Model-Based QoE Scoring for Real-Time Cloud Application Tuning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for evaluating Quality of Experience (QoE) in cloud computing environments are limited by subjective assessments that are time-consuming, laborious, and not applicable in real-time, and lack an end-to-end system that considers both infrastructure and application performance along with user satisfaction.
Innovation Solution
A model-based system that automatically measures QoE using BCI, IoT, and IoS data to determine configuration changes at the infrastructure and application layers, optimizing performance to meet user expectations and KPIs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subjective assessment methods (MOS, user interviews, surveys) are used to evaluate QoE, then user satisfaction can be captured, but the process becomes time-consuming, laborious, and not applicable in real-time
Solution Approach 1:
The patent replaces manual subjective assessment mechanisms with automated objective measurement systems. Instead of human observers conducting interviews and surveys, the system uses automated data collection from BCI devices, IoT sensors, and application logs to measure QoE continuously and in real-time, eliminating the time-consuming nature of subjective methods while maintaining measurement accuracy
Solution Approach 2:
The system enables self-service QoE measurement by automatically collecting data from multiple sources (BCI, IoT, application performance metrics) and processing it through machine learning models without requiring human intervention. The system serves itself by autonomously monitoring, analyzing, and reporting QoE metrics, making the process efficient and scalable
2Adaptability or versatility
If traditional QoE assessment methods are used, then user feedback can be obtained, but they lack an end-to-end system that integrates infrastructure and application performance with user satisfaction
Solution Approach 1:
The patent merges multiple previously separate assessment systems into a unified end-to-end QoE evaluation platform. It combines infrastructure monitoring, application performance tracking, and user satisfaction measurement (via BCI and IoT data) into a single integrated system that correlates technical metrics with user experience, providing comprehensive QoE assessment across the entire service delivery chain
Solution Approach 2:
The system implements multi-functionality by designing a platform that simultaneously performs infrastructure monitoring, application performance analysis, user physiological state detection, and QoE prediction. This universal system serves multiple purposes: real-time QoE measurement, anomaly detection, root cause analysis, and optimization recommendation, reducing the need for separate specialized tools
3Productivity
If cloud providers focus on cost-driven competition, then pricing can be optimized, but service quality differentiator is lost
Solution Approach 1:
The patent implements continuous feedback loops where QoE measurements from BCI and IoT devices are fed back to the cloud platform, which then adjusts infrastructure and application configurations to maintain or improve user experience. This real-time feedback mechanism ensures service quality consistency by automatically responding to changes in user state or system performance, preventing quality degradation even during cost-optimized operations
Solution Approach 2:
The system introduces dynamic adaptation by continuously adjusting service configurations based on real-time QoE measurements. Instead of static quality levels, the system dynamically optimizes resource allocation and service parameters to maintain consistent user experience across varying conditions, enabling reliability to adapt to changing demands while managing costs
Data Source
AI summary
A method for improving user experience associated with an application supported by a system is provided. The method comprises obtaining input data associated with a group of one or more features and determining a performance score associated with the obtained input data using a performance score generating model. The performance score indicates an estimated quality of user experience (QoE) of the application. The method further comprises determining whether to apply a configuration change based on the determined performance score. The configuration change is associated with the application and/or the system.