User Equipment Data Transport Node Selection via Performance Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users experience varying network performance when connected to different Data Transport Nodes (DTNs) due to random user distributions and changing network demands, leading to inconsistent performance for performance-sensitive applications like high-definition video streaming and gaming.
Innovation Solution
User Equipment (UE) selects the most suitable DTN based on performance predictions by determining if a data session is performance-sensitive and then choosing a DTN that can provide the best or adequate performance for that session, using methods such as monitoring data sessions, generating lists of available DTNs, and selecting the optimal one for uplink or downlink performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If a UE connects to a DTN based on traditional signal strength or load balancing, then network coverage is maintained, but performance-sensitive applications experience inconsistent throughput and latency
Solution Approach 1:
The patent changes the selection parameter from traditional signal strength or load metrics to predicted throughput and latency metrics. The UE monitors multiple DTNs and selects based on predicted performance parameters that directly correlate with application experience, thereby improving both throughput and performance consistency for sensitive applications.
Solution Approach 2:
The patent performs preliminary performance prediction and monitoring before actual data transmission. The UE proactively evaluates multiple DTNs using historical data and current network conditions to predict future performance, allowing selection of the optimal DTN before performance-sensitive applications begin, thus ensuring consistent performance.
2Productivity
If a UE monitors and predicts performance of multiple DTNs before selection, then application performance is optimized, but device complexity and processing overhead increase
Solution Approach 1:
The patent implements self-service by having the UE autonomously monitor, predict, and select DTNs based on its own observations and predictions. The device uses its own historical data and current measurements to make intelligent selections without requiring complex network-side coordination, thereby optimizing performance while managing complexity at the device level.
Solution Approach 2:
The patent applies partial monitoring by focusing only on the most relevant performance metrics (throughput and latency predictions) rather than comprehensive network parameter analysis. The UE monitors a subset of key parameters sufficient for performance-sensitive applications, achieving good performance optimization without the overhead of exhaustive monitoring of all network parameters.
3Adaptability or versatility
If network selection is based on ECN data or load information, then congestion avoidance is improved, but measurement precision and accuracy of performance prediction are reduced
Solution Approach 1:
The patent uses feedback from actual data transmission performance to refine predictions. The UE monitors real-world throughput and latency experienced on selected DTNs and uses this feedback to improve future predictions. This continuous learning approach enhances both congestion avoidance capability and prediction accuracy over time, as the system adapts to actual network conditions rather than relying solely on theoretical metrics.
Data Source
Figure 1~2
Figure 3~4
Figure 5~7
AI summary
A wireless device determines whether a data session is performance sensitive, and in response to the data session being performance sensitive, selects a data transport node (DTN) and performs the data session using the selected DTN. Whether a data session is performance sensitive may be determined using information on a process associated with the data session, by monitoring the data session, or both. The DTN may be selected from a list of available DTNs according to which DTN will probably provide the highest performance for the data session, or according to which DTN or DTNs will probably provide adequate performance. Micro speed tests, historical performance information, or other criteria may be used to predict the performance of the available DTNs. When multiple DTNs satisfy the selection criteria, a good neighbor policy may be used to select the DTN.