Signal Alignment Using Relative Delay Histograms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current signal alignment methods in telecommunications networks fail to accurately align reference and received signals, particularly for audio and video, leading to suboptimal quality assessment and degradation analysis.
Innovation Solution
A method involving confidence value determination and similarity measure calculation between frames, using correlation coefficients, to align signals through coarse and fine alignment processes, with sub-sampling and normalization techniques to compensate for spatial and frequency shifts, and identify active frames for effective time-alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional signal alignment methods are used, then the alignment process is simple, but the alignment accuracy is insufficient leading to suboptimal quality assessment
Solution Approach 1:
The alignment process is divided into distinct stages: generating similarity measures between signal frames, identifying peaks in the similarity measure, assigning confidence values to peaks, and determining matching values based on confidence thresholds. This segmentation allows each stage to be optimized independently while maintaining overall accuracy.
Solution Approach 2:
The method performs preliminary actions by generating similarity measures and identifying peaks before final alignment determination. Confidence values are assigned to peaks in advance, allowing the system to pre-filter and prioritize potential matches before committing to final alignment decisions, thereby improving accuracy without proportionally increasing complexity.
2Measurement precision
If comprehensive signal analysis is performed to improve quality assessment accuracy, then the quality measure is more accurate, but the processing time increases
Solution Approach 1:
The system performs partial analysis by focusing computational resources on identifying and analyzing only the most significant peaks above a confidence threshold, rather than exhaustively analyzing all possible frame comparisons. This selective approach maintains quality assessment accuracy while reducing overall processing time.
Solution Approach 2:
The method applies different processing intensities to different parts of the signal analysis. High-confidence peaks receive detailed examination with multiple verification steps, while low-confidence regions are processed more quickly with fewer computational operations, optimizing the balance between accuracy and processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This invention relates to alignment of signals, particularly for use in a quality assessment system. The invention provides a method and apparatus for aligning a first signal comprising a sequence of frames with a second signal comprising a sequence of frames, the method comprising the steps of: determining a similarity measure between each of a plurality of frames of the first signal and each of a plurality of frames of the second signal; assigning a matching value to each frame of the first signal wherein the matching value indicates a relative position of a matching frame in the second signal, by repeating the sub-steps of: generating a relative delay histogram the histogram comprising a set of values corresponding to each of a set of relative delays by: selecting a subset of frames of the first signal and for each frame of said subset identifying the frame of the second signal having the greatest similarity with said frame; determining the relative delay between the identified frame of the second signal and said frame of the first signal; and incrementing the value of the histogram corresponding to said relative delay; identifying one or more peaks in the relative delay histogram; and assigning the matching value to each frame contributing to a peak in dependence upon said identified peaks.