ETSI GS F5G 025 V1.1.1 (2026-09)

• Abstract:

  DGS/F5G-0025

• Status: Not Published
• Technical Committee: F5G - Fifth Generation Fixed Network
• Current Stage: 12 - Citation in the OJ (auto-insert)
• Due Date: 09-Oct-2026
• Completion Date: 15-Sep-2026

GROUP SPECIFICATION
Fifth Generation Fixed Network (F5G);
F5G-A Architecture of Computing Collaboration in
PON Network
Disclaimer
The present document has been produced and approved by the Fifth Generation Fixed Network (F5G) ETSI Industry
Specification Group (ISG) and represents the views of those members who participated in this ISG.
It does not necessarily represent the views of the entire ETSI membership.

2 ETSI GS F5G 025 V1.1.1 (2026-09)

Reference
DGS/F5G-0025
Keywords
F5G, F5G Advanced, network, PON

ETSI
650 Route des Lucioles
F-06921 Sophia Antipolis Cedex - FRANCE

Tel.: [phone]  Fax: [phone]

Siret N° 348 623 562 00017 - APE 7112B
Association à but non lucratif enregistrée à la
Sous-Préfecture de Grasse (06) N° w061004871

Important notice
The present document can be downloaded from the
ETSI Search & Browse Standards application.
The present document may be made available in electronic versions and/or in print. The content of any electronic and/or
print versions of the present document shall not be modified without the prior written authorization of ETSI. In case of any
existing or perceived difference in contents between such versions and/or in print, the prevailing version of an ETSI
deliverable is the one made publicly available in PDF format on ETSI deliver repository.
Users should be aware that the present document may be revised or have its status changed,
this information is available in the Milestones listing.
If you find errors in the present document, please send your comments to
the relevant service listed under Committee Support Staff.
If you find a security vulnerability in the present document, please report it through our
Coordinated Vulnerability Disclosure (CVD) program.
Notice of disclaimer & limitation of liability
The information provided in the present deliverable is directed solely to professionals who have the appropriate degree of
experience to understand and interpret its content in accordance with generally accepted engineering or
other professional standard and applicable regulations.
No recommendation as to products and services or vendors is made or should be implied.
No representation or warranty is made that this deliverable is technically accurate or sufficient or conforms to any law
and/or governmental rule and/or regulation and further, no representation or warranty is made of merchantability or fitness
for any particular purpose or against infringement of intellectual property rights.
In no event shall ETSI be held liable for loss of profits or any other incidental or consequential damages.

Any software contained in this deliverable is provided "AS IS" with no warranties, express or implied, including but not
limited to, the warranties of merchantability, fitness for a particular purpose and non-infringement of intellectual property
rights and ETSI shall not be held liable in any event for any damages whatsoever (including, without limitation, damages
for loss of profits, business interruption, loss of information, or any other pecuniary loss) arising out of or related to the use
of or inability to use the software.
Copyright Notification
No part of this document may be reproduced in any form, by any means and in any media, without the prior written
authorization of ETSI and except as expressly permitted below.
By way of exception and when the document is a normative deliverable (European Standard (EN),
Technical Specification (TS), Group Specification (GS) or ETSI Standard (ES)), ETSI authorizes to reproduce
and incorporate into products, services and technical documentation only those extracts (e.g. templates) that are strictly
necessary for the technical implementation of the normative deliverable, to ensure compliance with the latter.
Nothing in this notice shall be construed as limiting any mandatory exceptions to copyright provided by applicable law.

