hash
stringlengths
32
32
doc_id
stringlengths
5
12
section
stringlengths
5
1.47k
content
stringlengths
0
6.67M
a42d942a68c46b17b4d16e1f71329da3
38.765
4.1 Definitions of performance metrics
In the evaluation of NR ISAC, the following performance metrics are considered: - Horizontal/vertical positioning accuracy is defined as the absolute value of the difference between the estimated horizontal/vertical position and the corresponding true position of a sensing target. NOTE: There should be only one estimated horizontal/vertical position corresponding to the true position of a sensing target. - Velocity accuracy is defined as the absolute value of the difference between the estimated velocity and the corresponding true velocity of a sensing target. For single TRP monostatic sensing, both the radial velocity accuracy and the 3D velocity accuracy can be estimated. The true radial velocity is the projection of true 3D velocity on the direction from TRP to target for TRP monostatic. NOTE: There should be only one estimated velocity corresponding to the true velocity of a sensing target. - Missed detection probability is defined as the conditional probability of not detecting the presence of a target when the target is actually present in the simulation area. where, - is the number of missed targets in the drop n, i.e., the true target not associated with any detected object. - is the number of true targets in the drop n. - is total number of drops with at least one target per drop - False alarm probability Type 1 is defined for cases without true target dropped in simulation area. An object is detected when there is no target present in simulation area is considered a false alarm. where, - equal to 1 if at least one object is detected when there is no target dropped in the simulation area in the drop n, otherwise equal to 0. - is the total number of drops without targets in the simulation area. - False alarm probability Type 2 is defined for cases with true target dropped in simulation area. An object is detected but not associated with any true target in the simulation area is considered as a false alarm. where, - is the number of detected objects but not associated with any true targets in the drop n. - is the total number of detected objects in the drop n. - is number of drops with at least one detected object. NOTE: Both False Alarm Probability Types are mandatory. Sensing resolution, sensing service latency and refreshing rate are not considered as performance metrics for the evaluation of NR ISAC. For the purpose of performance metric calculation, association of the detected object(s) and the true target(s) should fulfil at least the following conditions: - One true target is associated with at most one detected object. - One detected object is associated with at most one true target. - The same association applies to miss detection, false alarm probability Type 2 and positioning/velocity accuracy. Companies should report the method used for association of the detected object(s) and the true target(s).
a42d942a68c46b17b4d16e1f71329da3
38.765
4.2 Performance objectives
The following performance objectives are adopted for the evaluation of UAV use case with gNB-based monostatic sensing. Table 4.2-1: Performance objectives Performance metrics Values Missed detection Probability 5% False Alarm Probability Type 1 5% False Alarm Probability Type 2 5% Horizontal Positioning Accuracy 10 m with confidence level 90% Vertical Positioning Accuracy 10 m with confidence level 90% Velocity Accuracy 5 m/s with confidence level 90% NOTE: Confidence level of the X% represents X percentile point of the cumulative distribution function (CDF) of the estimation errors
a42d942a68c46b17b4d16e1f71329da3
38.765
5 Measurements
Editor’s note: This section is to include the definitions of measurement metrics, and measurement quantization except for the related evaluation results. From physical layer perspective, the following measurements that may be reported from RAN are identified in the study of NR ISAC. - Level A: Raw data per Tx antenna port per OFDM symbol per RX antenna port per TRP for a given time stamp - Option A1: Amplitude and phase samples in time/delay domain of the estimated channel, i.e., Amplitude and phase values of channel impulse response - Option A2: Amplitude and phase per subcarrier in frequency domain of the estimated channel - Level B: Amplitude and phase profile of delay, and/or Doppler, and/or angle per TRP for a given time stamp by using window(s) of the [consecutive] samples in delay, Doppler and/or angle domain - Option B1: Delay-Doppler profile per Tx antenna port per Rx antenna port, which includes the amplitude and phase samples distributed across different delays and Doppler shifts. - Option B2: Delay-Angle profile per Tx antenna port per OFDM symbol, which includes the amplitude and phase samples distributed across different delays and spatial angles (e.g., Angle of Arrival). - Option B3: Delay-Doppler-Angle profile per Tx antenna port, which includes the amplitude and phase samples distributed across delay, Doppler, and angle domains. - Option B4: Delay profile per Tx antenna port per OFDM symbol per Rx antenna port, which includes the amplitude and phase samples distributed across different delays. NOTE: Level B is applicable for either LCS or GCS - Level C: per detected path/point measurements per Tx antenna port per TRP which may be reflected/scattered from scattering point(s). - Option C1: Delay/range, Doppler/velocity, one or multiple 3D angles, and [power/confidence metric] per detected path for a given time stamp. A path is associated with one couple {delay/range, Doppler/velocity, [power/confidence metric]}, and one or multiple 3D angles. - Option C2: Doppler/velocity, position, and [power/confidence metric] per detected path for a given time stamp. A path is associated with one doppler/velocity, [power/confidence metric], one or multiple positions - Option C3: Delay/range, Doppler/velocity, 3D angle, and [power/confidence metric] per detected path for a given time stamp. A path is associated with one delay, and one or multiple triple {Doppler/velocity, 3D angle, [power/confidence metric]} - Option C4: Delay/range, Doppler/velocity, 3D angle, and [power/confidence metric] per detected point for a given time stamp A point is associated with one range/delay, one Doppler/velocity, one 3D angle - Option C5: Position, velocity, and [power/confidence metric] per detected point for a given time stamp. A point is associated with one position, one velocity NOTE: Position can be defined in either LCS or GCS. Angle can be defined in either LCS or GCS. 3D angle refers to a pair of horizontal and vertical angles. - Level D: Object/target level measurement [per TRP or per gNB]. One or more value pair(s) {position, velocity} in GCS for a given time stamp is reported for a detected object/target - Option D1: Only one value pair {position, velocity} in GCS for a given time stamp is reported for a detected object/target. The association of multiple measurements across different time stamps for the same detected object/target is not reported. - Option D2: Only one value pair {position, velocity} in GCS for a given time stamp is reported for a detected object/target. The association of multiple measurements across different time stamps for the same detected object/target is reported. - Option D3: One or more value pairs {position, velocity} in GCS for a given time stamp are reported for a detected object/target. The association of multiple measurements across different time stamps for the same detected object/target is reported. - Option D4: One or more value pairs {position, velocity} in GCS for a given time stamp are reported for a detected object/target. The association of multiple measurements across different time stamps for the same detected object/target is not reported. NOTE: Velocity is 3D velocity for Option D2/D3. NOTE: For Level A/B/C, Tx antenna port means reference signal antenna port for sensing purpose.
