058] Hence, there is a need to address one or more of the above challenges and provide communication apparatuses and communication methods for network-side additional conditions for a UE-sided AI / ML model, in particular, to preserve the proprietary information for applying “associated ID” for different cells.

[0081]

[0059] In various embodiments below, a gNodeB (gNB) may be referred to a NW-sided device or a NW device; whereas a user equipment (UE) may be referred to as a UE-sided device. A NW operator can deploy a plurality of gNBs, which are produced by same NW vendor or different NW vendors, in the network. In various embodiments below, a process involving a NW operator, when mentioned, refers to a process involving the gNB / NW device(s) deployed by the NW operator. The term “information of network vendor identification” may be used interchangeably with “network vendor information identifier”.

[0060] In various embodiments below, an associated ID may consist of a general part and a private part. The general part can represent general information such as public land mobile network (PLMN) index. The private part can represent specific configuration of NW-side additional conditions, so the NW operator can link between the network vendor information identifier (e.g., NW-vendor-specific ID(s)) and the private part and thus determine the associated ID based on the network vendor information identifier reported by or allocated to the NW vendor. In various embodiments below, one or more of the following IDs and indices can be included in an associated ID: ( / ) network (NW)-vendor-specific ID(s), a cell group index (e.g., a cell group index can include different cells), PLMN index, a country index, an area / region / zone index (e.g., district / city), an assigned index of a configuration which includes specific value(s) of NW-side additional conditions. For example, an associated ID #1 is formatted as “PLMN index” + “zone index” + [assigned index]”.

[0082]

[0061] Fig. 2 shows an exemplary schematic diagram of a communication apparatus 200 in accordance with various embodiments of the present disclosure. The communication apparatus may be implemented as a UE, a gNB / base station in accordance with various embodiments of the present disclosure. The communication apparatus 200 may include circuitry 214, at least one radio transmitter 202, at least one radio receiver 204, and at least one antenna 212 (for the sake of simplicity, only one antenna is depicted in Fig. 2 for illustration purposes). The circuitry 214 may include at least one controller 206 for use in software and / or hardware aided execution of tasks that the at least one controller 206 is designed to perform, including control of communications with one or more other communication apparatuses in a wireless network. The circuitry 214 may further include at least one transmission signal generator 208 and at least one receive signal processor 210. The at least one controller 206 may control the at least one transmission signal generator 208 for generating signals (for example, a sidelink / uplink / downlink signal) to be sent through the at least one radio transmitter 202 to one or more other communication apparatuses and the at least one receive signal processor 210 for processing signals (for example, a sidelink / uplink / downlink signal) received through the at least one radio receiver 204 from the one or more other communication apparatuses under the control of the at least one controller 206. The at least one transmission signal generator 208 and the at least one receive signal processor 210 may be stand-alone modules of the communication apparatus 200 that communicate with the at least one controller 206 for the above-mentioned functions, as shown in Fig. 4. Alternatively, the at least one transmission signal generator 208 and the at least one receive signal processor 210 may be included in the at least one controller 206. In various embodiments, when in operation, the at least one radio transmitter 202, at least one radio receiver 204, and at least one antenna 212 may be controlled by the at least one controller 206.

[0083]

[0062] The at least one radio transmitter 202 and the at least one radio receiver 204 may be included in a stand-alone module of the communication apparatus 200 to perform functions of both sending and receiving signals to and from another communication apparatus respectively. Such a module may be referred to as a transceiver 202, 204 in various embodiments of the present disclosure.

[0084]

[0063] It is appreciable to those skilled in the art that the arrangement of these functional modules is flexible and may vary depending on the practical needs and / or requirements. The data processing, storage and other relevant control apparatus can be provided on an appropriate circuit board and / or in chipsets.

[0085]

[0064] The communication apparatus 200 may be a base-station, gNB or NW-sided device deployed by a NW operator, and when in operation, provide functions required for NW-side additional conditions for a UE-sided artificial intelligence (Al) / machine learning (ML) model. Fig. 3 shows an exemplary flow diagram 300 illustrating a communication method for NW-side additional conditions for a UE-sided AI / ML model implemented by the communication apparatus 200 when it works as / is a base-station, gNB or NW-sided device, for example, deployed by a NW operator in accordance with various embodiments of the present disclosure.

[0086]

[0065] In particular, in step 302, the circuitry 214 (e.g., the at least one controller 206 of the circuitry 214) may, in operation, determine, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an Al or a ML model. In step 304, the at least one radio transmitter 202 may transmit the one or more associated identifier to another communication apparatus (e.g., UE or a UE-sided device).

[0087]

[0066] The condition may be one or a combination of at least: a configuration of reference signal (RS) resources of a first set and / or a second set, a Quasi Co-Location (QCL) configuration for the first and / or second sets, a configuration of an association between the first and second sets, and a number of historic measurement results of the first and / or second sets. The first set of RS resources are corresponding to the first set of DL Tx beams that can be ( / ) used for determining measurement results of the first set of DL Tx beams by UE, and (Ji) inputted in an AI / ML model training and / or an AI / ML model inference. The second set of resources are corresponding to the second set of DL Tx beams that can be inputted in an AI / ML model inference for DL Tx beam prediction.

[0088]

[0067] In an embodiment, in step 302, the at least one radio receiver 204 may receive the network vendor information identifier from a corresponding network vendor and the circuitry 214 (e.g., the at least one controller 206 of the circuitry 214) may determine, based on the received network vendor information identifier, the one or more associated identifiers. In another embodiment, in step 302, the circuitry 214 (e.g., the at least one controller 206 of the circuitry 214) may further allocate the network vendor information identifier to a corresponding network vendor, and determine, based on the allocated network vendor information identifier, the one or more associated identifiers. Additionally, or alternatively, the at least one radio receiver 204 may receive the one or more associated identifiers from the corresponding network vendor, the one or more associated identifiers being used as the network vendor information identifier, and the circuitry 214 (e.g., the at least one controller 206 of the circuitry 214) assigns the one or more associated identifiers to the corresponding network vendor.

[0089]

[0068] In one embodiment, each of the one or more associated identifiers includes the network vendor information identifier. The each of the one or more associated identifier may comprise one or more of: ( / ) an index of a cell group to which the corresponding network vendor belongs, (77) an index of a country, an area, a region or a zone in which the corresponding network vendor is located, (Hi) an index of public land mobile network, and (iv) an assigned index for a configuration relating to the condition for collecting the data fortraining and / or inferring the Al or ML model.

