QoE Optimization System Using Emulation Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing Quality of Experience (QoE) optimization techniques are inefficient as they rely on passive user complaints and complex algorithms, failing to accurately determine key performance indicators (KPIs) and system control parameters, leading to delayed performance adjustments and inadequate emulation validation.
Innovation Solution
A QoE optimization system and method that collects KPIs and system control parameters, uses a pre-trained optimization model for non-real-time adjustments, and employs emulators to emulate and validate system control parameters, improving user experience through AI-driven optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If passive user complaints are used for QoE optimization, then system complexity is reduced, but optimization speed and user experience quality deteriorate
Solution Approach 1:
The system proactively collects KPIs and determines optimization parameters before users experience quality degradation, rather than waiting for complaints. This preliminary action enables the system to preemptively adjust parameters to maintain QoE, resolving the contradiction by shifting from reactive to proactive optimization.
Solution Approach 2:
The system implements continuous feedback loops where KPIs are collected, analyzed, and used to dynamically adjust optimization parameters. This feedback mechanism enables real-time adaptation to changing network conditions, significantly improving optimization speed while maintaining manageable system complexity through automated closed-loop control.
2Measurement precision
If complex QoE optimization algorithms are used, then optimization accuracy is improved, but processing time and system complexity increase
Solution Approach 1:
The system pre-calculates and stores optimization parameters offline for various network scenarios before deployment. During runtime, it performs rapid parameter selection from pre-computed options rather than executing complex real-time calculations, thereby maintaining high optimization accuracy while dramatically reducing processing time.
Solution Approach 2:
The optimization problem is divided into offline training phase and online selection phase. The complex algorithm work is segmented to be performed offline where time is not critical, while the online phase only requires fast parameter selection based on current KPIs, resolving the time-accuracy tradeoff.
3Device complexity
If conventional comparison or polynomial regression approaches are used, then system simplicity is maintained, but ability to handle high-dimensional mapping deteriorates
Solution Approach 1:
The system introduces neural network models as intermediary components that bridge the gap between simple system architecture and complex high-dimensional optimization tasks. The neural networks handle the computationally intensive high-dimensional mapping offline, while the main system remains relatively simple and only performs inference during runtime.
Solution Approach 2:
Conventional mathematical approaches (comparison, polynomial regression) are replaced with machine learning-based neural network models. This substitution enables the system to handle high-dimensional non-linear mappings that are intractable with traditional methods, while the overall system structure remains manageable through modular deployment.
4Ease of manufacture
If periodic network parameter collection is used, then implementation simplicity is improved, but ability to perform emulation validation deteriorates
Solution Approach 1:
The system creates virtual copies (emulators) of network components to simulate and validate optimization strategies before deploying them to the real network. This copying approach enables comprehensive validation without adding significant implementation complexity, as the emulators run in isolated test environments.
Solution Approach 2:
Emulation validation is performed preliminarily before actual network deployment. The system tests optimization parameters in simulated environments first, identifying potential issues before they affect real users. This preliminary validation step ensures reliability while maintaining implementation simplicity through standardized testing procedures.
Data Source
AI summary
A Quality of Experience (QoE) optimization system and method are provided. An electronic device inputs key performance indicators (KPIs) and system control parameters collected from a core network, a base station and a user equipment (UE) into a QoE optimization model. The QoE optimization model then optimizes the system control parameters based on the KPIs and a user QoE fed back from the UE to output optimized system control parameters. Furthermore, a strategy emulator controls at least one of a base station emulator and a UE emulator, so as to emulate the QoE optimization model using the at least one of the base station emulator and the UE emulator. Non-real-time optimization adjustments to the QoE optimization model are made based on the result of the emulation performed by the at least one of the base station emulator and the UE emulator.