a42d942a68c46b17b4d16e1f71329da3
38.765
6 Performance evaluation
Editor’s note: This section is to summarize the evaluation assumptions for UAV sensing per RAN1 agreements. Details in Annex A. Editor’s note: This section is to summarize on evaluation results for UAV sensing per RAN1 agreements. Details in Annex B. - It includes the evaluation results based on NR waveform and DL NR reference signal - Depending on RAN1 discussions, it can also include other results based on other waveform and reference signals - It includes the evaluation results considering measurement quantization.
a42d942a68c46b17b4d16e1f71329da3
38.765
6.1 Evaluation methodologies
The following general procedure for performance evaluation of NR ISAC is provided. 1) Simulation parameter configuration 2) Sensing scenario generation, including the deployment of sensing Tx/Rx (STXs/SRXs) 3) Dropping N target(s), where N is equal to 0 or larger than 0 4) Channel generation and STX/SRX determination for the targets 5) Sensing signal generation and passing the sensing signal to the generated channel 6) Sensing signal processing at each SRX, optionally, sensing signal processing based on fusion from multiple STXs/SRXs - E.g., optionally, sensing signal processing based on tracking 7) Sensing performance metric calculation.
a42d942a68c46b17b4d16e1f71329da3
38.765
6.2 Evaluation Assumptions
The evaluation assumptions for the evaluation of UAV use case with gNB-based monostatic sensing are provided in Annex A. In order to define the sensing resource ratio used in the evaluation of NR ISAC, three kinds of resources are defined - Type_1: Resources that are used for sensing signal transmission - Type_2: Part of Type_1 resources that are used for communication purpose NOTE: It is possible the Type_2 resource doesn’t exist. - Type_3: Resources that are not used for sensing signal transmission, and cannot be used for communication purpose due to sensing operation Two options are provided to calculate the sensing resource ratio. Both options should be reported by companies. - Option 1: (Type_1 + Type_3) resources over all radio DL and UL resources - Option 2: (Type_1 - Type_2 + Type_3) resources over all radio DL and UL resources NOTE: If Type_2 resource doesn’t exist, two options are the same In the evaluation on NR ISAC, company should report which sensing RS resources are considered as Type 2 resource and related reason. Multiple configurations of the following parameters are defined to respectively analyse the evaluation results. Two baseline configurations are defined for evaluation purpose. Table 6.1-1: Baseline configurations of evaluation assumptions Parameters Baseline configuration 1 Baseline configuration 2 1 Scenario (UMa-AV, 500m) (UMa-AV, 500m) 2 Sensing Tx/Rx operating simultaneously or not Sensing Tx/Rx operating simultaneously Sensing Tx/Rx operating simultaneously 3 Carrier frequency 4 or 4.9GHz 4 or 4.9GHz 4 Max BS Tx power 52 dBm 37 dBm 5 BS antenna configuration (M, N, P, Mg, Ng, Mp, Np) Tx: (8,8,2,1,1;4,8) Rx: (8,8,2,1,1;4,8) = (0.5, 0.8)λ, +45°/-45° polarization (M, N, P, Mg, Ng, Mp, Np) Tx: (8,8,2,1,1;4,8) Rx: (8,8,2,1,1;4,8) = (0.5, 0.8)λ, +45°/-45° polarization 6 Number of targets per sector in the center site N=5 N=5 7 Target vertical distribution 25-300m 25-300m ◦ Additional configurations can be defined for other assumptions of the parameters, based on reported evaluation results in RAN1 #124.
a42d942a68c46b17b4d16e1f71329da3
38.765
6.3 Performance evaluation results
a42d942a68c46b17b4d16e1f71329da3
38.765
6.3.1 Baseline configuration 1
[Editor’s note] this section is to capture observations on results for baseline configuration 1
a42d942a68c46b17b4d16e1f71329da3
38.765
6.3.2 Baseline configuration 2
[Editor’s note] this section is to capture observations on results for baseline configuration 2 6.3.3 [Configuration x] [Editor’s note] this section is to capture observations on results for other configurations if agreed. This section may be split to one or more sections.
a42d942a68c46b17b4d16e1f71329da3
38.765
7 Network architecture
Editor’s note: This section is to capture the study outcome of network architecture. Applicability to gNB bistatic sensing may be considered as part of this network architecture without additional architecture impacts. No inter-gNB coordination will be studied.
a42d942a68c46b17b4d16e1f71329da3
38.765
8 RAN-CN procedures and signalling
Editor’s note: This section is to capture the study outcome of procedures and signaling aspects between RAN and CN for gNB-based monostatic sensing.