[0090]

[0069] Yet in another embodiment, the network vendor information identifier is mapped to the one or more associated identifiers based on a table-based mapping or a rule, and, in step 302, the circuitry 214 (e.g., the at least one controller 206 of the circuitry 214) may determine, based on the network vendor information identifier and the mapping, the one or more associated identifiers.

[0091]

[0070] The communication apparatus 200 may be a UE or a UE-sided device and when in operation, provide functions required for NW-side additional conditions for a UE-sided artificial intelligence (Al)Zmachine learning (ML) model. Fig. 3 shows an exemplary flow diagram 300 illustrating a communication method for NW-side additional conditions for a UE-sided AI / ML model implemented by the communication apparatus 200 when it works as / is a base-station, gNB or NW-sided device, for example, deployed by a NW operator in accordance with various embodiments of the present disclosure. Fig. 4 shows an exemplary flow diagram 400 illustrating a communication method for NW-side additional conditions for a UE-sided AI / ML model implemented by the communication apparatus 200 when it works as / is a UE or UE- sided device in accordance with various embodiments of the present disclosure.

[0092]

[0071] In one embodiment, the at least one radio transmitter 202 may transmit the collected data to an entity which trains the Al or ML model, the entity being a server of a network vendor of the second communication apparatus or a third-party server; the at least one radio receiver 204 may receive a trained Al or ML model from the entity; and / or the at least one radio transmitter 202 may transmit the trained Al or ML model to the first communication apparatus.

[0093]

[0072] In particular, in step 402, the at least one radio transmitter 202 may receive, from another communication apparatus (e.g., gNB, base station, NW-sided device), one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model, and circuitry, which in operation, collects data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

[0094]

[0073] According to the present disclosure, a NW operator and each of NW vendor share information of NW vendor identification or an identifier (ID) of such information (herein referred to as NW vendor information identifier) such as NW vendor specific ID, and the NW operator utilizes the information of NW vendor or NW vendor information identifier to determine a set of associated IDs, each of the set of associated IDs represents a NW- side additional conditions, for the corresponding NW vendor.

[0095]

[0074] Figure 5 shows a block diagram 500 illustrating an operation among NW-sided devices and UE-sided devices for NW-side additional conditions for a UE-sided artificial intelligence (Al)Zmachine learning (ML) model according to various embodiments of the present disclosure. NW vendor A (e.g., NW vendor A provides its own NW-sided device to a NW operator (e.g., Nokia provides gNB solutions to Singtel)) may share two NW-vendor- specific IDs (# / , #j) and NW vendor B (e.g., NW vendor B provides its own NW-sided device to a NW operator (e.g., Ericsson provides another gNB solutions to Singtel)) may share one NW-vendor specific ID to the NW operator. Each of the NW-vendor-specific IDs corresponds to one or more associated IDs, and each associated ID represents a NW-side additional condition for the corresponding NW-vendor (NW-vendor-specific ID). For example, NW- vendor-specific ID #i of NW vendor A corresponds to associated IDs #1-4, NW-vendor specific ID #j of NW vendor A corresponds to associated ID #6 and #n, and NW-vendor-specific ID #k of NW vendor B corresponds to associated ID #5. The NW operator upon receipt of the NW- vendor-specific IDs may determine the one or more associated IDs corresponding to the NW- vendor-specific IDs. The one or more associated IDs will then be signalled to a UE(s) for collecting data for training and / or inferring a UE-sided Al or ML model.

[0096]

[0075] It is noted that the gNB / NW-sided device deployed by the NW operator only signals the set of associated IDs to UEs corresponding to set of configurations of NW-side additional conditions. Hence, UE knows each configuration (i.e., specific values) from the set of configurations corresponding to each associated ID from the set of associated IDs. Regarding how to use the configured associated IDs,

[0097] • For UE-side model training, UE collects data (e.g., measurement results of beams of Set A and / or Set B) corresponding to the set of associated IDs, then it transfers the collected data to a third party server or UE vendor sever to train AI / ML model(s) through internet in a higher layer, where the third party server could be a server or a cloud service of a third-party company such as Amazon Web Services, Microsoft Azure, and UE vendor server could be a server or a cloud service of UE manufacturing company.

[0098] • For UE-side model inference, the gNB / NW-sided device(s) deployed by the NW operator configures one or more associated IDs from the set of associated IDs to UE, UE performs inference of the trained Al model(s) (i.e., process of UE-side model inference) corresponding to the one or more associated IDs.

[0099]

[0076] Advantageously, the NW vendors do not share with each other associated IDs which contains the proprietary information of NW-vendor-specific implementation for beam shaping as associated IDs are determined by NW operator based on the NW vendor information identifiers and it is up to NW operator to determine the associated IDs. The NW operator preserves information of which NW vendor(s) are deploying in which area / site in different cells. As a result, the proprietary information for applying “associated ID” for different cells can be preserved.

[0100]

[0077] It is noted that it is up to NW operator to determine “associated IDs”. The set of associated IDs are corresponding to a set of configurations of NW additional conditions for the information of NW vendor identification or NW vendor information identifier. The NW operator does not share UE(s) the information of NW vendor identification and how NW operator utilizes the information to determine a set of associated IDs. The UE(s) only receives the set of associated IDs. It is also noted that there could be multiple NW vendors in the NW, wherein each NW vendor has the information of NW vendor, as shown in Figure 5.

[0101]

[0078] For an associated ID given to a DE, consistency is kept between different times and between training and inference. For example, if the associated ID is not allocated for the UE, it can be changed. If the associated ID is not used any more for the DE, it can be recycled the same ID with the different vendors. Regarding NW vendor(s), if a cell is equipped with solutions (including hardware and / or software solutions) (e.g., transmission reception point (TRP), gNB distributed unit (gNB-DU), gNB centralized unit (gNB-CU), etc.) that are provided by a single vendor, each of NW vendor(s) can be referred as a single vendor. If a cell is equipped with solutions (including hardware and / or software solutions) that are provided from different NW vendors (e.g., TRP, gNB-DU and gNB-CU), each of NW vendor(s) can be referred as "solution provider" or "a division of operator" or each of "group of NW vendors responsible for a cell or cell groups", e.g., for use case of O-RAN solutions.

