Phase Difference Measurement for Network Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for computing time deviation (TDEV) in packet-based networks are inefficient due to packet delay variation and queuing delays, which affect the accuracy of timing synchronization and quality of service.
Innovation Solution
The system estimates phase differences between clocks of a master and slave device, computes average phase differences within specific percentiles, and adjusts timing signals to improve synchronization, using techniques like one-step and two-step Precision Time Protocol (PTP) and circuit emulation services (CES), while employing sliding window statistics and circular buffers to reduce computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods are used to compute time deviation in packet-based networks, then the computation is simpler, but the accuracy of timing synchronization deteriorates due to packet delay variation and queuing delays
Solution Approach 1:
The patent segments the set of phase difference measurements by applying percentile thresholds to divide them into subsets. This segmentation allows the system to exclude outlier measurements caused by packet delay variation and queuing delays, thereby improving timing synchronization accuracy without requiring complex filtering algorithms.
Solution Approach 2:
The patent computes average phase differences only for specific percentile subsets of measurements rather than using all measurements. This partial action approach focuses computational resources on the most reliable measurements, improving accuracy while maintaining reasonable computational complexity.
2Measurement precision
If all phase difference measurements are used to compute average phase difference, then the computation is more comprehensive, but the accuracy deteriorates due to noise from packet delay variation
Solution Approach 1:
The patent extracts and excludes outlier measurements from the set of all phase difference measurements by applying percentile thresholds. This extraction removes noisy data points caused by packet delay variation and queuing delays, improving the accuracy of the average phase difference computation while reducing the quantity of measurements used.
3Productivity
If conventional TDEV computation is used, then the method is more straightforward, but the efficiency deteriorates due to computational overhead in packet-based networks
Solution Approach 1:
The patent performs preliminary sorting of phase difference measurements and pre-computation of percentile thresholds before calculating the average. This preliminary action organizes the data in advance, enabling more efficient computation of the average phase difference for the selected percentile subset, thereby improving overall computation efficiency.
Data Source
AI summary
In response to communications from a first device to a second device, respective phase differences are estimated between a first clock of the first device and a second clock of the second device. A first average phase difference is computed within a percentile of a first subset of the respective phase differences. The percentile is less than 100. A second average phase difference is computed within the percentile of a second subset of the respective phase differences. The second subset is a modification of the first subset. The second average phase difference is computed in response to the first average phase difference and the modification.