a42d942a68c46b17b4d16e1f71329da3
38.765
9 Conclusions
Editor’s note: TBA. Annex <A>: Evaluation assumptions Editor’s note: This annex is to include the agreed evaluation assumptions for UAV sensing per RAN1 agreements, which complements Clause 6. - It includes the evaluation assumptions based on NR waveform and DL NR reference signal - Depending on RAN1 discussions, it can also include other assumptions based on other waveforms and reference signals In this clause, the evaluation assumptions for the evaluation of UAV use case with gNB-based monostatic sensing are provided. When sensing Tx/Rx operates simultaneously, the assumptions are summarized in Table A-1. Table A-1: Evaluation assumptions Parameters Assumptions FR1 FR2-1 (Optional) Scenario UMa-AV, optional RMa-AV UMi-AV Carrier frequency 4 or 4.9 GHz Optional for FR1: 6 GHz 30 GHz System bandwidth 100 MHz 400 MHz Numerology SCS = 30 kHz SCS = 120 kHz BS Layout Hexagonal grid, 7 macro sites, 3 sectors per site. 3 sectors with 30, 150, 270 degrees Inter-BS (2D) distance UMa-AV: 500 m, optional 1000 m RMa-AV: 1732 m 200 m Wrap-round No wrap-round BS antenna height UMa-AV: 25 m RMa-AV: 35 m 10 m BS antenna configuration (M, N, P, Mg, Ng; Mp, Np) for 4GHz, 4.9GHz - Configuration A - Tx: (8,8,2,1,1;4,8) - Rx: (8,8,2,1,1;4,8) - Configuration B as optional - Tx: (12,16,2,1,1;2,16) - Rx: (12,16,2,1,1;2,16) (M, N, P, Mg, Ng; Mp, Np) for 6GHz - Configuration A - Tx: (8,8,2,1,1;4,8) - Rx: (8,8,2,1,1;4,8) - Configuration B as optional - Tx: (16,16,2,1,1;4,16) - Rx: (16,16,2,1,1;4,16) = (0.5, 0.8)λ, +45°/-45° polarization Optional: = (0.5, 0.5)λ, +45°/-45° polarization (M, N, P, Mg, Ng; Mp, Np) - Configuration A - Tx: (16,16,2,1,1;1,1) - Rx: (16,16,2,1,1;1,1) = (0.5, 0.5 or 0.8)λ, +45°/-45° polarization BS antenna radiation pattern Table 9 in Report ITU-R M.2412 BS antenna mechanic tilt 90° in GCS (pointing to horizontal direction) BS antenna electrical tilt Option 1: no electrical tilt Option 2: 102° in GCS Polarized antenna model Model-2 in clause 7.3.2 in TR 38.901 Antenna isolation 65 dB, 80 dB 80 ~ 100 dB Max BS Tx power 37 dBm, 52 dBm. See note 1 30 dBm Inter-site interference Not modelled Self-interference The residual leakage interference/noise is modelled e.g. by additional additive white Gaussian noise, -94+X dBm in 100 MHz, X is up to company report. Companies to provide details on their modelling. See note 2 Co-site inter-sector interference Not modelled Adjacent channel interference Not modelled BS receiver noise figure 5 dB 7 dB Sensing target Target type UAV with small size (0.3m x 0.4m x 0.2m) 3D distribution N targets per sector in the center site only - N = 5 - Optional: N is uniformly distributed from 1 to 10 Horizontal plane: Uniformly distributed in a sector Vertical plane: Uniformly distributed between 25m and 300m, optionally between 1.5m and 300m. Mobility Horizontal speed: uniformly distributed between 0 and 180km/h Vertical speed: 0km/h Minimum BS-target 3D distance 10 m Minimum target-target (3D) distance 10 m Outdoor/indoor proportion 100% outdoor LOS/NLOS LOS and NLOS Orientation Random in horizontal domain RCS model RCS model 1 for UAV with small size gNB-target link TRP-UAV link in Table 7.9.3-2 in TR 38.901, using Clause B.1.3 in TR 36.777 Concatenation of TX-target and target-RX links Up to company choice between two options for concatenation defined in Step 9 in clause 7.9.4.1 in TR 38.901 The power threshold for path dropping after concatenation for target channel -25 dB and -40 dB are respectively used for the two options for concatenation NOTE 1: The above options are calculated with BS_maxpower = BS Rx saturation power + antenna isolation by assuming the BS Rx saturation power = -28dBm and the antenna isolation = 65dB and 80dB, respectively. NOTE 2: X = -Infinity corresponds to not modelling self-interference. Besides the evaluation parameters provided in Table A-1, the following assumptions are up to company report: - To model self-interference, value of X to derive the power of the additional additive white Gaussian noise to model the residual leakage interference/noise. - Length of Coherent Processing Interval (CPI). - Tx beam information (number of Tx beams, wide/narrow Tx beam) being used at TRP. - RE mapping of sensing RS, and assumed TDD UL/DL configuration if applicable. - Sensing resource ratio. - High-level sensing signal/data processing method, e.g., 2D FFT, MUSIC, and any other methods. - Optionally, the maximum BS Tx power when it is assumed that Tx and Rx don’t operate simultaneously. Companies should report how the maximum BS Tx power is derived. - Sensing signal processing and ISAC channel generation. - Whether a same target is modelled in the ISAC channel of single, multiple or all STXs/SRXs? - Company should report how to determine the single or multiple STXs/SRXs for a target. - If the evaluation results are derived by measurement reports from multiple/all STXs/SRXs, companies should report how measurement reports from multiple/all STXs/SRXs are used. - Beam set at TRxP for FR2-1. - Additional configuration information for FR2-1. - How target trajectory is modeled if evaluation results on UAV tracking is reported. Annex <B>: Evaluation results Editor’s note: This annex is to include the detailed evaluation results for UAV sensing per RAN1 agreements, which complements Clause 6. Annex <X>: Change history Change history Date Meeting TDoc CR Rev Cat Subject/Comment New version 2025-10 RAN1 #122-bis R1-2507422 TR Skeleton 0.0.1 2025-11 RAN1 #123 R1-2509522 RAN1 agreements till RAN1 #122bis 0.1.1 2025-12 RAN1 #123 R1-2509632 RAN1 agreements in RAN1 #123 0.2.0
d8a2582e68c28448160ac2de7acb3004
38.760-3
1 Scope
The present document covers the Radio Access Architecture and Interface aspects of the study item “Study on 6G Radio” [2]. 2 Time Lines
d8a2582e68c28448160ac2de7acb3004
38.760-3
2.1 Expected Output and Time scale
New specifications Type TS/TR number Title For info at TSG# For approval at TSG# Remarks External TR 38.960 Study on 6G Radio TSG#115 TSG#116 This TR is led by RAN TR editor: Kumagai, Shinya, NTT DOCOMO, shinya.kumagai.yw@nttdocomo.com Note: The pCR(s) to this TR will be provided by RAN WGs at the end of the SI. Internal TR 38.870-1 Study on 6G Radio RAN1 aspects TSG#114 TSG#115 This TR is led by RAN1 TR editor: Kumagai, Shinya, NTT DOCOMO, shinya.kumagai.yw@nttdocomo.com Internal TR 38.870-2 Study on 6G Radio RAN2 aspects TSG#115 TSG#116 This TR is led by RAN2 TR editor: Chai, Li, China Mobile, chaili@chinamobile.com Internal TR 38.870-3 Study on 6G Radio RAN3 aspects TSG#115 TSG#116 This TR is led by RAN3 TR editor: Kulakov, Alexey, Vodafone, Alexey.Kulakov1@vodafone.com Internal TR 38.870-4 Study on 6G Radio RAN4 aspects TSG#115 TSG#116 This TR is led by RAN4 TR editor: Schumacher, Joseph, AT&T, jq304t@att.com TSG#115: March 2027, TSG#116 June 2027
d8a2582e68c28448160ac2de7acb3004
38.760-3
2.2 Interim Milestone
TSG#112 (June/2026): RAN3 to provide interim study results to allow TSGs to make a decision on: - RAN-CN interface: P2P vs SBI - RAN internal interfaces: CU-DU split, CP-UP split.