[0102]

[0079] Figure 6 shows a flow diagram 600 illustrating a process between a gNB / NW device (herein referred to as NW device) and a UE for NW-side additional conditions for a UE-sided artificial intelligence (Al) / machine learning (ML) model according to an embodiment of the present disclosure. In step 601 , the NW device (NW operator) utilizes information of NW vendor identification (e.g., NW-vendor specific ID(s)) to define a set of associated IDs for each of NW vendor(s). The set of associated IDs are corresponding to a set of configurations of NW additional conditions for the NW-vendor-specific ID(s). In step 602, the NW device signals the set of associated IDs to the UE. In step 603, the UE then collects data (e.g., measurement results of beams of Set A and / or Set B) corresponding to the set of associated IDs. In step 604, the UE transfers the collected data and the set of associated IDs to UE-vendor server or third-party server through internet in a higher layer, wherein the third-party server could be a server or a cloud service of a third-party company such as Amazon Web Services, Microsoft Azure, and UE vendor server could be a server or a cloud service of UE manufacturing company. In step 605, the UE-sided AI / ML models are trained at UE-vendor server or third- party server based on the collected data corresponding to the associated ID(s). In step 606, the UE-vendor server or third-party server delivers trained AI / ML model(s) corresponding to the associated ID(s) to the UE. In step 607, the UE then reports information of supporting trained AI / ML model(s) corresponding to the associated ID(s) to the NW device. In step 608, the NW signals one or more associated IDs from the set of associated IDs to UE(s), for inference of UE-sided model at the UE. In step 609, the UE performs model inference by using the trained AI / ML model(s) corresponding to the one or more associated ID(s). In step 610, the UE reports outcome (i.e., DL Tx beam prediction) of the performed model inference corresponding to the one or more associated ID(s). Note that step 603 to step 606 are related to process of UE-sided training model(s) and reporting of UE capability of trained AI / ML model(s), while step 607 to step 610 are related to process of a UE-sided inference model and reporting of inference results which is output of the UE-sided inference model.

[0103]

[0080] Figure 7 shows a flow chart 700 illustrating a process carried out by a gNB / NW-sided device deployed by a NW operator for NW-side additional conditions for a UE-sided AI / ML model according to an embodiment of the present disclosure. For sake of simplicity, the term “NW operator” may be used to refer to the gNB / NW-sided device deployed by the NW operator to carry out the process. In step 702, the NW operator utilizes information of NW vendor identification to determine a set of associated IDs for each NW vendor(s). The set of associated IDs are corresponding to a set of configurations of NW additional conditions for the NW-vendor-specific ID(s). In step 704, the NW-sided device signals the set of associated IDs to a UE. In step 706, the NW-sided device receives UE’s report of information of supporting trained AI / ML model(s) corresponding to the associated ID(s). In step 708, the NW-sided device signals one or more associated IDs, from the set of associated IDs to UE(s), for inference of UE-sided model at the UE. In step 710, the NW-sided device receives UE’s report of outcome (i.e., DL Tx beam prediction) of the performed model inference corresponding to the one or more associated ID(s).

[0104]

[0081] Figure 8 shows a flow chart 800 illustrating a process carried out by a UE for NW- side additional conditions for a UE-sided AI / ML model according to an embodiment of the present disclosure. In step 802, a UE receives the set of associated IDs from a gNB / NW-sided device. In step 804, the UE collects data (e.g., measurement results of beams of Set A and / or Set B) corresponding to the set of associated IDs. The UE transfers the collected data and the set of associated IDs to UE-vendor server or third-party server. In step 806, the UE downloads trained AI / ML model(s) corresponding to the associated ID(s) to the UE, from UE-vendor server or third-party server. In step 808, the UE reports information of supporting trained AI / ML model(s) corresponding to the associated ID(s). In step 810, the UE receives one or more associated IDs, from the set of associated IDs to UE(s), for inference of UE-sided model at the UE. In step 812, the UE performs model inference by using the trained AI / ML model(s) corresponding to the one or more associated ID(s). In step 814, the UE reports outcome (i.e., DL Tx beam prediction) of the performed model inference corresponding to the one or more associated ID(s).

[0105]

[0082] The following paragraphs describes a first embodiment of the present disclosure, where the NW vendor information identifier is reported to a NW operator by each of NW vendors. Advantageously, this provides flexibility to NW vendor to allocate the NW vendor information identifiers such as NW-vendor-specific IDs.

[0106]

[0083] In the first embodiment of the present disclosure, the NW operator utilizes the NW vendor information identifier reported by each of the NW vendors to determine a set of associated ID. In an implementation, the NW vendor information identifier is included in apart of each associated ID of the set of associated IDs. Advantageously, such implementation allows each associated ID to reflect the NW vendor information identifier of the NW vendor it is mapped to so the NW operator can minimize effort to manage pool of associated IDs. It is assumed that for a NW operator, sequences of n bit-lengths (e.g., n=4 / 8 / 16 / 32 / 64, etc) can be used to allocate associated IDs and 3 bit-lengths of the n-bit lengths corresponds to NW-vendor-specific ID or NW vendor information ID.

[0107]

[0084] Figure 9 shows an exemplary associated ID 900 according to an embodiment of the present disclosure. In this example, the associated ID 900 contains 32 bits (n=32). The first three bits “111 ” correspond NW-vendor-specific ID for the NW operator, so there are 29 remaining bits for the assigned index, shown as “aabbcc...xxyyzz” in Figure 9. Therefore, in this implementation, each of associated IDs corresponding to a NW vendor information ID contains the NW vendor information ID, thus when the NW operator receives the NW vendor information ID reported by each NW vendor, it can use to determine the set of associated IDs.

[0108]

[0085] In another implementation, the NW vendor information identifier is allocated / mapped with the set of associated IDs based on a table-based mapping or a rule. Advantageously, such implementation provides high flexibility to NW operator to allocate a set of associated IDs for each NW vendor information identifier.

[0109]

[0086] Table 1 shows an example mapping of NW-vendor-specific ID to a set of associated IDs of 32 bit-lengths.

[0110] [Table 1]

[0111]

[0087] Therefore, in this implementation, each NW vendor information ID corresponds to a set of associated IDs. When the NW operator receives the NW vendor information ID reported by each NW vendor, it can use to determine the set of associated IDs.

[0112]

[0088] Yet in another implementation, the NW-vendor can report and recommend a certain set of associated IDs (e.g., an optimal set of associated ID) to the NW operator, and the NW operator can assign the certain set of associated IDs to the NW vendor. When the NW operator receives the NW vendor information ID reported by the NW vendor, it can use to determine the set of associated IDs. For example, when deploying AI / ML model at NW-vendor side, the NW vendor might identify good AI / ML models which, which can be trained, based on such (optimally) certain set of associate IDs, hence it reports and recommends such set of associated IDs to the NW operator, and the NW operator then assign such set of associated IDs to the NW vendor. Such set of associated IDs determined and assigned by the NW operator is not known by the NW vendor(s). Advantageously, this will utilize configuration of NW-side additional conditions which are recommended / reported by the NW vendor and maximize accuracy of DL beam prediction of UE-sided model.

