Web Page Latency Quality Estimation via Probability Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods, such as the Navigation Timing API and QoE estimation models, face challenges in accurately estimating user experience quality due to varying web loading latencies even under the same network conditions, making it difficult to evaluate and improve communication service quality.
Innovation Solution
A quality estimation apparatus that acquires a set of web page loading latencies and calculates the probability that the quality of experience value is a predetermined value or more, using a predefined function to estimate user experience quality for a set of web loading latencies with different values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple web loading latency measurements are performed to evaluate average quality, then measurement reliability is improved, but measurement precision deteriorates due to varying latency values
Solution Approach 1:
The patent transforms the latency measurement from a single deterministic value to a probabilistic distribution. By representing latency as a probability distribution function rather than a fixed value, the system can capture the variability inherent in network conditions while maintaining statistical precision through the distribution parameters.
Solution Approach 2:
The patent introduces a probability distribution function as an intermediary between the raw latency measurements and the quality evaluation. This intermediary model aggregates multiple measurements into a unified probabilistic representation, enabling reliable quality assessment without losing the precision of individual measurement variations.
2Difficulty of detecting and measuring
If web loading latency is measured using Navigation Timing API, then measurement capability is improved, but quality of experience estimation accuracy deteriorates
Solution Approach 1:
The patent changes the parameter representation from deterministic latency values to probabilistic distributions. By modeling latency as a probability distribution rather than a single value, the system better captures the stochastic nature of network performance and its impact on user experience quality.
Solution Approach 2:
The patent establishes a feedback loop where latency measurements inform the probability distribution model, which in turn provides quality estimates that can guide further measurements. This iterative process refines the accuracy of QoE estimation by continuously adjusting the distribution parameters based on observed latency patterns.
Data Source
AI summary
A quality estimation apparatus includes an acquisition unit for acquiring a set of web page loading latencies, and a calculation unit for calculating a probability that a quality of experience value is a predetermined value or more for each latency in the set, as a probability that the quality of experience value is the predetermined value or more for the set, and enables estimation of the user experience quality for the set of web page loading latencies.


