Passive QoE Estimation via Network Intermediary
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating webpage load time are inadequate, as they either rely on indirect metrics, user complaints, privacy-invasive measurements, or active tests that introduce network traffic and have limitations such as time lags and biases, or are restricted by encryption in HTTPS environments.
Innovation Solution
A computer-implemented method that predicts Quality of Experience (QoE) by generating a feature vector from connection-level statistics measured at a network node, estimating download performance metrics, and calculating QoE values, which includes detecting transport-layer connections, generating connection-level data, and forming feature vectors to estimate webpage load times and QoE without direct user measurement or active testing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a measurement application is installed on the user's device to directly measure webpage load times, then measurement precision is improved, but ease of operation deteriorates due to privacy concerns and user perception of encroachment
Solution Approach 1:
The patent introduces an intermediary measurement entity (separate from both the user device and the web server) that passively monitors and measures webpage load times. This intermediary captures network traffic and derives load time metrics without requiring direct installation on user devices, thereby maintaining measurement precision while eliminating privacy concerns and user acceptance issues.
2Measurement precision
If Java script is inserted at the start of the webpage request to monitor HTTP requests, then measurement precision is improved, but adaptability deteriorates for HTTPS websites where object requests are encrypted
Solution Approach 1:
The patent positions an intermediary measurement entity at the network level that can observe and analyze encrypted HTTPS traffic without requiring decryption. This intermediary captures the timing and characteristics of encrypted requests and responses, enabling accurate webpage load time measurement across both HTTP and HTTPS protocols, thus maintaining adaptability and versatility.
3Measurement precision
If active tests are used to measure webpage load times, then measurement precision is improved, but productivity deteriorates due to inability to operate continuously and introduction of additional network traffic
Solution Approach 1:
The patent implements a continuous passive measurement system that operates continuously without interruption. The intermediary measurement entity continuously monitors network traffic passing through it, capturing webpage load time data for all users in real-time without requiring active test initiation. This eliminates the discontinuity and additional traffic overhead associated with active testing methods.
4Ease of operation
If user complaints are counted to estimate webpage load times, then ease of operation is improved, but measurement precision deteriorates due to time lag and negative bias
Solution Approach 1:
The patent implements preliminary direct measurement of webpage load times by the intermediary entity before user complaints can be generated. By measuring the actual load time objectively and directly, the system eliminates the time lag between poor network conditions and user complaint submission, as well as the negative bias where users only report bad experiences rather than good ones.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods are disclosed, and one includes a computer-implemented method for predicting quality of experience (QoE) for application downloading a webpage from a server stack, including determining a stat data corresponding to a traffic through a network node, generating a feature vector using the stat data, estimating a download performance metric for the application in a downloading of the webpage, using feature vector data, and estimating a QoE value, using the estimated performance metric.