[0113]

[0089] The following paragraphs describes a second embodiment of the present disclosure, where the NW vendor information identifier is allocated by the NW operator. Advantageously, it provides more flexibility to NW operator to allocate information of NW vendor identification.

[0114]

[0090] In the second embodiment of the present disclosure, the NW operator utilizes the NW vendor information identifier allocated by the NW operator to each NW vendor to determine a set of associated ID. Similar to the first embodiment, in an implementation, the NW vendor information identifier allocated by the NW operator is included in apart of each associated ID of the set of associated IDs. Advantageously, such implementation allows each associated ID to reflect the linkage to the NW vendor information identifier of the NW vendor it is mapped to so the NW operator can minimize effort to manage pool of associated IDs. It is assumed that for a NW operator, sequences of n bit-lengths (e.g., n=4 / 8 / 16 / 32 / 64, etc) can be used to allocate associated IDs and 3 bit-lengths of the n-bit lengths corresponds to NW-vendor-specific ID or NW vendor information ID. Therefore, in this implementation, each of associated IDs corresponding to a NW vendor information ID contains the NW vendor information ID, thus when the NW operator receives the NW vendor information ID allocated to a NW vendor, it can use to determine the set of associated IDs corresponding to the NW vendor information ID of the NW vendor.

[0115]

[0091] In another implementation, the NW vendor information identifier is allocated / mapped with the set of associated IDs based on a table-based mapping or a rule, and each NW vendor information ID corresponds to a set of associated IDs. When the NW operator receives the NW vendor information ID allocated to a NW vendor, it can use to determine the set of associated IDs corresponding to the NW vendor information ID of the NW vendor. Advantageously, such implementation provides high flexibility to NW operator to allocate a set of associated IDs for each NW vendor information identifier.

[0116]

[0092] Yet in another implementation, the NW operator can directly assign a certain set of associated IDs (e.g., an optimal set of associated ID) to a specific NW vendor or its NW vendor information identifier. When the NW operator receives the NW vendor information ID reported by the NW vendor, it can use to determine the set of associated IDs. For example, when deploying AI / ML model for a specific NW vendor at NW operator in a prelaunch phase (i.e. , before a commercial phase of services and sales of NW operator), the NW operator might identify good AI / ML models, which can be trained, based on such (optimally) certain set of associate IDs, hence it directly assigns such set of associated IDs to the NW vendor information identifier of the NW vendor. Such set of associated IDs determined and assigned by the NW operator is not known by the NW vendor(s). Advantageously, this provides high flexibility to NW operator to assign the associated IDs for different configurations of NW-side additional conditions.

[0117]

[0093] In an alternative embodiment, instead of information of NW vendor identification (e.g., NW vendor information identifier), the NW operator may utilize one or more of ( / ) cell group index (e.g., a cell group index can include different cells), ( / / ) area / region / zone index (e.g., district / city) and public land mobile network (PLMN) index. This provides a better flexibility of NW planning about the set of associated IDs (or the corresponding UE- sided AI / ML models trained based on the set of associated IDs) which can be used locally (e.g., in some cells in a cell group, area, etc.).

[0118] RRC connection setup and reconfiguration procedures

[0119]

[0094] Interactions between a UE, gNB, and AMF (an 5G core (5GC) entity) in the context of a transition of the UE from RRC_IDLE to RRC_CONNECTED for the NAS part are described (see 3GPP TS 38.300 v15.6.0).

[0120]

[0095] RRC is a higher layer signaling (protocol) used for UE and gNB configuration. In particular this transition involves the AMF preparing the UE context data (including e.g. PDU session context, the Security Key, UE Radio Capability and UE Security Capabilities, etc.) and sending it to the gNB with the INITIAL CONTEXT SETUP REQUEST. Then, the gNB activates the AS security with the UE, which is performed by the gNB transmitting to the UE a SecurityModeCommand message and by the UE responding to the gNB with the SecurityModeComplete message. Afterwards, the gNB performs the reconfiguration to setup the Signaling Radio Bearer 2, SRB2, and Data Radio Bearer(s), DRB(s) by means of transmitting to the UE the RRCReconfiguration message and, in response, receiving by the gNB the RRCReconfigurationComplete from the UE. For a signalling-only connection, the steps relating to the RRCReconfiguration are skipped since SRB2 and DRBs are not setup. Finally, the gNB informs the AMF that the setup procedure is completed with the INITIAL CONTEXT SETUP RESPONSE.

[0121]

[0096] In the present disclosure, thus, an entity (for example Access and Mobility Management Function (AMF), Session Management Function (SMF), etc.) of a 5thGeneration Core (5GC) is provided that comprises control circuitry which, in operation, establishes a Next Generation (NG) connection with a gNodeB, and a transmitter which, in operation, transmits an initial context setup message, via the NG connection, to the gNodeB to cause a signaling radio bearer setup between the gNodeB and a user equipment (UE). In particular, the gNodeB transmits a Radio Resource Control, RRC, signaling containing a resource allocation configuration information element to the UE via the signaling radio bearer. The UE then performs an uplink transmission or a downlink reception based on the resource allocation configuration. QoS control

[0122]

[0097] The 5G QoS (Quality of Service) model is based on QoS flows and supports both QoS flows that require guaranteed flow bit rate (GBR QoS flows) and QoS flows that do not require guaranteed flow bit rate (non-GBR QoS Flows). At NAS level, the QoS flow is thus the finest granularity of QoS differentiation in a PDU session. A QoS flow is identified within a PDU session by a QoS flow ID (QFI) carried in an encapsulation header over NG-U interface.

[0123]

[0098] For each UE, 5GC establishes one or more PDU Sessions. For each UE, the NG-RAN establishes at least one Data Radio Bearers (DRB) together with the PDU Session, and additional DRB(s) for QoS flow(s) of that PDU session can be subsequently configured (it is up to NG-RAN when to do so). The NG-RAN maps packets belonging to different PDU sessions to different DRBs. NAS level packet filters in the UE and in the 5GC associate UL and DL packets with QoS Flows, whereas AS-level mapping rules in the UE and in the NG- RAN associate UL and DL QoS Flows with DRBs.

[0124] Open- RAN

[0125]

[0099] The base station described in each exemplary embodiment (for example, a 5G NR base station called gNB) may be formed of three functional modules: Centralized Unit (CU), Distributed Unit (DU), and Radio Unit (RU).

