VoLTE Quality Assessment via Time Segment Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing VoLTE voice and video quality evaluation methods are inaccurate due to their focus on complete call services, which fail to capture poor user perception during periods of high packet loss, as they average out quality over the entire call, masking significant impairments in specific time segments.
Innovation Solution
A data processing method that classifies data packets into different time segments to determine feature parameters for each segment, allowing for separate MOS evaluations, thereby improving the accuracy of voice and video quality assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If MOS evaluation is performed based on feature parameters of a complete call service, then the evaluation covers the entire call duration, but the accuracy of voice and video quality perception is reduced due to averaging effects that mask impairments in specific time segments
Solution Approach 1:
The patent divides a complete call service into multiple time segments based on packet loss characteristics. Each segment is evaluated separately to identify periods of poor quality perception. This segmentation allows the system to capture transient impairments that would be masked in a complete-call average, thereby improving measurement precision without excessive complexity increase.
Solution Approach 2:
The patent applies different evaluation criteria to different time segments based on their packet loss characteristics. High packet loss segments are identified and weighted differently from low packet loss segments, allowing local quality variations to be reflected in the overall evaluation. This resolves the contradiction by making the evaluation method sensitive to local impairments while maintaining a comprehensive view of the entire call.
2Ease of operation
If packet loss rate is averaged over the entire call service, then the evaluation is simplified, but the ability to capture poor user perception during high packet loss periods is lost
Solution Approach 1:
The patent segments the call into time periods based on packet loss thresholds, preserving information about when poor quality occurred. Instead of a single averaged value, the system maintains segment-level metrics that capture transient impairments while still providing a simplified overall evaluation framework.
Solution Approach 2:
The patent changes the evaluation parameter from a simple average packet loss rate to a segment-based quality metric that weights different time periods differently. This transformation preserves critical quality information while maintaining operational simplicity through automated segment identification and weighted scoring.
3Measurement precision
If separate MOS evaluations are performed for each time segment, then the accuracy of quality assessment is improved, but the computational complexity increases
Solution Approach 1:
The patent segments the call into a limited number of time periods based on packet loss characteristics, performing MOS evaluation only on these segments rather than continuous monitoring. This reduces computational resource consumption compared to full continuous analysis while maintaining improved accuracy over complete-call averaging.
Solution Approach 2:
The patent performs evaluations on selected time segments rather than all possible segments, applying the principle of partial action. By focusing computational resources on segments with significant packet loss or quality variations, the system achieves improved accuracy without the excessive computational cost of evaluating every possible time window.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
Embodiments of this application disclose a data processing method and a client. The method in the embodiments of this application includes: if obtaining a first data packet and a second data packet of a target call service initiated by a terminal, classifying, by a client, the first data packet into a first data packet set and classifying the second data packet into a second data packet set based on a first time segment and a second time segment, where a first obtaining moment of the first data packet belongs to the first time segment, a second obtaining moment of the second data packet belongs to the second time segment, the first time segment and the second time segment are different time segments used by the client to obtain the data packets of the target call service, and an intersection set of the first time segment and the second time segment is empty; determining, by the client, a first feature parameter corresponding to the first data packet set and a second feature parameter corresponding to the second data packet set; and calculating, based on a preset audio and/or video quality evaluation algorithm, audio and/or video quality corresponding to the first feature parameter, and audio and/or video quality corresponding to the second feature parameter.