Network Path Selection Using Predicted QoE Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network analysis methods struggle to effectively determine the impact of communication path changes on user experience and network performance, particularly in predicting Quality of Experience (QoE) metrics, which are crucial for optimizing network pathways and adhering to Quality of Service (QoS) policies.
Innovation Solution
A system that predicts Quality of Experience (QoE) metrics for communication paths in a network by measuring network metrics such as bandwidth, latency, and jitter, and applying QoS policies to recommend optimal communication paths, taking into account historical path changes and their potential impact on network performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If communication path changes are made to improve user experience, then network performance may improve, but frequent path changes can adversely affect the network as a whole and harm user experience for many users
Solution Approach 1:
The system performs preliminary analysis of communication path changes by predicting QoE metrics before actually switching paths. This allows the system to evaluate potential impacts on network performance and user experience in advance, preventing harmful frequent changes while still enabling beneficial path optimizations.
Solution Approach 2:
The system implements a feedback mechanism where QoE metrics are continuously measured and used to inform future path selection decisions. By monitoring the actual impact of path changes on user experience and network performance, the system can adjust its behavior to avoid harmful frequent changes while maintaining service quality.
2Reliability
If network modifications are made to improve user experience, then application performance may improve, but determining the effect on user experience and network as a whole becomes prohibitively difficult
Solution Approach 1:
The system introduces QoE metrics as an intermediary measure that bridges the gap between network modifications and actual user experience impact. These metrics serve as a quantifiable proxy that simplifies the assessment of how path changes affect both application performance and overall network health, making impact evaluation tractable.
Solution Approach 2:
The system transforms the complex, multi-dimensional problem of assessing network modification impacts into a more manageable form by focusing on key QoE parameters. By measuring and analyzing specific parameters like latency, jitter, and packet loss, the system can effectively evaluate user experience impact without needing to analyze every aspect of network behavior.
3Speed
If computing devices frequently change communication paths to optimize performance, then individual user experience may improve, but the network as a whole is adversely effected
Solution Approach 1:
The system applies preliminary anti-action by predicting negative QoE outcomes before allowing path changes to occur. When the prediction indicates that a path change would harm network stability or user experience, the system preemptively prevents the change, counteracting the tendency toward frequent destabilizing modifications.
Solution Approach 2:
The system implements dynamic path selection that adapts to current network conditions while incorporating predictions of future impacts. Rather than static or purely reactive path changes, the system dynamically evaluates whether a path change is likely to improve or harm performance, enabling flexible optimization while maintaining network stability.
Data Source
AI summary
Methods and systems for changing communication paths in a network based on predicted Quality of Experience metrics are described herein. Computing devices in a network may communicate via one or more communication paths and using one or more applications. One or more Quality of Experience metrics may be determined for the one or more applications. Network metrics for the network may be measured and, based on one or more Quality of Service policies for the network, predicted Quality of Experience metrics may be determined using, e.g., a model network. A communication path recommendation may be output based on the predicted Quality of Experience metrics. For example, the recommendation may cause an application to change from a first communication path to a second communication path.