[0126]

[0100] CU may also be referred as, for example, a centralized node, an aggregated node, a centralized station, an aggregated station, or a central unit. DU may also be referred as, for example, O-DU (O-RAN Distributed Unit), a distributed node, a distributed station, or a distributed unit. RU may also be referred as, for example, O-RU (O-RAN Radio Unit), a radio apparatus, a radio node, a radio station, an antenna unit, or a radio unit.

[0127]

[0101] Several split options are defined for the functional split configuration (or functional split point) between CU, DU, and RU. The term “functional split point” may also be referred to as “split”, “option”, or “split option”.

[0128]

[0102] Examples of the “split option” include the following split options 1 to 8. The functionality of the base station described in each exemplary embodiment may be split into functions as CU, DU, and RU by one of the following split options 1 to 8. For example, each of CU, DU, and RU may be subjected to functional splitting or functional splitting only between CU and DU or only between DU and RU is possible. (1 ) Split Option 1 : between RRC (radio resource control) and PDCP

[0129] (2) Split Option 2: between PDCP and RLC (High-RLC)

[0130] (3) Split Option 3: between High-RLC and Low-RLC

[0131] (4) Split Option 4: between RLC (Low-RLC) and MAC (High-MAC)

[0132] (5) Split Option 5: between High-MAC and Low-MAC

[0133] (6) Split Option 6: between MAC (Low-MAC) and PHY (High-PHY)

[0134] (7) Split Option 7: between High-PHY and Low-PHY

[0135] (8) Split Option 8: between PHY (Low-PHY) and RF

[0136]

[0103] The functional split point between CU and O-DU may be Split Option 2. The link between CU and O-DU is referred to as midhaul and the F1 interface is defined by the 3GPP. Further, the link between O-DU and O-RU is referred to as fronthaul and its functional split point may be Split Option 7-2x adopted as the O-RAN fronthaul specifications.

[0137]

[0104] Fig. 11 illustrates an example in which the base station functionality of the gNB is subjected to functional splitting into CU, O-DU, O-RU by Split Option 2 and Split Option 7-2x.

[0138]

[0105] CU may include, for example, an RRC (radio resource control) function, an SDAP (service data adaptation protocol) function, and a PDCP (packet data convergence protocol) function.

[0139]

[0106] O-DU may include, for example, an RLC (radio link control) function, a MAC function, and a higher physical layer (HIGH-PHY) function. Further, the HIGH-PHY function may include an encoding function, a scrambling function, a modulation function, a layer mapping function, a precoding function, and an RE (resource element) mapping function for downlink (DL) transmission. The HIGH-PHY function may also include a decoding function, a descrambling function, a demodulation function, a layer demapping function, and an RE (resource element) demapping function for uplink (UL) reception.

[0140]

[0107] O-RU may include, for example, a LOW-PHY function and an RF function. Further, the LOW-PHY function may include a beamforming function, IFFT (Inverse Fast Fourier Transform) + CP (Cyclic Prefix) addition functions, and a D / A (Digital to Analog) conversion function for downlink transmission. Further, the LOW-PHY function may include an A / D (Analog to Digital) conversion function, CP removal + FFT (Fast Fourier Transform) functions, and a beamforming function for uplink reception.

[0108] Note that, in a case where O-DU does not include the precoding function, O-RU may include the precoding function.

[0141]

[0109] O-RU may include an LBT (listen before Talk)-related function.

[0142]

[0110] eCPRI (Evolved Common Public Radio Interface) is defined as a communication scheme between O-DU and O-RU in Split Option 7-2x.

[0143]

[0111] In Split Option 7-2x, a sampling sequence of the in-phase (I) and quadrature (Q) components of an OFDM signal in the frequency domain as well as information used for beamforming in the antenna, a time synchronization signal, and the like are transmitted and received by eCPRI.

[0144]

[0112] Information transmitted by signals (PDCCH, PUCCH, PDSCH, PUSCH, MAC CE, RRC, and the like) described in each exemplary embodiment may be transmitted by using the User Plane (U-Plan) or Control Plane (C-Plane) of eCPRI between O-DU and O-RU.

[0145]

[0113] In a case where a function described in each exemplary embodiment is executed in O-RU by function splitting, O-DU may control O-RU by transmitting information for controlling the function by means of a control signal (for example, eCPRI) between O-DU and O-RU.

[0146]

[0114] In a case where a function described in each exemplary embodiment is executed by function splitting in O-DU, O-RU may receive a result of the execution of the function in O-DU by means of a control signal (for example, eCPRI) and may control O-RU based on the received result.

[0147]

[0115] CU, O-DU, and O-RU may be deployed in physically different apparatuses, the respective functions of which are connected by optical fibres or the like, or some or all of the functions may be deployed in a physically identical apparatus.

[0148]

[0116] CU and O-DU may be logical entities implemented as software operating on a server, such as a cloud, as a virtual Radio Access Network (vRAN). Further, some or all of the functions of CU and O-DU may be provided as services of a Network Functions Virtualization (NFV) function.

[0117] The transceiver may not be a radio transceiver and may be, for example, a network transceiver, an optical transceiver, or the like. The radio resource allocated by O-DU may be a resource for radio communication between O-RU and the DE.

[0149] SBFD

[0150]

[0118] Operations on uplink, downlink, and sidelink symbols in one exemplary embodiment of the present disclosure may be applied to symbols (for example, SBFD (Subband nonoverlapping full duplex) symbols, Subband full duplex) on which an SBFD operation or control is performed. For SBFD symbols, the frequency domain (or frequency resource or frequency bandwidth) is divided into a plurality of frequency domains (also referred to as, for example, sub-bands, RB sets, sub-bandwidths, or sub-BWPs (Bandwidth parts)). The terminal performs transmission and reception in a direction (for example, a downlink or uplink direction) in units of sub-bands that are the divided domains. For SBFD symbols, the terminal may perform transmission / reception in one direction of uplink and downlink directions, and may not perform transmission / reception in the other direction. The base station, on the other hand, may be capable of performing both uplink and downlink transmissions / receptions simultaneously. SBFD symbols may have a fewer frequency domain usable for downlink compared to symbols for which only downlink transmission / reception is performed. Further, SBFD symbols may have a fewer frequency domain usable for uplink compared to symbol for which only uplink transmission / reception is performed.

[0151]

[0119] Further, for SBFD symbols, the terminal may perform uplink and downlink transmissions / receptions simultaneously. At this time, the frequency domain transmitted by the terminal and the frequency domain received by the terminal may not be adjacent and a frequency interval (also referred to as a frequency gap) may be provided therebetween.