© ETSI 2026.
All rights reserved.
ETSI
3 ETSI GS F5G 025 V1.1.1 (2026-09)
Contents
Intellectual Property Rights . 4
Foreword . 4
Modal verbs terminology . 4
1 Scope . 5
2 References . 5
2.1 Normative references . 5
2.2 Informative references . 5
3 Definition of terms, symbols and abbreviations . 6
3.1 Terms . 6
3.2 Symbols . 6
3.3 Abbreviations . 6
4 Overview of Computing Collaboration Architecture . 7
4.1 Relationship with F5G-A Architecture. 7
4.2 Architecture of computing collaboration . 8
4.2.1 Architecture framework of computing collaboration . 8
4.2.2 General procedure of computing coordination . 11
5 Scenarios and Computation Capabilities . 12
5.1 F5G Use Cases relevant to computing collaboration. 12
5.2 F5G-A Use Cases relevant to computing collaboration . 13
5.3 Additional use case: demonstration of applying computing-networking collaboration in Cloud-VR . 14
6 Collaboration Requirements . 15
6.1 Collaboration Scenarios . 15
6.2 General Technical Requirements for Computing-Networking Collaboration . 15
7 Components Enhancement . 18
7.1 General . 18
7.2 Enhancement on Heterogeneous computing capability . 19
7.3 Enhancement on computing functionality (including perception capability) . 20
7.4 Enhancement on dynamic computing collaboration in PON network . 21
7.4.1 Overview . 21
7.4.2 Generic interface framework for computing collaboration . 21
7.4.3 Enhancement on the access collaboration interface . 23
7.4.4 Enhancement on the in-premises collaboration interfaces . 23
7.4.5 Enhancement on the remote management and northbound interfaces . 24
7.4.6 Generic mechanism, message format and data model requirements . 25
8 Conclusions . 25
Annex A (informative): Change history . 26
History . 27

ETSI
4 ETSI GS F5G 025 V1.1.1 (2026-09)
Intellectual Property Rights
Essential patents
IPRs essential or potentially essential to normative deliverables (European Standard (EN), Technical Specification (TS),
Group Specification (GS) or ETSI Standard (ES)) may have been declared to ETSI. The declarations pertaining to these
essential IPRs, if any, are publicly available for ETSI members and non-members, and can be found in
ETSI SR 000 314: "Intellectual Property Rights (IPRs); Essential, or potentially Essential, IPRs notified to ETSI in
respect of ETSI standards", which is available from the ETSI Secretariat. Latest updates are available on the
ETSI IPR online database.
Pursuant to the ETSI Directives including the ETSI IPR Policy, no investigation regarding the essentiality of IPRs,
including IPR searches, has been carried out by ETSI. No guarantee can be given as to the existence of other IPRs not
referenced in ETSI SR 000 314 (or the updates on the ETSI Web server) which are, or may be, or may become,
essential to the present document.
Trademarks
The present document may include trademarks and/or tradenames which are asserted and/or registered by their owners.
ETSI claims no ownership of these except for any which are indicated as being the property of ETSI, and conveys no
right to use or reproduce any trademark and/or tradename. Mention of those trademarks in the present document does
not constitute an endorsement by ETSI of products, services or organizations associated with those trademarks.
DECT™, PLUGTESTS™, UMTS™ and the ETSI logo are trademarks of ETSI registered for the benefit of its
Members. 3GPP™, LTE™ and 5G™ logo are trademarks of ETSI registered for the benefit of its Members and of the
3GPP Organizational Partners. oneM2M™ logo is a trademark of ETSI registered for the benefit of its Members and of ®
the oneM2M Partners. GSM and the GSM logo are trademarks registered and owned by the GSM Association.
Foreword
This Group Specification (GS) has been produced by ETSI Industry Specification Group (ISG) Fifth Generation Fixed
Network (F5G).
Modal verbs terminology
In the present document "shall", "shall not", "should", "should not", "may", "need not", "will", "will not", "can" and
"cannot" are to be interpreted as described in clause 3.2 of the ETSI Drafting Rules (Verbal forms for the expression of
provisions).
"must" and "must not" are NOT allowed in ETSI deliverables except when used in direct citation.