d8a2582e68c28448160ac2de7acb3004
38.760-3
3 References
The following documents contain provisions which, through reference in this text, constitute provisions of the present document. - References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific. - For a specific reference, subsequent revisions do not apply. - For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document. [1] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications". [2] 3GPP RP-252912: “New SID: Study on 6G Radio” [3] 3GPP TR 38.914: “Study on 6G Scenarios and Requirements”
d8a2582e68c28448160ac2de7acb3004
38.760-3
4 Definitions of terms, symbols and abbreviations
d8a2582e68c28448160ac2de7acb3004
38.760-3
4.1 Terms
For the purposes of the present document, the terms given in TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in TR 21.905 [1]. example: text used to clarify abstract rules by applying them literally. Standalone 6G RAN: refers to the support of 6GR only, connecting exclusively with a Core Network for 6G. Service awareness: refers to enabling the RAN to acquire and use relevant information to appropriately handle a service. Resilience: refers to enabling the RAN to recover from failures and to ensure accessibility and service continuity.
d8a2582e68c28448160ac2de7acb3004
38.760-3
4.2 Symbols
For the purposes of the present document, the following symbols apply: <symbol> <Explanation>
d8a2582e68c28448160ac2de7acb3004
38.760-3
4.3 Abbreviations
For the purposes of the present document, the abbreviations given in TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in TR 21.905 [1]. 6GR 6G Radio
d8a2582e68c28448160ac2de7acb3004
38.760-3
5 Objectives and requirements
The detailed objectives of the study are: Single technology framework based on a stand-alone architecture to support the agreed existing and new services, and to satisfy the usage scenarios, requirements, deployment scenarios and design principles with acceptable performance/complexity trade-off, as determined by the RAN requirements in [RP-250810] and [TR38.914], including: [RAN1], [RAN2], [RAN3], [RAN4]
d8a2582e68c28448160ac2de7acb3004
38.760-3
5.1 General Principles
All requirements of TR 38.914 will serve as the basis for RAN3 architecture design principles. NOTE 1: If needed, RAN3 will check with relevant working groups on issues concerning security and data privacy. The 6G RAN architecture and interfaces shall allow for different implementations, e.g. virtualized, cloud-based or dedicated hardware. The 6G RAN architecture shall enable the introduction of new services flexibly and efficiently, e.g., minimizing the impact on previously deployed 6G features and services where possible. The 6G RAN architecture shall support self-configuration and self-optimization. FFS on which features should be recommended to be standardized from day1. The RAN shall support adaptation of RAN resources allocation based on service characteristics awareness to meet specific service requirements. The RAN shall support recovery mechanisms to ensure accessibility and service continuity under RAN disruption conditions. [FFS on other conditions]. The 6G RAN architecture supports data collection according to the following principles: - Reusability of collected data is supported. - Data collected or generated by the 6G RAN can be used by the RAN and it can be made available to other entities, if needed. FFS which entities. - The 6G RAN can request data, and use the data collected from other entities. FFS which entities. NOTE 2: The collection, storage and usage of data follows the security principles which will be defined by relevant WGs.
d8a2582e68c28448160ac2de7acb3004
38.760-3
5.2 Deployment Scenarios
d8a2582e68c28448160ac2de7acb3004
38.760-3
5.2.1 General
This section may be used to describe the details/solutions related to deployment scenarios as per 38.914. The 6G RAN architecture shall strive to support the deployment scenarios defined in TR 38.914. - FFS on the implications of this requirement on 6G RAN architecture. - FFS whether all deployment scenarios of TR 38.914 can be supported.
d8a2582e68c28448160ac2de7acb3004
38.760-3
5.2.2 Shared RAN deployment
6GR should support shared RAN deployments, including support of multiple participating operators. A shared RAN should be able to efficiently interoperate with a non-shared RAN. Mobility and service continuity between the non-shared RAN and the shared RAN shall be supported. Figure 5.2.x-1: Shared RAN deployment
d8a2582e68c28448160ac2de7acb3004
38.760-3
5.3 Overall RAN Architecture
Editor’s Note: Figure 5.3-1 and title aim at providing a single framework based on a stand-alone architecture for existing and new services, taking into account legacy communication services, and is subject to study progress including new services. The names of the logical nodes and the interfaces are FFS. The 6G RAN consists of 6G RAN logical nodes. A 6G RAN logical node provides 6G U-plane and C-plane protocol terminations towards the UE. The 6G RAN logical nodes may be connected with each other by means of a 6G RAN node to 6G RAN node interface. The 6G RAN logical nodes are connected to the CN for 6G by means of an interface between 6G RAN node and CN for 6G. The 6G-RAN logical nodes may be connected to one or more CN(s) for 6G. The 6G RAN architecture is illustrated in Figure 5.3-1. Figure 5.3-1 The RAN Architecture 5.3.1 RAN-CN Functional Split  The aim of this section is to describe functions split between RAN and CN. For 5G legacy features which are to be supported in 6G, base the feature design on the RAN-CN functional split for the 5G RAN.
d8a2582e68c28448160ac2de7acb3004
38.760-3
5.3.2 RAN Functions
The aim of this section is to describe the functions supported in RAN.
d8a2582e68c28448160ac2de7acb3004
38.760-3
6 RAN Architecture and Interfaces
The aim of this section is to describe all aspects related to the Interfaces between RAN and CN. 6.1 RAN-CN Interface
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.1.1 General Principles
The aim of this section is to describe general design principles and requirements for RAN-CN Interface. The general principles for the specification of the 6G RAN-CN interface are as follows: - the 6G RAN-CN interface supports the exchange of signalling information between the RAN and CN; - the 6G RAN-CN interface supports control plane and user plane separation; - the 6G RAN-CN interface supports future enhancements; - the 6G RAN-CN interface supports all possible RAN deployment scenarios; - the 6G RAN-CN interface supports RAN sharing between multiple operators; - the 6G RAN-CN interface supports the operation of network slicing; - the 6G RAN-CN interface supports enhanced service awareness in RAN; - the 6G RAN-CN control plane interface supports reliable signalling transmission; - the 6G RAN-CN interface is designed with a clear functional split between RAN and CN.