[0152]

[0120] Further, sidelink transmission / reception may also be included as a transmission / reception direction in units of sub-bands which are the divided domains.

[0153] XDD: Cross Division Duplex

[0154]

[0121] Operations on uplink, downlink, and sidelink symbols in one exemplary embodiment of the present disclosure may be applied to symbols (for example, Full duplex symbols) on which a Full duplex operation or control is performed. For Full duplex symbols, both the terminal and the base station are capable of performing uplink and downlink transmissions / receptions simultaneously. For Full duplex symbols, the terminal and the base station may operate to perform transmission / reception simultaneously in available frequency domains (or frequency resources or frequency bandwidths) or may operate to perform transmission / reception simultaneously in one or some of frequency domains (that is, may operate to perform transmission or reception in the other frequency domains). At this time, the frequency domain transmitted by the base station or the terminal and the frequency domain received by the base station or the terminal may not be adjacent and a frequency interval (also referred to as a frequency gap) may be provided therebetween. Further, for example, for the purpose of reduction in interference or the like, one of the terminal and the base station may operate to perform transmission / reception simultaneously (that is, the other may operate to perform transmission or reception).

[0155]

[0122] Further, the Full duplex operation may be applied to an operation in which the terminal is capable of performing sidelink transmission / reception simultaneously. Further, the Full duplex operation may be applied to an operation in which the terminal is capable of performing sidelink transmission / reception and uplink or downlink transmission / reception simultaneously.

[0156] Control Signals

[0157]

[0123] In the present disclosure, the downlink control signal (information) related to the present disclosure may be a signal (information) transmitted through PDCCH of the physical layer or may be a signal (information) transmitted through a MAC Control Element (CE) of the higher layer or the RRC. The downlink control signal may be a pre-defined signal (information).

[0158]

[0124] The uplink control signal (information) related to the present disclosure may be a signal (information) transmitted through PUCCH of the physical layer or may be a signal (information) transmitted through a MAC CE of the higher layer or the RRC. Further, the uplink control signal may be a pre-defined signal (information). The uplink control signal may be replaced with uplink control information (UCI), the 1ststage sidelink control information (SCI) or the 2ndstage SCI.

[0159] Base Station

[0160]

[0125] In the present disclosure, the base station may be a Transmission Reception Point (TRP), a cluster head, an access point, a Remote Radio Head (RRH), an eNodeB (eNB), a gNodeB (gNB), a Base Station (BS), a Base Transceiver Station (BTS), a base unit or a gateway, for example. Further, in sidelink communication, a terminal may be adopted instead of a base station. The base station may be a relay apparatus that relays communication between a higher node and a terminal. The base station may be a roadside unit as well. Uplink / Downlink / Sidelink

[0161]

[0126] The present disclosure may be applied to any of uplink, downlink and sidelink.

[0162]

[0127] The present disclosure may be applied to, for example, uplink channels, such as PUSCH, PUCCH, and PRACH, downlink channels, such as PDSCH, PDCCH, and PBCH, and side link channels, such as Physical Sidelink Shared Channel (PSSCH), Physical Sidelink Control Channel (PSCCH), and Physical Sidelink Broadcast Channel (PSBCH).

[0163]

[0128] PDCCH, PDSCH, PUSCH, and PUCCH are examples of a downlink control channel, a downlink data channel, an uplink data channel, and an uplink control channel, respectively. PSCCH and PSSCH are examples of a sidelink control channel and a sidelink data channel, respectively. PBCH and PSBCH are examples of broadcast channels, respectively, and PRACH is an example of a random access channel.

[0164] Data Channels / Control Channels

[0165]

[0129] The present disclosure may be applied to any of data channels and control channels. The channels in the present disclosure may be replaced with data channels including PDSCH, PUSCH and PSSCH and / or control channels including PDCCH, PUCCH, PBCH, PSCCH, and PSBCH.

[0166] Reference Signals

[0167]

[0130] In the present disclosure, the reference signals are signals known to both a base station and a mobile station and each reference signal may be referred to as a Reference Signal (RS) or sometimes a pilot signal. The reference signal may be any of a DeModulation Reference Signal (DMRS), a Channel State Information - Reference Signal (CSI-RS), a Tracking Reference Signal (TRS), a Phase Tracking Reference Signal (PTRS), a Cell-specific Reference Signal (CRS), and a Sounding Reference Signal (SRS).

[0168] Time Intervals

[0169]

[0131] In the present disclosure, time resource units are not limited to one or a combination of slots and symbols, and may be time resource units, such as frames, super-frames, subframes, slots, time slots, sub-slots, mini-slots, or time resource units, such as symbols, Orthogonal Frequency Division Multiplexing (OFDM) symbols, Single Carrier-Frequency Division Multiple Access (SC-FDMA) symbols, or other time resource units. The number of symbols included in one slot is not limited to any number of symbols exemplified in the embodiment(s) described above, and may be other numbers of symbols. Frequency Bands

[0170]

[0132] The present disclosure may be applied to any of a licensed band and an unlicensed band.

[0171] Communication

[0172]

[0133] The present disclosure may be applied to any of communication between a base station and a terminal (Uu-link communication), communication between a terminal and a terminal (Sidelink communication), and Vehicle to Everything (V2X) communication. The channels in the present disclosure may be replaced with PSCCH, PSSCH, Physical Sidelink Feedback Channel (PSFCH), PSBCH, PDCCH, PUCCH, PDSCH, PUSCH, and PBCH.

[0173]

[0134] In addition, the present disclosure may be applied to any of a terrestrial network or a network other than a terrestrial network (NTN: Non-Terrestrial Network) using a satellite or a High Altitude Pseudo Satellite (HAPS). In addition, the present disclosure may be applied to a network having a large cell size, and a terrestrial network with a large delay compared with a symbol length or a slot length, such as an ultra-wideband transmission network.

[0174] Antenna Ports

[0175]

[0135] An antenna port refers to a logical antenna (antenna group) formed of one or more physical antenna(s). That is, the antenna port does not necessarily refer to one physical antenna and sometimes refers to an array antenna formed of multiple antennas or the like. For example, it is not defined how many physical antennas form the antenna port, and instead, the antenna port is defined as the minimum unit through which a terminal is allowed to transmit a reference signal. The antenna port may also be defined as the minimum unit for multiplication of a precoding vector weighting.

