Federated Learning Handover Across Wireless Access Nodes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Federated learning (FL) processes in wireless communication networks face issues due to wireless devices becoming unavailable during training, leading to interruptions, latencies, and inaccurate models, especially in scenarios with mobile devices.
Innovation Solution
A method for federated learning that allows resuming the FL process through a secondary access node by transmitting FL information between a wireless device and a secondary access node, enabling continued participation even when the primary connection is lost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Use of energy by moving object
If a wireless device goes into idle mode during FL training, then the device can save energy and reduce signaling overhead, but the FL process is interrupted and latency increases
Solution Approach 1:
The system performs preliminary actions by establishing a second wireless access radio link with a second access node before the device actually goes idle. FL information is proactively transmitted to the second access node in advance, so when the device transitions to idle mode and reconnects, the FL process can resume immediately without interruption or latency.
2Reliability
If a wireless device hands over to a different access node, then the device can maintain continuous connectivity, but the FL process is interrupted and data integrity is compromised
Solution Approach 1:
The second access node acts as an intermediary during handover. When the device hands over from the first access node to the second, the second access node receives and stores the FL information locally. This intermediary storage mechanism ensures that FL data integrity is maintained throughout the handover process, allowing seamless resumption of the FL process without data loss.
3Reliability
If FL information is transmitted to multiple access nodes, then the FL process can resume after handover, but signaling overhead and network complexity increase
Solution Approach 1:
The system applies local quality by transmitting FL information selectively to only those access nodes that are relevant for potential handovers. Each access node processes and stores FL information locally based on its specific role in the network, rather than universally distributing to all nodes. This reduces unnecessary signaling overhead and network complexity while maintaining FL process continuity.
Data Source
AI summary
A computer implemented method for federated machine learning, FL, in a wireless communication system, the method comprising establishing a first wireless access radio link between a first access node and a wireless device, initiating an FL process involving the first access node and the wireless device, transmitting FL information from the first access node to the wireless device, where the FL information comprises data indicative of the FL process, establishing a second wireless access radio link between a second access node and the wireless device, where the second access node is communicatively coupled to the first access node, exchanging at least part of the FL information between the wireless device and the second access node, and resuming the FL process involving the first access node and the wireless device by communication via the second access node over the second wireless access radio link.