Throughput Mode Distribution for QoE Estimation in SD-WAN
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining Quality of Experience (QoE) in software-defined wide area networks (SD-WANs) rely on static Service Level Agreement (SLA) templates, which are often weakly correlated with actual user experience, especially for applications like voice and video, and fail to accurately discern QoE.
Innovation Solution
A device calculates distributions of bitrates associated with an application's traffic across network paths, detects throughput modes, and associates these with quality of experience labels to estimate QoE metrics without relying on active QoS probing or static SLA templates, using traffic telemetry information and machine learning techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If static SLA templates are used to measure QoE, then the measurement process is simple and standardized, but the accuracy and correlation with actual user experience deteriorates
Solution Approach 1:
The patent changes the measurement parameters from static SLA templates to dynamic throughput mode distributions. Instead of using fixed thresholds and categories, the system continuously monitors and analyzes actual throughput values, identifying modes and their probabilities to accurately reflect user experience variations.
Solution Approach 2:
The patent replaces the mechanical/procedural approach of static SLA template matching with a statistical/probabilistic approach. By using throughput mode distributions and their associated probabilities, the system substitutes rigid categorical assessment with flexible statistical modeling that adapts to actual network conditions.
2Measurement precision
If active QoS probing is used to estimate QoE, then the measurement can be performed, but the network overhead and complexity increases
Solution Approach 1:
The patent enables the network to self-measure QoE by utilizing existing throughput telemetry data that is already being collected for routing decisions. Instead of requiring separate active probing mechanisms, the system leverages the network's own operational data to infer QoE metrics.
Solution Approach 2:
The patent makes the throughput telemetry system multi-functional by using the same data collection infrastructure for both routing optimization and QoE estimation. The existing telemetry mechanisms serve dual purposes, eliminating the need for dedicated probing systems.
3Measurement precision
If throughput mode distribution analysis is used, then QoE accuracy improves, but the computational complexity and processing requirements increase
Solution Approach 1:
The patent applies partial action by focusing computational resources on identifying and analyzing only the significant throughput modes rather than processing every individual throughput measurement in detail. By concentrating on mode identification and their probability distributions, the system achieves accurate QoE estimation with reduced computational overhead.
Data Source
AI summary
In one embodiment, a device calculates one or more distributions of bitrates associated with an application whose traffic is conveyed via one or more paths in a network. The device detects throughput modes of the application, based on the one or more distributions of bitrates associated with the application. The device associates each throughput mode with a quality of experience label, to form a plurality of pairs of throughput modes and quality of experience labels. The device estimates a quality of experience metric for the application, based on a bitrate of the application and the plurality of pairs of throughput modes and quality of experience labels.