[0176]

[0136] The present disclosure can be realized by software, hardware, or software in cooperation with hardware. Each functional block used in the description of each embodiment described above can be partly or entirely realized by an LSI such as an integrated circuit, and each process described in each embodiment may be controlled partly or entirely by the same LSI or a combination of LSIs. The LSI may be individually formed as chips, or one chip may be formed so as to include a part or all of the functional blocks. The LSI may include a data input and output coupled thereto. The LSI here may be referred to as an IC, a system LSI, a super LSI, or an ultra-LSI depending on a difference in the degree of integration. However, the technique of implementing an integrated circuit is not limited to the LSI and may be realized by using a dedicated circuit, a general-purpose processor, or a special-purpose processor. In addition, a FPGA (Field Programmable Gate Array) that can be programmed after the manufacture of the LSI or a reconfigurable processor in which the connections and the settings of circuit cells disposed inside the LSI can be reconfigured may be used. The present disclosure can be realized as digital processing or analogue processing. If future integrated circuit technology replaces LSIs as a result of the advancement of semiconductor technology or other derivative technology, the functional blocks could be integrated using the future integrated circuit technology. Biotechnology can also be applied.

[0177]

[0137] The present disclosure can be realized by any kind of apparatus, device or system having a function of communication, which is referred to as a communication apparatus.

[0178]

[0138] The communication apparatus may comprise a transceiver and processing / control circuitry. The transceiver may comprise and / or function as a receiver and a transmitter. The transceiver, as the transmitter and receiver, may include an RF (radio frequency) module including amplifiers, RF modulators / demodulators and the like, and one or more antennas.

[0179]

[0139] Some non-limiting examples of such a communication apparatus include a phone (e.g., cellular (cell) phone, smart phone), a tablet, a personal computer (PC) (e.g., laptop, desktop, netbook), a camera (e.g., digital still / video camera), a digital player (digital audio / video player), a wearable device (e.g., wearable camera, smart watch, tracking device), a game console, a digital book reader, a telehealth / telemedicine (remote health and medicine) device, and a vehicle providing communication functionality (e.g., automotive, airplane, ship), and various combinations thereof.

[0180]

[0140] The communication apparatus is not limited to be portable or movable, and may also include any kind of apparatus, device or system being non-portable or stationary, such as a smart home device (e.g., an appliance, lighting, smart meter, control panel), a vending machine, and any other “things” in a network of an “Internet of Things (loT)”.

[0181]

[0141] The communication may include exchanging data through, for example, a cellular system, a wireless LAN system, a satellite system, etc., and various combinations thereof.

[0182]

[0142] The communication apparatus may comprise a device such as a controller or a sensor which is coupled to a communication device performing a function of communication described in the present disclosure. For example, the communication apparatus may comprise a controller or a sensor that generates control signals or data signals which are used by a communication device performing a communication function of the communication apparatus.

[0183]

[0143] The communication apparatus also may include an infrastructure facility, such as a base station, an access point, and any other apparatus, device or system that communicates with or controls apparatuses such as those in the above non-limiting examples.

[0184]

[0144] In the following paragraphs, certain exemplifying embodiments are explained with reference to terms related to 5G core network and the present disclosure regarding communication apparatuses and methods for NW-side additional conditions for a userequipment (UE)-sided artificial intelligence (Al) or machine learning (ML) model, namely:

[0185] Example 1. Afirst communication apparatus comprising: circuitry, which in operation, determines, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or a machine learning (ML) model; and a transceiver, which in operation, transmits the one or more associated identifiers to a second communication apparatus.

[0186] Example 2. The first communication apparatus of Example 1 , wherein the transceiver receives the network vendor information identifier from a corresponding network vendor, and the circuitry determines, based on the received network vendor information identifier, the one or more associated identifiers.

[0187] Example 3. The first communication apparatus of Example 1 , wherein the circuitry allocates the network vendor information identifier to a corresponding network vendor; and determines, based on the allocated network vendor information identifier, the one or more associated identifiers.

[0188] Example 4. The first communication apparatus of Examples 2 or 3, wherein the transceiver receives the one or more associated identifiers from the corresponding network vendor, the one or more associated identifiers being used as the network vendor information identifier, and the circuitry assigns the one or more associated identifiers to the corresponding network vendor. Example 5. The first communication apparatus of any one of Examples 1-3, wherein each of the one or more associated identifiers includes the network vendor information identifier.

[0189] Example 6. The first communication apparatus of Example 5, wherein the each of the one or more associated identifier further comprises one or more of: ( / ) an index of a cell group to which the corresponding network vendor belongs, ( / / ) an index of a country, an area, a region or a zone in which the corresponding network vendor is located, (Hi) an index of public land mobile network, and (iv) an assigned index for a configuration relating to the condition for collecting the data for training and / or inferring the Al or ML model.

[0190] Example 7. The first communication apparatus of any one of Examples 1-3, wherein the network vendor information identifier is mapped to the one or more associated identifiers based on a table-based mapping or a rule, and the circuitry determines, based on the network vendor information identifier and the mapping, the one or more associated identifiers.

[0191] Example 8 The first communication apparatus of any one of Examples 1-7, wherein the condition is one or a combination of at least: a configuration of reference signal (RS) resources of a first set and / or a second set, a Quasi Go-Location (QCL) configuration for the first and / or second sets, a configuration of an association between the first and second sets, and a number of historic measurement results of the first and / or second sets.

[0192] Example 9. A second communication apparatus comprising: a transceiver, which in operation, receives, from a first communication apparatus, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model; and circuitry, which in operation, collects data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

[0193] Example 10. The second communication apparatus of Example 9, wherein the transceiver further ( / ) transmits the collected data to an entity which trains the Al or ML model, the entity being a server of a network vendor of the second communication apparatus or a third-party server’ ( / / ) receives a trained Al or ML model from the entity; and / or (Hi) transmits the trained Al or ML model to the first communication apparatus. Example 11. The second communication apparatus of Example 10, wherein the circuitry is further configured to predict one or more beams based on the trained Al or ML model and the one or more associated identifiers.

[0194] Example 12. A communication method implemented by a first communication apparatus, comprising: determining, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or a machine learning (ML) model; and transmitting the one or more associated identifiers to a second communication apparatus.

[0195] Example 13. A communication method implemented by a second communication apparatus, comprising: receiving, from a first communication apparatus, one or more associated identifiers associated with a network vendor information identifier, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model; collecting the data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

[0196] Example 14. Afirst communication apparatus comprising: circuitry, which in operation, determines the one or more associated identifiers based on one or a combination of ( / ) an index of a cell group in which includes a plurality of cells, ( / ' / ) an index of an area, a region or a zone includes another plurality of cells, and ( / / / ) an index of a public land mobile network of a corresponding network operator; and a transceiver, which in operation, transmits the one or more associated identifiers to a second communication apparatus.