ETSI
5 ETSI GS F5G 025 V1.1.1 (2026-09)
1 Scope
The present document specifies the architecture framework and requirements of computing collaboration in PON
optical access networks and key computing functions and capabilities of PON network components.
2 References
2.1 Normative references
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
referenced document (including any amendments) applies.
Referenced documents which are not found to be publicly available in the expected location might be found in the
ETSI docbox.
NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee
their long-term validity.
The following referenced documents are necessary for the application of the present document.
[1] ETSI GS F5G 036 (V1.1.1) (2026-04): "Fifth Generation Fixed Network (F5G); F5G Advanced
Network Architecture Release 4".
2.2 Informative references
References are either specific (identified by date of publication and/or edition number or version number) or
non-specific. For specific references, only the cited version applies. For non-specific references, the latest version of the
referenced document (including any amendments) applies.
NOTE: While any hyperlinks included in this clause were valid at the time of publication, ETSI cannot guarantee
their long-term validity.
The following referenced documents may be useful in implementing an ETSI deliverable or add to the reader's
understanding, but are not required for conformance to the present document.
[i.1] ETSI GR F5G 008 (V1.1.1) (2024-06): "Fifth Generation Fixed Network (F5G); F5G Use Cases
Release #2".
[i.2] ETSI GR F5G 020 (V1.1.1) (2024-10): "Fifth Generation Fixed Network (F5G); F5G Advanced
Use Cases; Release 3".
[i.3] Recommendation ITU-T G.988 (2022) including Amendments: "ONU management and control
interface (OMCI) specification".
[i.4] Recommendation ITU-T G.9943 (10/2025): "High speed fibre-based in-premises transceivers
network management".
[i.5] Recommendation ITU-T G.9949 (10/2025): "Wireless local area network management control
interface (WMCI) for in-premises network".
[i.6] BBF TR-069: "CPE WAN Management Protocol (CWMP)".
[i.7] BBF TR-181: "Device Data Model for TR-069".
[i.8] BBF TR-369: "User Services Platform (USP)".
[i.9] BBF TR-383: "Common YANG Modules for Access Networks".
[i.10] BBF TR-385: "ITU-T PON YANG Modules".
ETSI
6 ETSI GS F5G 025 V1.1.1 (2026-09)
3 Definition of terms, symbols and abbreviations
3.1 Terms
Void.
3.2 Symbols
Void.
3.3 Abbreviations
For the purposes of the present document, the following abbreviations apply:
3D Three Dimensions
th
6G 6 generation mobile networks
AI Artificial Intelligence
AP Access Point
CDN Content Delivery Network
CO Central Office
CPN Customer Premises Network
CPU Central Processing Unit
CWMP CPE WAN Management Protocol
DBA Dynamic Bandwidth Allocation
E2E End-to-End
E-ONU Edge ONU
FMCI Fibre Management Control Interface
FTTR Fibre-To-The-Room
GEM GPON Encapsulation Method
GPU Graphics Processing Unit
ICT Information and Communications Technology
IoT Internet of Things
IP Internet Protocol
KPI Key Performance Indicator
M&C Management and Control
MCA Management, Control & Analytics
MFU Main Fibre Unit
NOTE: The MFU was called Primary ONU (P-ONU) in previous releases of the F5G and F5G-A architecture.
mTLS mutual Transport Layer Security
NaaS Network as a Service
NE Network Element
NMP Network Management Platform
NMS Network Monitoring System
NPU Neural Processing Unit
O&M Operations and Maintenance
ODN Optical Distribution Network
OLT Optical Line Termination
OMCI Optical Network Terminal Management & Control Interface
ONT Optical Network Terminal
NOTE: An ONT is many times also referred to in the industry as an Optical Network Unit (ONU).
ONU Optical Network Unit
OTN Optical Transport Network
PON Passive Optical Network
P-ONU Primary ONU
QoD Quality on Demand
ETSI
7 ETSI GS F5G 025 V1.1.1 (2026-09)
QoE Quality of Experience
QoS Quality of Service
RBAC Role-Based Access Control
SAP Service Access Point
SFU Sub Fibre Unit
NOTE: The SFU used to be called Edge ONU (E-ONU) in previous release of the F5G and F5G-A architecture.
SLA Service Level Agreement
SMP Service Mapping Point
SNR Signal-to-Noise Ratio
SPP Service Processing Point
