INACTIVE eDRX Configuration via Federated Learning in 5G

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing eDRX configurations for user equipment (UEs) in 5G networks are not optimally set, leading to inefficiencies in power consumption and latency tradeoffs, as INACTIVE eDRX configurations are not shared globally, hindering centralized optimization and compromising data privacy.

Innovation Solution

Implement an iterative learning process using a federated learning framework, where a core network node updates a global model with local model parameters from access network nodes to determine optimized INACTIVE eDRX configurations, ensuring data privacy and security by only exchanging model parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If INACTIVE eDRX configurations are not shared globally, then data privacy and security are maintained, but centralized optimization of eDRX configurations becomes impossible

Engineering Contradiction:

Improvedata privacy and securityVSAvoidcentralized optimization capability

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system segments the eDRX configuration optimization into decentralized local optimizations at each gNB while maintaining a centralized coordination mechanism through the AMF. Each gNB independently determines INACTIVE eDRX configurations based on local conditions, preserving data privacy, while the AMF collects statistics and coordinates between different gNBs to achieve global optimization without sharing raw UE data.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The AMF acts as an intermediary between gNBs and the optimization process. It collects statistics from multiple gNBs about eDRX performance, processes this information to identify optimization opportunities, and coordinates configuration changes across different gNBs. This mediator enables centralized optimization insights while maintaining decentralized configuration decisions and data privacy.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Use of energy by moving object

If eDRX cycle length is increased to reduce power consumption, then battery life is extended, but downlink latency increases

Engineering Contradiction:

Improvepower consumptionVSAvoiddownlink latency

Core Design Contradiction:

Use of energy by moving objectVSLoss of time

Solution Approach 1:

The system dynamically adjusts eDRX cycle lengths based on real-time conditions. The AMF collects statistics on UE activity patterns, paging requirements, and network conditions to determine optimal eDRX configurations for each gNB. This dynamic adaptation allows the system to extend eDRX cycles when conditions permit (reducing power consumption) while reducing cycles when latency is critical (improving responsiveness).

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes eDRX configuration parameters (cycle length, paging time window) based on statistical analysis of network conditions and UE behavior. By analyzing data from multiple gNBs, the AMF identifies patterns that indicate when longer eDRX cycles are safe and when shorter cycles are needed, optimizing the tradeoff between power saving and latency for different scenarios.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If INACTIVE eDRX configurations are determined locally at gNB, then device complexity is reduced, but global optimization is lost

Engineering Contradiction:

Improveconfiguration determination complexityVSAvoidglobal optimization capability

Core Design Contradiction:

Device complexityVSProductivity

Solution Approach 1:

The optimization function is segmented between local gNBs and centralized AMF. Each gNB performs simple local determination of eDRX configurations based on immediate conditions, keeping device complexity low. The AMF performs more complex global optimization by collecting statistics from multiple gNBs and identifying cross-gNB optimization opportunities, dividing computational complexity across the network hierarchy.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Each gNB is empowered to independently determine appropriate INACTIVE eDRX configurations for its served UEs based on local conditions and capabilities. This self-service approach reduces the complexity burden on individual nodes while the AMF provides coordination and global optimization insights through statistical analysis and inter-gNB coordination.

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS12501513B2Determination of inactive EDRX configurations for user equipment

Publication Date: 2025.12.16 TELEFONAKTIEBOLAGET LM ERICSSON (PUBL)

AI summary

There is provided techniques for determining INACTIVE eDRX configurations for UEs. The method is performed by a core network node. The method comprises performing an iterative learning process with the access network nodes to determine the INACTIVE eDRX configurations. The iterative learning process comprises transmitting a global parameter vector of a machine learning model to the access network nodes for determining the INACTIVE eDRX configurations of the UEs. The global parameter vector defines coefficients for the machine learning model for the current iteration. The iterative learning process comprises receiving local model parameter vectors with locally updated coefficients for the machine learning model from the access network nodes. The iterative learning comprises updating the global parameter vector for the next iteration as a function of all received local model parameter vectors until a stopping criterion is fulfilled.