[0197] Example 15. The first communication apparatus of Example 14, wherein the transceiver receives, from a corresponding network vendor, the one or a combination of ( / ) the index of the cell group in which includes the plurality of cells, ( / / ) the index of the area, the region or the zone includes the another plurality of cells, and (Hi) the index of the public land mobile network of the corresponding network operator, and the circuitry determines, based on the received one or a combination of ( / ) the index of the cell group in which includes the plurality of cells, (ii) the index of the area, the region or the zone includes the another plurality of cells, and (Hi) the index of the public land mobile network of the corresponding network operator, the one or more associated identifiers.

[0198] Example 16. The first communication apparatus of Example 14, wherein the circuitry allocates the one or a combination of ( / ) the index of the cell group in which includes the plurality of cells, (ii) the index of the area, the region or the zone includes the another plurality of cells, and (Hi) the index of the public land mobile network of the corresponding network operator to a corresponding network vendor; and determines, based on the allocated one or a combination of ( / ) the index of the cell group in which includes the plurality of cells, (ii) the index of the area, the region or the zone includes the another plurality of cells, and (iii) the index of the public land mobile network of the corresponding network operator, the one or more associated identifiers.

[0199] Example 17. The first communication apparatus of Examples 15 or 16, wherein the transceiver further receives the one or more associated identifiers from the corresponding network vendor, and the circuitry associates the network vendor information identifier to the one or more associated identifiers and determines, based on the network vendor information identifier, the one or more associated identifiers.

[0200] Example 18. The first communication apparatus of any one of Examples 14-16, wherein each of the one or more associated identifiers includes the one or a combination of (i) the index of the cell group in which includes the plurality of cells, (ii) the index of the area, the region or the zone includes the another plurality of cells, and (iii) the index of the public land mobile network of the corresponding network operator.

[0201] Example 19. The first communication apparatus of Example 18, wherein the each of the one or more associated identifier further comprises (iv) an assigned index for a configuration relating to the condition for collecting the data for training and / or inferring the Al or ML model.

[0202] Example 20. The first communication apparatus of any one of Examples 14-17, wherein the one or a combination of (i) the index of the cell group in which includes the plurality of cells, (ii) the index of the area, the region or the zone includes the another plurality of cells, and (Hi) the index of the public land mobile network of the corresponding network operator is mapped to the one or more associated identifiers based on a table-based mapping or a rule, and the circuitry determines, based on the mapping and the one or a combination of ( / ) the index of the cell group in which includes the plurality of cells, (ii) the index of the area, the region or the zone includes the another plurality of cells, and (Hi) the index of the public land mobile network of the corresponding network operator, the one or more associated identifiers.

[0203] Example 21 . The first communication apparatus of any one of Examples 14-20, wherein the condition is one or a combination of at least: a configuration of reference signal (RS) resources of a first set and / or a second set, a Quasi Co- Location (QCL) configuration for the first and / or second sets, a configuration of an association between the first and second sets, and a number of historic measurement results of the first and / or second sets.

[0204]

[0145] It will be appreciated by a person skilled in the art that numerous variations and / or modifications may be made to the present disclosure as shown in the specific embodiments without departing from the spirit or scope of the disclosure as broadly described. The present embodiments are, therefore, to be considered in all respects illustrative and not restrictive.

Claims

CLAIMS1. A first communication apparatus comprising: circuitry, which in operation, determines, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or a machine learning (ML) model; and a transceiver, which in operation, transmits the one or more associated identifiers to a second communication apparatus.

2. The first communication apparatus of claim 1 , wherein the transceiver receives the network vendor information identifier from a corresponding network vendor, and the circuitry determines, based on the received network vendor information identifier, the one or more associated identifiers.

3. The first communication apparatus of claim 1 , wherein the circuitry allocates the network vendor information identifier to a corresponding network vendor; and determines, based on the allocated network vendor information identifier, the one or more associated identifiers.

4. The first communication apparatus of claim 2, wherein the transceiver receives the one or more associated identifiers from the corresponding network vendor, the one or more associated identifiers being used as the network vendor information identifier, and the circuitry assigns the one or more associated identifiers to the corresponding network vendor.

5. The first communication apparatus of claim 1 , wherein each of the one or more associated identifiers includes the network vendor information identifier.

6. The first communication apparatus of claim 5, wherein each of the one or more associated identifiers comprises one or more of: ( / ) an index of a cell group to which the corresponding network vendor belongs, ( / / ) an index of a country, an area, a region or a zone in which the corresponding network vendor is located, ( / / / ) an index of public land mobile network, and ( / ) an assigned index for a configuration relating to the condition for collecting the data for training and / or inferring the Al or ML model.

7. The first communication apparatus of claim 1 , wherein the network vendor information identifier is mapped to the one or more associated identifiers based on a tablebased mapping or a rule, and the circuitry determines, based on the network vendor information identifier and the mapping, the one or more associated identifiers.

8. The first communication apparatus of claim 1 , wherein the condition is one or a combination of at least: a configuration of reference signal (RS) resources of a first set and / or a second set, a Quasi Co-Location (QCL) configuration for the first and / or second sets, a configuration of an association between the first and second sets, and a number of historic measurement results of the first and / or second sets.

9. A second communication apparatus comprising: a transceiver, which in operation, receives, from a first communication apparatus, one or more associated identifiers, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model; and circuitry, which in operation, collects data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

10. The second communication apparatus of claim 9, wherein the transceiver further ( / ) transmits the collected data to an entity which trains the Al or ML model, the entity being a server of a network vendor of the second communication apparatus or a third-party server; ( / 7) receives a trained Al or ML model from the entity; and / or ( / / / ) transmits the trained Al or ML model to the first communication apparatus.11 . The second communication apparatus of claim 10, wherein the circuitry is further configured to predict one or more beams based on the trained Al or ML model and the one or more associated identifiers.

12. A communication method implemented by a first communication apparatus, comprising: determining, based on a network vendor information identifier, one or more associated identifiers, each of the one or more associated identifiers representing acondition for collecting data for training and / or inferring an artificial intelligence (Al) or a machine learning (ML) model; and transmitting the one or more associated identifiers to a second communication apparatus.

13. A communication method implemented by a second communication apparatus, comprising: receiving, from a first communication apparatus, one or more associated identifiers associated with a network vendor information identifier, each of the one or more associated identifiers representing a condition for collecting data for training and / or inferring an artificial intelligence (Al) or machine learning (ML) model; collecting the data based on the one or more associated identifiers, the data being measurement results of one or more reference signals.

Images