TLS Transport Layer Security
VLAN Virtual Local Area Network
VoIP Voice over Internet Protocol
VR Virtual Reality
WAN Wide Area Network
WLAN Wireless Local Area Network ®
WMCI Wi-Fi Management & Control Interface
XR eXtended Reality
YANG Yet Another Next Generation
4 Overview of Computing Collaboration Architecture
4.1 Relationship with F5G-A Architecture
The F5G-A network architecture consists of three basic planes from F5G, an Underlay Plane, a Service Plane and a
Management, Control and Analytics (F5G-A MCA) Plane. A cross-plane computing is specified as main characteristic
in F5G-A network architecture. The compute-oriented resources are located on any of the network nodes or are
co-located with the network nodes.
Figure 1: F5G-Advanced network architecture
As defined in the F5G-A network architecture [1], computing collaboration in the PON network requires the following
functionalities:
• Compute resources are distributed throughout the network running on network elements or dedicated compute
hardware.
• Computing collaborating with the network, enables the use of more intelligent algorithms including AI-based
functions to steer and control the network. The network requires coordinated configuration of computing,
storage, and networking. Computing and storage shall be addressed from the physical hardware (general
purpose, AI-optimized and special purpose), cloud computing platform, and software packages.
ETSI
8 ETSI GS F5G 025 V1.1.1 (2026-09)
• Support of a Northbound Interface to configure Network as a Service (NaaS) and to enable the establishment
of connections with a certain Quality on Demand (QoD). This supports the real-time dynamic scheduling of
computing power.
• Enabling operators to control and manage the Customer Premise Network (CPN) requires the capability to
manage the services on the CPN segment. Computing resources in CPN and computing resource requirement
of users are mastered by the service provider or network operator.
In the F5G-A network, the different planes should support the following features:
1) F5G-A Underlay Plane in the access and CPN network segment: the physical network plane, including all
physical network nodes and devices. In PON/FTTR networks, F5G-A Underlay Plane includes OLT, ONU,
MFU and SFU network nodes, which, shall support:
- Computing capability.
- Guaranteed communication capability.
- Storage capacity.
- Analytical capability.
- Data collecting capacity.
- An software operating system with standardized interfaces, where applications are installed.
2) F5G-A Service Plane: An Access Network Node typically contains SAP, SPP and SMP. It identifies services,
adds or removes encapsulations, and directs the computing-related traffic to the proper underlay fabric,
channels, and compute resources.
3) F5G-A Management, Control and Analytics (MCA) Plane: The MCA Plane serves as the intelligent core of
the network, performing management, control and analytics of the complete F5G-A network. It is comprised
of four logical components: Digital Twin, Autonomous Management and Control, AI Analyser and Slicing.
For the PON/FTTR computation collaboration architecture, this plane should support:
- Strong computing capability.
- Strong storage capacity.
- Strong analytical capability based on AI and Digital Twin.
- Computing resource managements.
4.2 Architecture of computing collaboration
4.2.1 Architecture framework of computing collaboration
As shown in Figure 2, the main characteristic of the computing collaboration architecture is the distributed computing
power across different devices in the optical access and CPN network, including the Network Management Platform
(NMP), OLT, ONT(MFU) and SFU. For the collaboration two options are foreseen. First option is extending the
existing management and control protocols with computing-collaboration-specific data models, so that the procedures
defined in clause 4.2.2 are realized. Second option is establishing an IP-based computing collaboration data channel for
high-frequency telemetry data exchange and AI model distribution, when the capacity or the latency of the management
channel is not sufficient, the IP tunnel in Figure 2 is visualizing this option.
ETSI
9 ETSI GS F5G 025 V1.1.1 (2026-09)