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.1.2 RAN-CN Interface Functions
RAN-CN control plane interface supports following functions: - UE context management : The functionality to manage UE context between the RAN and CN; - Transport of NAS messages: The functionality to transfer NAS messages between the CN and UE, subject to SA2 progress; - PDU Session Management: The functionality to manage PDU sessions and respective RAN resources, subject to SA2 progress.
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.1.3 RAN-CN Interface Options
This chapter includes description of RAN-CN interface options including protocol stacks, considering new and existing services. 6.1.3.1 RAN-CN Control Plane Editor’s Note: The encoding (e.g., ASN.1) for control plane interface options below is FFS. 6.1.3.1.1 Point to Point (P2P) A RAN-CN P2P interface refers to application layer communication between the 6G RAN node and the CN entity for 6G by means of elementary procedures, either triggered by the 6G RAN node or by the CN entity for 6G. Editor’s Note 1: FFS whether multiple CN entities can be involved. Potential options for the 6G P2P protocol stack are as follows: Editor's Note 2: Other options are not precluded. - SCTP based - QUIC based 6.1.3.1.2 Service Based Interface (SBI) A RAN-CN SBI (service-based interface) refers to application layer communication between the 6G RAN node and the CN entity for 6G by means of services provided/exposed by either the 6G RAN node or the CN entity for 6G. Editor’s Note 1: FFS whether multiple CN entities can be involved. Potential options for the 6G SBI protocol stack are as follows: Editor's Note 2: Other options are not precluded. - TCP+ HTTP/2 based - QUIC+HTTP/3 based
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.1.3.2 RAN-CN User Plane
Editor’s Note 1: The aim of this section is to describe the user plane protocol stack for the RAN-CN interface. Editor’s Note 2: On hold waiting for CT4 input. The user plane interface between the 6G RAN and CN for 6G is based on a point-to-point tunnel.
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.1.4 Evaluation of RAN-CN Interface Options
This chapter includes evaluation and comparison of RAN-CN interface options described in clause 6.1.3.1
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.2 RAN Internal Interfaces
This chapter shall comprehensively address study on CU-DU and CP-UP aspects including functional split considering interim milestone TSG#112 (June/2026) to take a decision during TSG#115: March 2027 This chapter should also address the interfaces between 6G RAN nodes (equivalent to Xn).
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.2.1 General Principles
The aim of this section is to describe general design principles and requirements for RAN Internal Interfaces
d8a2582e68c28448160ac2de7acb3004
38.760-3
6.2.2 Disaggregated RAN Architecture
RAN3 acknowledges that there are benefits of HLS: - RAN3 will list benefits The main areas of study (as a starting point) RAN3 is going to address within this study item for HLS are: - UE context handling between CU and DU - F1-U interface optimizations (e.g. flow control) NOTE: Additional areas of study are FFS.
d8a2582e68c28448160ac2de7acb3004
38.760-3
7 AI/ML for RAN
d8a2582e68c28448160ac2de7acb3004
38.760-3
7.1 High-level principles
The following high-level principles apply for the 6G RAN AI/ML study: - The study focuses on AI/ML functionality and corresponding input/output/feedback data. - The study focuses on data needed at the Model Training function. The aspects of how the Model Training function uses this data to train a model are out of RAN3 scope. - The study focuses on data needed at the Model Inference function. The aspects of how the Model Inference function uses this data to derive outputs are out of RAN3 scope. - The input/output/feedback data and the location of the Model Training and Model Inference function should be studied case-by-case. - The design of AI/ML algorithms and models for RAN3-led use cases is implementation-specific and out of RAN3 scope.
d8a2582e68c28448160ac2de7acb3004
38.760-3
7.2 AI/ML use cases
Identify Use Case(s) of interest (either existing or new) with compelling trade-off between e.g., performance, complexity, etc… Editor’s Note: Focus on clarifying the areas where AI/ML can be applied, the scenario description and the problem statement.
d8a2582e68c28448160ac2de7acb3004
38.760-3
7.3 AI/ML framework
d8a2582e68c28448160ac2de7acb3004
38.760-3
8 Radio access network procedures
The intention of this section is to describe as a minimum Mobility scenarios. The section might also cover migration options if agreed by the plenary. 8.1 Mobility
d8a2582e68c28448160ac2de7acb3004
38.760-3
8.1.1 Intra 6G Mobility
d8a2582e68c28448160ac2de7acb3004
38.760-3
8.1.2 Inter-RAT Mobility
This section is to describe inter-RAT mobility, at least between the 6GR and NR.
d8a2582e68c28448160ac2de7acb3004
38.760-3
8.1.2.1 Mobility between 6GR and 5G NR
X Conclusions This section is to provide interim study results to allow TSGs to make a decision during TSG#112. X.1 RAN-CN interface X.2 RAN internal interfaces Annex A: Void This section is intended to capture considerations related to aspects that fall outside the scope of 3GPP specifications but may nonetheless impact the content or conclusions of this Technical Report (TR). Annex B (informative): Change history Change history Date Meeting TDoc CR Rev Cat Subject/Comment New version Oct 2025 RAN3#129bis R3-257234 0.1.1 Nov 2025 RAN3#130 0.2.0
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
1 Scope
The present document reports the RAN1 aspects of the study item “Study on 6G Radio” [2].
