E2 Node Federated Learning via RIC-Controlled Capability Signaling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is currently no mechanism to initiate a federated learning process within the radio access network (RAN) when the aggregator resides outside the RAN nodes, leading to inefficiencies in resource consumption and data exchange.
Innovation Solution
Implementing a method to enable federated learning operations within the O-RAN architecture by exposing the capabilities of E2 nodes as local clients to a near-real-time RIC, which acts as the central aggregator, and extending the E2 interface with new procedures and messages to support FL operation provisioning and execution.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If federated learning is implemented with aggregator outside RAN nodes, then data privacy is preserved and resource consumption is reduced, but there is no mechanism to initiate the federated learning process within RAN
Solution Approach 1:
The patent introduces an intermediary mechanism in the form of extended E2 interface messages (E2 setup request, E2 setup response, RIC control request) that enable the RAN nodes to indicate their federated learning capabilities and allow the external aggregator to initiate the federated learning process. This intermediary communication framework resolves the contradiction by providing a structured initiation mechanism while maintaining the external aggregator architecture that reduces resource consumption.
2Productivity
If large volumes of data are transferred externally for training, then centralized model training is achieved, but resource consumption and data exchange inefficiencies increase
Solution Approach 1:
The patent extracts the federated learning initiation and coordination functionality from the RAN nodes themselves and places it in the external aggregator system. The RAN nodes only send capability indications and participate in the learning process when initiated, rather than continuously exchanging large volumes of data. This extraction principle enables centralized model training capability while minimizing data transfer and resource consumption by only exchanging necessary signaling messages.
3Adaptability or versatility
If E2 interface is extended with new procedures and messages, then federated learning operation provisioning is supported, but device complexity increases
Solution Approach 1:
The patent segments the federated learning support functionality into distinct, modular message exchanges on the E2 interface. Rather than implementing a complex integrated system, the solution divides the functionality into separate procedures: capability indication in E2 setup request/response messages and initiation commands in RIC control request messages. This segmentation enables federated learning support while keeping each individual message and procedure relatively simple and manageable.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a method comprising receiving, from one or more E2 nodes of a radio access network, an E2 setup request message indicating a capability of the one or more E2 nodes for operating as a federated learning client in a federated learning process; determining to initiate the federated learning process for the one or more E2 nodes that indicated the capability for operating as the federated learning client; and transmitting, to the one or more E2 nodes, an RIC control request message comprising an indication to initiate the federated learning process.