UE Access Performance Prediction Circuit for Traffic Steering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
User equipment (UE) faces challenges in rapidly changing 3GPP and non-3GPP access network environments, leading to delayed response times and degraded end-user experience due to difficulties in properly steering, switching, or splitting traffic across these networks.
Innovation Solution
Incorporating an access performance prediction circuit that uses machine learning to predict the future performance of 3GPP and non-3GPP access networks, allowing the UE to take proactive actions such as traffic steering, switching, or splitting based on predicted availability, round-trip time, and congestion, thereby improving user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the ATSSS capable UE performs access performance measurements to decide traffic distribution, then the traffic steering decision can be made based on actual measurements, but the response time increases and end-user experience is degraded due to rapid network environment changes
Solution Approach 1:
The patent applies preliminary action by performing access performance prediction before actual traffic steering decisions are needed. The UE uses machine learning models to predict future access performance (availability, RTT, throughput) based on current and historical measurements, allowing the system to proactively prepare steering decisions in advance rather than reactively responding to measurements when network conditions have already changed.
2Reliability
If the UE waits for measurement results to make traffic steering decisions, then decisions are based on accurate current performance data, but the rapid changes in 3GPP and non-3GPP access networks cause delayed responses
Solution Approach 1:
The patent implements feedback by continuously monitoring actual access performance measurements and using this feedback to update and refine the machine learning prediction models. The system compares predicted performance with actual measured performance, allowing the models to learn from discrepancies and improve future predictions, thereby maintaining both reliability and speed in traffic steering decisions.
3Adaptability or versatility
If the UE takes action based on current access performance measurements, then the traffic distribution responds to current network conditions, but the rapid changes in network environment result in suboptimal decisions by the time actions are taken
Solution Approach 1:
The patent applies parameter changes by transforming the approach from using current measured performance parameters to using predicted future performance parameters for traffic steering decisions. The machine learning models predict key performance parameters (availability, round-trip time, throughput) for future time intervals, allowing the UE to adapt traffic distribution to anticipated network conditions rather than outdated current conditions.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A user equipment, UE, (100) includes an access performance prediction circuit (102) and a wireless communication circuit (104) . The access performance prediction circuit (102) predicts performance of a 3rd generation partnership project, 3GPP, access and performance of a non-3GPP access. The wireless communication circuit (104) takes action in response to predicted performance of the 3GPP access and predicted performance of the non-3GPP access.