Figure 2: Architecture framework of computing collaboration
The main components achieving computing collaboration include:
1) Computing collaboration module: enables computing collaboration message exchange and processing.
2) Application module: processes and analyses data and makes collaboration decisions.
3) Data management: manages the data lifecycle and enables data subscription.
4) Configuration module: based on the decision, it generates configuration data for the devices in the optical
access network.
The overall architecture of the convergence technology between the access network and the broadband customer
network computing network is shown in Figure 3. According to functional divisions, it is composed of the "Forwarding
Layer", "Computing Layer", "Management and Control Layer", and "Orchestration Layer".

Figure 3: Schematic diagram of the overall architecture of the convergence technology between
the access network and the broadband customer network computing network
ETSI
10 ETSI GS F5G 025 V1.1.1 (2026-09)
The Orchestration Layer orchestrates functions of the management and control system and the cloud management
system.
EXAMPLE: Collaboratively invoking intra-network computing power and off-network computing power,
end-to-end collaboration between an access network controller and another network controller, and
collaboration between intra-network edge enhanced computing power and access network
connections are implemented.
The Management and Control Layer includes the cloud management platform, the access network management
system, and the home network management system. The access network management system manages, configures, and
maintains the access network NEs, service pipes, and computing power through the controller's southbound control
interface and a standard information model. It also perceives, calculates, and analyses the status of the connected
service through data interfaces. The access network management system accepts the unified scheduling of the
orchestration and collaboration system through the northbound interface. The home network management system
manages the FTTR MFU and SFU devices, manages, allocates, and maintains the FTTR devices, and coordinates and
schedules the computing power between the internal computing power modules.
The Computing Layer includes the computing power within the access network (both, inside and outside the network),
as well as the computing power at the device-side. The computing power within the network includes the local
computing power at the OLT device and the device-side computing power at the home network element. The
computing power outside the network includes the cloud computing power at edge clouds, central clouds, and public
clouds. The Computing Layer provides perception, measurement, scheduling, and management of computing power.
The computing power within the access network is opened to the users accessing the edge computing power as a service
or product. Or it serves as edge computing centre for home or an industry usage. It can be used for running home
network applications and vertical industry applications, enabling real-time, secure, and reliable lightweight computing.
The Forwarding Layer includes access network elements, ODN, and service pipelines. The access network elements
include OLT devices, and home network elements (ONT/FTTR devices). The service pipelines include OLT user-side
application-level service pipelines, home network user-side application-level service pipelines, and OLT network-side
service pipelines with different SLA levels. The Forwarding Layer provides access, forwarding, and aggregation of user
terminal service data, and provides application-level SLA deterministic assurance for different application data.
The entire access network and broadband customer network need to integrate multiple ICT capabilities, such as sensing,
computing, and storage, to implement computing-network collaboration. The specific list of capabilities is as follows:
• Awareness: real-time awareness of user experience is supported for application-level service flows. Supports
the awareness of the device status based on the device environment.
• Calculation: For service flows, the system collects and analyses real-time user data and evaluates user
experience based on user usage. Accurately allocate network resources based on resource usage. The training
and use of AI/ML models should be supported.
• Connection: application-level slice connections should be supported. Real-time connection parameters should
be adjusted.
• Storage: it should support the storage of perception results and real-time data. It should also support the
storage of data related to computation results. It should further support mechanisms for data synchronization
and reporting, to ensure real-time closed-loop control and management functionality.
• Control: it should support the ability to identify specific events, such as poor-service quality identification. It
should also support dynamic adjustments to service flows and networks based on real-time computation
results. Additionally, it should support the allocation and control of resources such as computation and storage.
At the same time, the service processing capability of the entire network is achieved through independent closed-loop
and cross-layer coordination at various levels, including the network management level, the central office equipment,
and the broadband customer network.
ETSI
11 ETSI GS F5G 025 V1.1.1 (2026-09)
4.2.2 General procedure of computing coordination
To implement a computing coordination procedure, different devices in the optical access network have dedicated roles.
Figure 4 shows a general procedure of computing coordination:
1) Step 1: Enable function for dedicated ONT(MFU). Computing collaboration is a network service offered to
end users by the service provider. The enabling of the function is triggered by the NMP and configured by the
service provider. The targeted ONT(MFU) shall be informed via the OLT.
2) Step 2: The OLT informs the corresponding ONT(MFU) to enable the computing collaboration function
triggering the negotiation of functional capabilities.
3) Step 3: The corresponding ONT(MFU) exchanges messages with the OLT via the management channel using
OMCI to discover computing collaboration capabilities, negotiate the profile and create a data channel for
computing collaboration.
4) Step 4: Based on collaboration function, the OLT subscribes the necessary data from ONT(MFU) for future
decision making.
5) Step 5: the ONT(MFU) collects data from the on-premises network, performs necessary data filtering and
characteristic extraction and sends the processed data to the OLT. Based on the received data, the OLT makes
decisions and provides the strategy for network configuration
6) Step 6: The OLT forwards the decision to ONT(MFU) for dedicated services.
EXAMPLE: Decisions include new network configurations and examples for dedicated services are network
optimization services.
7) Step 7: The OLT performs checks the outcome and provides a report to the NMP periodically.

