Dynamic Application Selection via Network Impairment Metrics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network management systems rely on static Service Level Agreements (SLAs) to ensure Quality of Experience (QoE) for applications, which fail to account for application-specific factors and network dynamics, leading to suboptimal user experience due to reliance on generic thresholds that do not consider the variability in network impairments and application resilience.
Innovation Solution
A device identifies impairment scenarios in a network, estimates QoE metrics for multiple applications, and selects the most suitable application based on these metrics to provide users with the best experience, leveraging machine learning models and cross-layer telemetry to dynamically adjust routing and application usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If static Service Level Agreements (SLAs) with generic thresholds are used to ensure Quality of Experience (QoE), then network management is simplified and easier to operate, but the system fails to account for application-specific factors and network dynamics, leading to suboptimal user experience
Solution Approach 1:
The patent transforms static SLAs into dynamic systems that automatically adapt to changing network conditions and application characteristics. The system continuously monitors network parameters (delay, loss, jitter) and application performance metrics, then dynamically adjusts routing decisions and application selections based on real-time conditions, resolving the contradiction between operational simplicity and QoE reliability.
Solution Approach 2:
The system changes the parameters used for QoE assessment from generic SLA thresholds to application-specific parameters that account for different application resilience characteristics. By incorporating application-aware parameters and using machine learning models to predict QoE based on multiple factors (network conditions, application type, user device characteristics), the system achieves both simplicity and reliability.
2Device complexity
If generic SLA thresholds are applied to all applications, then network management complexity is reduced, but the system cannot account for application-specific resilience characteristics such as different codec tolerances to packet loss
Solution Approach 1:
The patent segments the network management approach by application type, creating application-specific QoE models and thresholds. Different application categories (voice, video, data) receive tailored management strategies based on their specific requirements and resilience characteristics. This segmentation allows the system to handle application-specific nuances without overwhelming complexity, as each segment can be managed with appropriate granularity.
Solution Approach 2:
The system enables applications to effectively 'self-serve' by automatically selecting the most appropriate application or routing path based on current network conditions and application characteristics. The machine learning models autonomously make decisions about application selection and routing without requiring manual configuration for each application, reducing management complexity while maintaining high adaptability.
3Speed
If traditional SLA-based routing is used, then routing decisions are simple and fast, but the system cannot predict or mitigate SLA failures, leading to degraded user experience during network impairments
Solution Approach 1:
The system performs preliminary actions by proactively predicting potential SLA failures before they occur. Machine learning models analyze current network trends and predict future QoE degradation, allowing the system to pre-switch to alternative applications or routing paths before actual failures occur. This predictive capability maintains fast routing decisions while improving reliability through advance preparation.
Solution Approach 2:
The system implements continuous feedback loops where QoE metrics are monitored in real-time, fed back to the machine learning models, which then adjust routing decisions dynamically. This feedback mechanism enables the system to respond quickly to changing conditions while maintaining reliable QoE, as the feedback drives continuous optimization without slowing down routing decisions.
Data Source
AI summary
In one embodiment, a device may identify a plurality of impairment scenarios for a network. The device may estimate quality of experience metrics for a plurality of applications accessible via the network for each of the plurality of impairment scenarios. The device may select a particular application from among the plurality of applications based on a comparison between the quality of experience metrics for the plurality of applications. The device may provide an indication for presentation by a user interface that a user should use the particular application from among the plurality of applications.