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
2 References
The following documents contain provisions which, through reference in this text, constitute provisions of the present document. - References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific. - For a specific reference, subsequent revisions do not apply. - For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document. [1] 3GPP TR 21.905: “Vocabulary for 3GPP Specifications”. [2] 3GPP RP-251881: “New SID: Study on 6G Radio”. [3] 3GPP TR 38.914: “Study on 6G Scenarios and Requirements”.
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
3 Definitions of terms, symbols and abbreviations
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
3.1 Terms
For the purposes of the present document, the terms given in TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in TR 21.905 [1]. example: text used to clarify abstract rules by applying them literally.
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
3.2 Symbols
For the purposes of the present document, the following symbols apply: <symbol> <Explanation>
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
3.3 Abbreviations
For the purposes of the present document, the abbreviations given in TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in TR 21.905 [1]. <ABBREVIATION> <Expansion>
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
4 Overview of 6GR air interface
Editor’s Note: High level design proposals/principles/target and overall design of 6G air interface
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5 Physical Layer structure for 6GR
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.1 Waveforms
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.2 Frame structure
Editor’s Note: Including numerology and frame structure (for all duplex types)
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.3 Channel coding
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.4 Modulations
Editor’s Note: joint channel coding and modulation will be captured in this Clause, if any
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.5 Initial access
Editor’s Note: including synchronization signal and raster, broadcast signals/channel and physical random access channel, as well as initial access procedure, random access procedures, system information and paging
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.6 Physical layer control, data scheduling and HARQ operation
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.7 MIMO operation
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.8 Duplexing
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.9 Spectrum utilization and aggregation
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.10 NTN
Editor’s Note: harmonized 6G Radio design for TN and NTN, including their integration can be captured in another section. Aspects unique to NTN, if any, can be captured in this section.
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
5.11 Other physical layer signals, channels and procedures
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
6 Energy efficiency
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
7 AI/ML [for 6GR]
Editor’s Note: including use case identification with compelling trade-off between e.g., performance, complexity, etc…, as well as AI/ML framework: Extensible AI/ML enablers based on the identified Use Case(s), including LCM procedures, Data collection, and data management
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
8 Sensing
Editor’s Note: including PHY functions and procedures for sensing technology (e.g., waveform. reference signals, measurement feedback, etc), as well as aspects of integration with communication services
8e813a83597c9d6f04fe2bcb430572a3
38.760-1
9 Performance evaluation
Editor’s Note: at least energy efficiency, spectrum efficiency, and coverage compared to 5G NR, as well as sensing performance Annex <A>: Simulation scenarios and assumptions Editor’s Note: LLS and SLS assumptions used for evaluations Annex <B>: Change history Change history Date Meeting TDoc CR Rev Cat Subject/Comment New version 2025-10 RAN1#122bis R1-2507813 TR Skeleton 0.0.1 2025-11 RAN1#123 R1-2509279 TR Skeleton for RAN1 endorsement 0.0.2 2025-11 RAN1#123 R1-2509569 Updated TR Skeleton for RAN1 endorsement 0.0.3
21583c06c9a14c098fd1bd4da409262a
29.530
1 Scope
The present document specifies the stage 3 protocol and data model for the Naf Service Based Interface. It provides stage 3 protocol definitions and message flows, and specifies the API for each service offered by the AF. The 5G System stage 2 architecture and procedures are specified in 3GPP TS 23.288 [14], 3GPP TS 23.501 [2] and 3GPP TS 23.502 [3]. The Technical Realization of the Service Based Architecture and the Principles and Guidelines for Services Definition are specified in 3GPP TS 29.500 [4] and 3GPP TS 29.501 [5].
21583c06c9a14c098fd1bd4da409262a
29.530
2 References
The following documents contain provisions which, through reference in this text, constitute provisions of the present document. - References are either specific (identified by date of publication, edition number, version number, etc.) or non‑specific. - For a specific reference, subsequent revisions do not apply. - For a non-specific reference, the latest version applies. In the case of a reference to a 3GPP document (including a GSM document), a non-specific reference implicitly refers to the latest version of that document in the same Release as the present document. [1] 3GPP TR 21.905: "Vocabulary for 3GPP Specifications". [2] 3GPP TS 23.501: "System Architecture for the 5G System; Stage 2". [3] 3GPP TS 23.502: "Procedures for the 5G System; Stage 2". [4] 3GPP TS 29.500: "5G System; Technical Realization of Service Based Architecture; Stage 3". [5] 3GPP TS 29.501: "5G System; Principles and Guidelines for Services Definition; Stage 3". [6] OpenAPI: "OpenAPI Specification Version 3.0.0", https://spec.openapis.org/oas/v3.0.0. [7] 3GPP TR 21.900: "Technical Specification Group working methods". [8] 3GPP TS 33.501: "Security architecture and procedures for 5G system". [9] IETF RFC 6749: "The OAuth 2.0 Authorization Framework". [10] 3GPP TS 29.510: "5G System; Network Function Repository Services; Stage 3". [11] IETF RFC 9113: "HTTP/2". [12] IETF RFC 8259: "The JavaScript Object Notation (JSON) Data Interchange Format". [13] IETF RFC 9457: "Problem Details for HTTP APIs". [14] 3GPP TS 23.288: "Architecture enhancements for 5G System (5GS) to support network data analytics services". [15] 3GPP TS 29.552: "5G System; Network Data Analytics signalling flows; Stage 3". [16] 3GPP TS 29.571: "5G System; Common Data Types for Service Based Interfaces; Stage 3". [17] 3GPP TS 29.523: "5G System; Policy Control Event Exposure Service; Stage 3". [18] 3GPP TS 29.520: "5G System; Network Data Analytics Services; Stage 3". [19] IETF RFC 9112: "HTTP/1.1". [20] IETF RFC 9110: "HTTP Semantics". [21] IETF RFC 9111: "HTTP Caching". [22] 3GPP TS 29.122: "T8 reference point for Northbound APIs". [23] 3GPP TS 29.554: "5G System; Background Data Transfer Policy Control Service; Stage 3".
