|
Work item:
|
Y.FMSC-InNetFL
|
|
Subject/title:
|
Requirements and Framework for In-Network Aggregated Federated Learning to Enable AI in Fixed, Mobile, and Satellite Convergence Networks
|
|
Status:
|
Under study [Issued from previous study period]
|
|
Approval process:
|
AAP
|
|
Type of work item:
|
Recommendation
|
|
Version:
|
New
|
|
Equivalent number:
|
-
|
|
Timing:
|
2025-Q3 (High priority)
|
|
Liaison:
|
3GPP, ITU-R
|
|
Supporting members:
|
Hankuk University of Foreign Studies (HUFS), KT, ETRI, China Mobile
|
|
Summary:
|
This Recommendation specifies the requirements and framework for in-network aggregated federated learning (FL) to enable artificial intelligence (AI) in fixed, mobile, and satellite convergence (FMSC) networks. In-network aggregated FL will enhance the network capability in FMSC to significantly reduce the outgoing data traffic from the lower network entities, consume less bandwidth, and enable more efficient model aggregation over multi-hop networks under limited communication resources.
|
|
Comment:
|
-
|
|
Reference(s):
|
|
|
Historic references:
|
|
Contact(s):
|
|
| ITU-T A.5 justification(s): |
|
|
|
|
First registration in the WP:
2022-12-02 16:33:25
|
|
Last update:
2025-04-02 13:39:37
|