Figure 4: General procedure of computing coordination
NOTE: The management of computing collaboration for SFU is similar to that of MFU. The configuration of
SFU is left for further study. The establishment of data channel for computing collaboration could be
between SFU and MFU or between SFU and OLT.
ETSI
12 ETSI GS F5G 025 V1.1.1 (2026-09)
5 Scenarios and Computation Capabilities
5.1 F5G Use Cases relevant to computing collaboration
Computing collaboration enhances QoE for digital services and enables more efficient network management. This
capability applies to a number of F5G use cases [i.1]:
• F5G R1 use case #1: Cloud Virtual Reality
This use case describes the Cloud VR scenario, offloading computing and cloud rendering in VR services from local
dedicated hardware to a shared cloud infrastructure. Cloud rendered video and audio outputs are encoded, compressed,
and transmitted to user terminals through fast and stable networks.
In this use case, computing collaboration identifies VR service priority and optimizes network configuration to provide
sufficient network capability for fulfilling service requirements in terms of throughput, latency and jitter.
• F5G R1 use case #10: Scenario based broadband
This use case describes the need for the network to adapt to service requirements from various coexisting frequently
used Internet applications, such as education, gaming, media streaming, and ordinary best-effort Internet (browsing, and
Email).
In this use case, computing collaboration identifies service types based on application characteristic. Network resources
are then allocated to support corresponding network demands with adjustments made based on flexible changes of the
broadband application in the Internet.
• F5G R1 use case #11: Enhanced traffic monitoring and network control in intelligent access network
This use case describes how dynamic data collection and analysis improves traffic monitoring capabilities in the optical
access network enabling further management decisions for optical components.
In this use case, computing collaboration analyses traffic characteristics to fine-tune network parameters including both
PON and FTTR networks.
• F5G R1 use case #14: Digitalized ODN/FTTX
This use case describes a method to digitalize ODN infrastructure through ODN labels and provides intelligent
identification, achieving 100 % accurate resource recording and digital management enabling automatic service
provision.
In this use case, computing collaboration provides visualization and management of fibre infrastructure, ensuring
accurate identification and recoding of ODN resources thereby facilitating rapid ODN deployment and simplified
maintenance.
• F5G R2 use case #15: XR-based virtual presence
This use case describes virtual presence based on XR, where users consume immersive content while producing
immersive content for other online end users.
In this use case, the computing collaboration addresses the challenges of XR-based remote collaboration by offering a
true feeling of being together achieved through provision of sufficient real-time bitrate and deterministic latency.
• F5G R2 use case #18: Virtual music
This use case describes sharing and creating music in real time by different parties that are not in the same location. The
applications include virtual orchestra, virtual music classes, virtual concerts, etc. Currently, this is hardly possible. This
require strict synchronization and latency requirements.
In this use case, the computing collaboration is used to enable a E2E deterministic latency.
ETSI
13 ETSI GS F5G 025 V1.1.1 (2026-09)
• F5G R2 use case #30: Bandwidth on demand
This use case describes adaptive network planning based on the static throughput characteristic for the majority of the
digital services, such as online gaming, periodic data backups, premium video streaming service, newspaper download,
pre-recorded video download, etc.
In this use case, the computing collaboration helps service providers optimize available resources, develop new business
propositions for end-users and improve the QoE for end-users.
• F5G R2 use case #32: AI-based PON optical path diagnosis
This use case describes identification of optical path fault information in the optical access network through data
collection training, and analysis via an AI platform. The AI platform learns the fault characteristics and accurately
models the ODN data with PON optical path data.
In this use case, the computing collaboration identifies the root cause of a PON optical path faults and enables rapid
fault location. Optical path performance degradation is monitored and identified before service interruption actually
occurs. The associated work order system automatically sends orders in advance for optical path fault repair preventing
customer service interruption and significantly improving repair efficiency
5.2 F5G-A Use Cases relevant to computing collaboration