21583c06c9a14c098fd1bd4da409262a
29.530
3 Definitions, symbols and abbreviations
21583c06c9a14c098fd1bd4da409262a
29.530
3.1 Definitions
For the purposes of the present document, the terms and definitions given in 3GPP TR 21.905 [1] and the following apply. A term defined in the present document takes precedence over the definition of the same term, if any, in 3GPP TR 21.905 [1]. For the purpose of the present document, the terms and definitions given in clause 3 of 3GPP TS 23.288 [14] also apply, including the ones referencing other specifications.
21583c06c9a14c098fd1bd4da409262a
29.530
3.2 Symbols
21583c06c9a14c098fd1bd4da409262a
29.530
3.3 Abbreviations
For the purposes of the present document, the abbreviations given in 3GPP TR 21.905 [1] and the following apply. An abbreviation defined in the present document takes precedence over the definition of the same abbreviation, if any, in 3GPP TR 21.905 [1]. AF Application Function AI/ML Artificial Intelligence/Machine Learning GPSI Generic Public Subscription Identifier NEF Network Exposure Function NWDAF Network Data Analytics Function REST Representational State Transfer VFL Vertical Federated Learning
21583c06c9a14c098fd1bd4da409262a
29.530
4 Overview
The Application Function Artificial Intelligence/Machine Learning (AI/ML) Services, as defined in 3GPP TS 23.288 [14], are provided by the Application Function (AF). The following AI/ML services are specified for the AF: Table 4.1-1: AI/ML Services provided by AF Service Name Description Service Operations Operation Semantics Example Consumer(s) Naf_VFLTraining This service is provided by an AF acting as VFL client and enables an NF service consumer to request the AF to participate in VFL model training as VFL client and train a local model. Subscribe(NOTE 1) Subscribe / Notify NWDAF, NEF Unsubscribe Notify Naf_VFLInference This service is provided by AF acting as VFL client and enables an NF service consumer to subscribe/unsubscribe for a VFL inference. Subscribe(NOTE 2) Subscribe / Notify NWDAF Unsubscribe Notify Naf_Inference This service is provided by AF acting as VFL server and enables an NF service consumer to subscribe/unsubscribe for a VFL inference. Subscribe(NOTE 3) Subscribe / Notify NWDAF, NEF Unsubscribe Notify Naf_Training This service is provided by AF acting as VFL server and enables an NF service consumer to subscribe/unsubscribe for a VFL training. Subscribe Subscribe / Notify NWDAF, NEF Unsubscribe Notify NOTE 1: This service implements also the Naf_VFLTraining_Request as specified in 3GPP TS 23.288 [14] by using immediate and one-time reporting requirement. NOTE 2: This service implements also the Naf_VFLInference_Request as specified in 3GPP TS 23.288 [14] by using immediate and one-time reporting requirement. NOTE 3: This service implements also the Naf_Inference_Request as specified in 3GPP TS 23.288 [14] by using immediate and one-time reporting requirement.
21583c06c9a14c098fd1bd4da409262a
29.530
5 Services offered by the AF
21583c06c9a14c098fd1bd4da409262a
29.530
5.1 Introduction
The AF offers to other NFs the following services: - Naf_VFLTraining; - Naf_VFLInference; - Naf_Training; - Naf_Inference. Table 5.1-1 summarizes the corresponding APIs defined for this specification. Table 5.1-1: API Descriptions Service Name Clause Description OpenAPI Specification File apiName Annex Naf_VFLTraining 5.2 AF VFL Training service TS29530_Naf_VFLTraining.yaml naf-vfl-train A.2 Naf_VFLInference 5.3 AF VFL Inference service TS29530_Naf_VFLInference.yaml naf-vflinference A.3 Naf_Training 5.4 AF training service TS29530_Naf_Training.yaml naf-train A.4 Naf_Inference 5.5 AF Inference service TS29530_Naf_Inference.yaml naf-inference A.5
21583c06c9a14c098fd1bd4da409262a
29.530
5.2 Naf_VFLTraining Service
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.1 Service Description
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.1.1 Overview
The Naf_VFLTraining service exposed by the AF acting as VFL client and enables an NF service consumer to: - request the creation/update of a VFL Training Subscription; and - receive VFL Training related event(s) reporting.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.1.2 Service Architecture
The 5G System Architecture is defined in 3GPP TS 23.501 [2]. The Network Data Analytics Exposure architecture is defined in 3GPP TS 23.288 [14]. The VFL signalling flows are defined in 3GPP TS 29.552 [15]. The Naf_VFLTraining service is part of the Naf service-based interface exhibited by the trusted Application Function (AF) or untrusted Application Function (AF). Known consumers of the Naf_VFLTraining service are: - Network Data Analytics Function (NWDAF) when the AF is trusted. - Network Exposure Function (NEF) when the AF is untrusted. Figure 5.2.1.2-1: Reference Architecture for the Naf_VFLTraining service; SBI representation Figure 5.2.1.2-2: Reference Architecture for the Naf_VFLTraining service: reference point representation
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.1.3 Network Functions
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.1.3.1 Application Function (AF)
The Application Function (AF) acting as VFL client provides VFL training for different analytics events to NF service consumers. The Application Function (AF) acting as VFL client allows NF service consumers to subscribe to and unsubscribe from VFL training event notifications.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.1.3.2 NF Service Consumers
The Network Data Analytics Function (NWDAF) and Network Exposure Function (NEF) support (un)subscription to the notification of different VFL training events.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2 Service Operations
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.1 Introduction
The service operations defined for the Naf_VFLTraining service are shown in table 5.2.2.1-1. Table 5.2.2.1-1: Naf_VFLTraining Service Operations Service Operation Name Description Initiated by Naf_VFLTraining_Subscribe This service operation enables the NF service consumer to request the creation/update of a VFL Training Subscription. e.g., NWDAF, NEF Naf_VFLTraining_Unsubscribe This service operation enables the NF service consumer to request the deletion of a VFL Training Subscription. e.g., NWDAF, NEF Naf_VFLTraining_Notify This service operation enables the NF service consumer to receive VFL Training related event(s) reporting. AF
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.2 Naf_VFLTraining_Subscribe
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.2.1 General
This service operation is used by an NF service consumer to request the creation/update of a VFL Training Subscription at the AF. The following procedures are supported by the "Naf_VFLTraining_Subscribe" service operation: - VFL Training Subscription Creation. - VFL Training Subscription Update.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.2.2 VFL Training Subscription Creation
