End User Experience Scoring for Delay Cause Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing metrics for estimating end user experience (EUE) quality, such as round-trip network delay (RTND), application delay, data delivery time, and retransmissions, do not provide clear indications of the causes of EUE degradation, making it difficult to identify and address performance issues effectively.
Innovation Solution
A scoring platform calculates client, network, and application delay scores based on specific metrics to determine an end user experience score, identifying the sources of delays and enabling targeted actions to improve user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional EUE metrics (RTND, application delay, data delivery time, retransmissions) are used, then EUE quality estimation is simplified, but the ability to identify specific delay causes is lost
Solution Approach 1:
The patent segments the end-to-end delay into four distinct components: client delay, network delay, server delay, and application delay. Each component is measured and scored separately using specific metrics and machine learning models, enabling precise identification of delay sources while maintaining automated monitoring through structured data collection and processing pipelines.
2Measurement precision
If detailed delay metrics are collected from all components, then delay cause identification accuracy is improved, but computing resource consumption increases
Solution Approach 1:
The patent uses synthetic traffic generators to create representative copies of actual traffic patterns, allowing delay measurement without processing every real data packet. Machine learning models are trained on sampled data to predict delay characteristics, significantly reducing computing resource requirements while maintaining measurement accuracy.
Solution Approach 2:
The system performs preliminary actions by pre-configuring measurement probes, synthetic traffic generators, and machine learning models before actual delay measurement begins. This preparation enables automated, efficient data collection and analysis without requiring intensive real-time processing resources during operation.
3Productivity
If automated scoring and action selection is implemented, then handling efficiency is improved, but system complexity increases
Solution Approach 1:
The scoring platform is designed as a universal system that handles multiple types of delay metrics (client, network, server, application), applies various machine learning models, and selects from diverse remediation actions. This multi-functional architecture consolidates complexity into a single automated platform that can manage EUE monitoring across different applications and network conditions using standardized processes.
Data Source
AI summary
A device receives client delay metrics associated with a client device, network delay metrics associated with a network, server delay metrics associated with a server device, and application delay metrics associated with an application provided by the server device, wherein the client device and the server device communicate via the network. The device calculates client delay scores based on the client delay metrics, and calculates network delay scores based on the network delay metrics. The device calculates server delay scores based on the server delay metrics, and calculates application delay scores based on the application delay metrics. The device calculates an end user experience score based on the client delay scores, the network delay scores, the server delay scores, and the application delay scores, and performs an action based on the end user experience score.