The F5G-A Use case #3 in ETSI GR F5G 020 [i.2] explores the collaboration of computing power in the PON access
network. The computing power, located in ONT, OLT or cloud platform, is used for different purposes, but primarily to
improve the service QoE. The computing power of ONT, OLT and cloud platform at each level are as follows:
• ONT: evaluates and analyses the status of the on-premises network based on its computing capabilities.
• OLT: The computing power within the OLT derives a deeper and more accurate view of on-premises network
status based on ONT feedback, and assist the OLT in performing certain operations such as service
identification, DBA, QoS, etc.
• Cloud platform: performs the deepest analysis of service quality, analysing the service over a long period of
time, providing more accurate conclusions, and request other systems/platforms to collaboratively execute the
necessary operations.
Furthermore, there are several additional related F5G-A use cases in ETSI GR F5G 020 [i.2] involving the computing
power in the PON access network: ®
• F5G-A R3 use case #2: Stable & reliable Wi-Fi connection over FTTR
This use case describes several scenarios (Gigabit/multi-Gigabit coverage, latency sensitive services, seamless hand- ®
over for service continuity, and robust control for IoT services) requiring stable and reliable Wi-Fi connections over
FTTR.
® ®
In this use case, the computing power identifies Wi-Fi problems and improve Wi-Fi quality, e.g. service type ®
identification and adaptation, and Wi-Fi interference recognition and avoidance. In addition, the computing power
facilitating seamless handover between different APs within a single FTTR network or across multiple FTTR networks.
• F5G-A R3 use case #9: Evaluation and assurance of user service experience
This use case describes proactive evaluation and assurance of user experience with fast service experience problem
identification and rectification in the PON access network.
In this use case, computing power in ONT, OLT and QoE analysis system (which may reside in the cloud platform)
executes algorithms for user experience evaluation, QoE problem root cause analysis and user experience rectification.
• F5G-A R3 use case #11: Dynamically digitalized ODN
This use case describes dynamically digitalized ODN to perform proactive and periodical ODN inspection.
With computing power in OLT (called Optical Artificial Intelligent) and the cloud platform, various physical
parameters collected from passive intelligent optical components (e.g. splitters) serve as input for AI algorithms to
perform fault prediction, fault detection and fault recovery in the ODN.
ETSI
14 ETSI GS F5G 025 V1.1.1 (2026-09)
• F5G-A R3 use case #12: On-premises Millimetre Wave (mmWave) WLAN
This use case introduces the mmWave WLAN for high throughput (e.g. 10 Gbps) and high QoE light-of-sight
® ®
connections, avoiding sub-6G Wi-Fi interference. Sub-6G Wi-Fi may still be used to assist beamforming and tracking
of mmWave signals.
In this use case, the computing power executes algorithms for mmWave beamforming and tracking.
• F5G-A R3 use case #17: QoD App-Flow service provisioning
This use case introduces dynamic App-Flow service provisioning for the home broadband users, to improve the user
experience.
In this use case, the computing power at both the NE-level (e.g. ONT and OLT) and the controller-level (which may run
in the cloud platform) are aware of the user service intention (e.g. SLA), perform App-Flow prediction and service
package development, and App-Flow SLA analysis.
In addition, F5G-A introduces new cloud services that requires high bandwidth and very strict latency, including two
use cases in ETSI GR F5G 020 [i.2] that are listed below:
• F5G-A R3 use case #6: Naked-eye 3D display
• F5G-A R3 use case #10: Cloud Desktop
To improve network performance for these services, one possible solution is to move the cloud application content and
the related servers (e.g. rendering servers) closer to users. This approach requires collaboration of computing power in
the PON access network.
EXAMPLE: Edge Content Delivery Network (CDN) storage and Graphics Processing Unit (GPU) computing
power for the cloud applications is installed on the OLT.
5.3 Additional use case: demonstration of applying
computing-networking collaboration in Cloud-VR
Figure 5 shows an example demonstrating the computing-networking collaboration functions in a Cloud-VR service. In
this case, the ONT discovers the on-boarding of a VR device. The peer OLT establishes the computing-networking
collaboration channels for service characteristic identification and make decisions for network optimization. The OLT
maintains the lifecycle of this computing-networking collaboration channel (including channel setup and deletion). The
OLT simultaneously investigates the data demand of the VR service. Furthermore, this is aligned and collaborated with
the transport network, like OTN to reserve a rigid data transmission channel from the rendering cloud.