Figure 5.2.2.2.2-1 depicts a scenario where an NF service consumer sends a request to the AF to request the creation of a VFL Training Subscription (see also clause 6.2H of 3GPP°TS°23.288°[14]). Figure 5.2.2.2.2-1: Procedure for VFL Training Subscription Creation 1. In order to subscribe to VFL Training, the NF service consumer shall send an HTTP POST request to the AF targeting the URI of the "VFL Training Subscriptions" collection resource, with the request body including the VflTrainingSubs data structure. 2a. Upon success, the AF shall respond with an HTTP "201 Created" status code with the response body containing a representation of the created "Individual VFL Training Subscription" resource within the VflTrainingSubs data structure, and an HTTP "Location" header field containing the URI of the created resource. 2b. On failure, the appropriate HTTP status code indicating the error shall be returned and appropriate additional error information should be returned in the HTTP POST response body, as specified in clause 6.1.7.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.2.3 VFL Training Subscription Update
Figure 5.2.2.2.3-1 depicts a scenario where an NF service consumer sends a request to the AF to request the update of an existing VFL Training Subscription (see also clause 6.2H of 3GPP°TS°23.288°[14]). Figure 5.2.2.2.3-1: Procedure for VFL Training Subscription Update 1. In order to request the update of an existing VFL Training Subscription, the NF service consumer shall send an HTTP PUT/PATCH request to the AF, targeting the URI of the corresponding "Individual VFL Training Subscription" resource, with the request body including either: - the updated representation of the resource within the VflTrainingSubs data structure, in case the HTTP PUT method is used; or - the requested modifications to the resource within the VflTrainingSubsPatch data structure, in case the HTTP PATCH method is used. 2a. Upon success, the AF shall update the targeted "Individual VFL Training Subscription" resource accordingly and respond with either: - an HTTP "200 OK" status code with the response body containing a representation of the updated "Individual VFL Training Subscription" resource within the VflTrainingSubs data structure; or - an HTTP "204 No Content" status code. 2b. On failure, the appropriate HTTP status code indicating the error shall be returned and appropriate additional error information should be returned in the HTTP PUT/PATCH response body, as specified in clause 6.1.7.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.3 Naf_VFLTraining_Unsubscribe
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.3.1 General
This service operation is used by an NF service consumer to request the deletion of a VFL Training Subscription at the AF. The following procedures are supported by the "Naf_VFLTraining_Unsubscribe" service operation: - VFL Training Subscription Deletion.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.3.2 VFL Training Subscription Deletion
Figure 5.2.2.3.2-1 depicts a scenario where an NF service consumer sends a request to the AF to delete an existing VFL Training Subscription (see also clause 6.2H of 3GPP°TS°23.288°[14]). Figure 5.2.2.3.2-1: Procedure for VFL Training Subscription Deletion 1. In order to request the deletion of an existing VFL Training Subscription, the NF service consumer shall send an HTTP DELETE request to the AF targeting the URI of the corresponding "Individual VFL Training Subscription" resource. 2a. Upon success, the AF shall respond with an HTTP "204 No Content" status code. 2b. On failure, the appropriate HTTP status code indicating the error shall be returned and appropriate additional error information should be returned in the HTTP DELETE response body, as specified in clause 6.1.7.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.4 Naf_VFLTraining_Notify
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.4.1 General
This service operation is used by the AF to notify a previously subscribed service consumer on: - VFL Training report(s). The following procedures are supported by the "Naf_VFLTraining_Notify" service operation: - VFL Training Notification.
21583c06c9a14c098fd1bd4da409262a
29.530
5.2.2.4.2 VFL Training Notification
Figure 5.2.2.4.2-1 depicts a scenario where the AF sends a request to notify a previously subscribed service consumer on VFL Training report(s) (see also clause 6.2H of 3GPP°TS°23.288°[14]). Figure 5.2.2.4.2-1: Procedure for VFL Training Notification 1. In order to notify a previously subscribed service consumer on VFL Training report(s), the AF shall send an HTTP POST request to the NF service consumer with the request URI set to "{notifUri}", where the "notifUri" variable is set to the value received from the NF service consumer during the creation/update of the corresponding VFL Training Subscription using the procedures defined in clauses 5.2.2.2, and the request body including the VflTrainingNotify data structure. 2a. Upon success, the NF service consumer shall respond to the AF with an HTTP "204 No Content" status code to acknowledge the reception of the notification. 2b. On failure, the appropriate HTTP status code indicating the error shall be returned and appropriate additional error information should be returned in the HTTP POST response body, as specified in clause 6.1.7.
21583c06c9a14c098fd1bd4da409262a
29.530
5.3 Naf_VFLInference Service
21583c06c9a14c098fd1bd4da409262a
29.530
5.3.1 Service Description
21583c06c9a14c098fd1bd4da409262a
29.530
5.3.1.1 Overview
The Naf_VFLInference service as defined in 3GPP TS 23.288 [14], is provided by the trusted Application Function (AF) or untrusted Application Function (AF) acting as VFL client. This service allows the NF service consumers acting as VFL servers to: - subscribe to and unsubscribe from different VFL inference events; - modify VFL inference subscriptions; and - be notified about events for corresponding VFL inference subscriptions.
21583c06c9a14c098fd1bd4da409262a
29.530
5.3.1.2 Service Architecture
The 5G System Architecture is defined in 3GPP TS 23.501 [2]. The Network Data Analytics Exposure architecture is defined in 3GPP TS 23.288 [14]. The VFL signalling flows are defined in 3GPP TS 29.552 [15]. The Naf_VFLInference service is part of the Naf service-based interface exhibited by the trusted Application Function (AF) or untrusted Application Function (AF). Known consumers of the Naf_VFLInference service are: - Network Data Analytics Function (NWDAF) when the AF is trusted. - Network Exposure Function (NEF) when the AF is untrusted. Figure 5.3.1.2-1: Reference Architecture for the Naf_VFLInference service; SBI representation Figure 5.3.1.2-2: Reference Architecture for the Naf_VFLInference service: reference point representation
21583c06c9a14c098fd1bd4da409262a
29.530
5.3.1.3 Network Functions