Call Quality Testing via Intermediary Reference Signal Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional voice communication networks, including PSTN, VOIP, and cellular networks, face issues such as signal distortion, latency, packet loss, jitter, and interference, which affect voice quality, and existing quality measurement metrics like PSQM, PAMS, and MOS have limitations in accurately assessing these issues, especially in packet-switched networks.
Innovation Solution
A call quality testing system that generates a call across a network, analyzes the received audio signal using a voice analyzer, and provides scoring metrics like PAMS, PSQM, and MOS, while also accounting for factors like echo cancellation and voice activity detection, to assess and improve voice quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional voice quality measurement metrics (PSQM, PAMS, MOS) are used to evaluate voice quality in packet-switched networks, then voice quality assessment is provided, but accurate assessment of packet loss, jitter, and latency effects is insufficient
Solution Approach 1:
The patent introduces an intermediary reference voice signal that traverses the same packet-switched network path as the test signal. By comparing the received test signal against this reference signal that has experienced identical network conditions, the system accurately measures the effects of packet loss, jitter, and latency without requiring complex mathematical models. This intermediary reference acts as a mediator that captures the actual network degradation.
Solution Approach 2:
The system creates a copy of the reference voice signal and sends it through the packet-switched network alongside the test signal. This copied reference signal serves as a baseline that has been subjected to the same network conditions, enabling direct comparison and accurate measurement of quality degradation caused by packet loss, jitter, and latency.
2Measurement precision
If active testing with reference voice signals is performed to measure voice quality, then speech quality scores are obtained, but the testing process is complex and requires active participation
Solution Approach 1:
The system employs voice activity detection that automatically detects and analyzes speech segments without requiring manual intervention or complex processing. The reference signal and test signal self-compare through the network, and the system automatically generates quality measurements. This self-service approach simplifies the testing process while maintaining measurement precision.
3Measurement precision
If multiple scoring metrics (PAMS, PSQM, PESQ, MOS) are calculated to comprehensively evaluate voice quality, then detailed quality assessment is provided, but processing time and computational resources increase
Solution Approach 1:
The system pre-calculates and stores multiple scoring metrics (PAMS, PSQM, PESQ, MOS) during the signal comparison process. By performing these calculations in advance as part of the single pass comparison between reference and test signals, the system provides comprehensive quality assessment without requiring separate processing steps, thus minimizing processing time while maintaining measurement comprehensiveness.
Data Source
AI summary
A method, apparatus and computer readable medium for call quality testing is presented. A query is transmitted over a communications network from a first location to a second location. The query results in an audio signal at the second location, which is received at the first location. The audio signal is analyzed by comparing the signal with a reference signal clip. A statistical parameter is generated, the statistical parameter indicative of a quality of the received signal.