Figure 5: An example of applying computing collaboration in Cloud-VR scenario
ETSI
15 ETSI GS F5G 025 V1.1.1 (2026-09)
6 Collaboration Requirements
6.1 Collaboration Scenarios
The collaboration scenarios include:
1) Collaboration between on-premises devices MFU and SFU.
2) Collaboration between PON devices OLT and ONTs/MFUs.
3) End-to-End collaboration across the optical access network between the PON network and the on-premises
network.
6.2 General Technical Requirements for Computing-Networking
Collaboration
The main characteristics of computing-networking collaboration are reflected in the use of agile, reliable, intelligent,
and secure network facilities to connect the centre and edge computing facilities. This enables unified orchestration and
control of multi-level computing resources, allowing computing power to be rapidly, securely, and intelligently
transmitted over the network. This, in turn, enables computing power to assist the network in achieving dynamic,
on-demand, and intelligent connections.
Specifically, it includes the following 4 dimensions of collaboration:
1) Computing power collaboration: computing power collaboration refers to the distribution of computing
power resources on the access network and broadband home network based on the perception of user
requirements and the distribution of computing power resources on the access network and broadband home
network.
Hierarchical scheduling and orchestration, select proper paths based on the orchestration result and routing
policy, and schedule computing power requirements to corresponding nodes.
2) Data collaboration: data collaboration refers to the matching between data and computing power through the
network, thereby forming collaboration between data distribution and computing.
Data collaboration includes data collection, storage, processing, and analysis, and effective matching between
data locations and computing power. In this way, effective collaboration between central offices and terminals,
and between systems and devices at the management and control layer is realized.
3) Intelligent collaboration: intelligent collaboration schedules services to appropriate nodes based on the
collaborative perception of services, computing resources, and network resources, implementing collaborative
invoking of intelligent capabilities.
4) Slice collaboration: slice collaboration implements slice service provisioning and control by sensing,
measuring, calculating, and controlling the transmission quality of network slices in real time, improving the
assurance of slice transmission quality.
ETSI
16 ETSI GS F5G 025 V1.1.1 (2026-09)
Table 1: Functional Requirements for Compute-Network Convergence by Collaboration Dimension
Dimension of Capability Description Orchestration M&C Layer Device
Collaboration Layer Layer (CO &
Terminal)
Compute Measurement Unified modelling of multi-source Mandatory Mandatory Mandatory
heterogeneous compute (CPU,
NPU, GPU). Unit: TOPS (int8),
FLOPS (FP32).
Awareness Real-time discovery and reporting Mandatory Mandatory Mandatory
of available compute resources (aggregate) (collect from (report local
and their status. devices) status)
Scheduling Assigning application compute Mandatory N/A N/A
requests (e.g. inference tasks) to
the optimal node based on policy
(latency, load, proximity).
Management Lifecycle management Mandatory Mandatory Mandatory
(provisioning, monitoring, (orchestration) (ops & (local
decommissioning) of compute maintenance) execution)
resources.
Data Collection Gathering operational data (state, N/A Mandatory Mandatory
attributes, KPI metrics) from (from (push to
devices and network paths. devices) M&C)
Storage Persistent storage of telemetry, N/A Mandatory Optional
logs, and user-data cache (e.g. (aggregated (local cache
video). Supports configurable data data) & logs)
retention (e.g. ≥ 30 days).
Processing Hierarchical data handling. Edge N/A Mandatory Optional
devices perform real-time filtering; (central (edge
M&C layer performs correlation analysis) filtering)
and analytics.
Transmission Secure, lossless transport of critical N/A Mandatory Mandatory
operational data. TLS 1.3 (secure (secure
encryption is mandatory. Zero- protocols) protocols)
packet-loss is targeted for critical
flows.
Intelligence Hierarchical
Automated, localized response to Mandatory Mandatory Mandatory
Closed-loop events at each level without central (global policy) (cross-device (local
intervention (e.g. ONT adjusts policy) automation) ®
Wi-Fi ; OLT adjusts DBA).
Cross-domain End-to-end coordination across Mandatory (via Mandatory Mandatory
Collaboration Broadband Customer Network and northbound) (cross- (execute
Access Network for key domain commands)
user/services. control)
Slice Lifecycle On-demand creation, modification, Mandatory Mandatory Mandatory
Management and deletion of application-level (orchestrate (provision per (instantiate/a
slices (e.g. for gaming, VR). E2E) domain) pply)
SLA
Real-time monitoring and Mandatory Mandatory Mandatory
Assurance deterministic guarantee of slice (policy (KPI (measure
SLAs (latency, jitter, loss). monitoring) collection) and report)
Closed-loop Automated "discovery -> Mandatory Mandatory
Mandatory
O&M provisioning -> intelligent O&M -> (policy engine) (analytics & (execution &
performance demonstration" for root cause) reporting)
slice services.
The collaboration scenarios include the collaboration between on-premises devices (i.e. MFU and SFU), collaboration
between PON devices (i.e. OLT and ONT) and collaboration across the E2E optical access network between PON
network and o
...