Clock Synchronization Using Kalman Filter and Moving Horizon Line Fitting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network-based clock synchronization methods are sensitive to measurement outliers and network variability, leading to suboptimal performance in synchronizing client clocks with a server clock over unreliable communication networks.
Innovation Solution
A method that determines clock offsets within a specific time horizon, weights them by measurement accuracy, and uses Moving Horizon Line Fitting to adjust the client clock, incorporating Kalman filter techniques to mitigate outlier effects and improve synchronization accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional filtering methods (Kalman Filter) are used to process clock offset measurements, then the filter provides optimal estimation for random Gaussian errors, but it becomes too sensitive to delay outliers and network variability causing synchronization performance degradation
Solution Approach 1:
The patent converts the harmful effect of outliers into a useful detection mechanism by analyzing the statistical properties of measurement residuals. When residuals exceed a threshold, the system identifies these as outliers and adjusts processing accordingly, transforming what was previously harmful noise into a detectable signal for quality control.
Solution Approach 2:
The system dynamically changes processing parameters based on network conditions. When outliers are detected, the patent modifies the filtering approach by either downweighting affected measurements or switching between different filtering strategies, thereby adapting to varying network reliability conditions while maintaining synchronization accuracy.
2Device complexity
If simple averaging or low-pass filtering is applied to clock offset measurements, then the implementation is simple, but the synchronization accuracy is reduced due to inability to handle outliers effectively
Solution Approach 1:
The patent introduces dynamic adaptation into the filtering process. Rather than using a static filtering approach, the system continuously monitors measurement quality and adjusts its processing strategy in real-time. This dynamic behavior allows the system to maintain high accuracy by switching between simple and complex processing modes based on actual network conditions.
Solution Approach 2:
The patent introduces an intermediary quality assessment layer between raw measurements and final synchronization. This intermediary component evaluates measurement reliability and mediates which measurements should influence the final clock offset, thereby improving accuracy without requiring complete redesign of the filtering architecture.
3Ease of operation
If batch line fitting methods are used to detect outliers, then outlier detection is simplified by assuming positive delay outliers, but the method assumes symmetric bi-directional links which may not hold in practice
Solution Approach 1:
The patent segments the outlier detection process into multiple independent stages: first detecting potential outliers through residual analysis, then verifying them through additional statistical tests, and finally applying appropriate corrections. This segmentation allows the system to handle asymmetric networks by treating each direction's measurements independently while maintaining overall simplicity.
Data Source
AI summary
The present invention is directed towards a method, a system, and a device which allow an improved synchronization of a system-wide timing information over a network. Hence, slave clocks can be synchronized to a high quality clock, such as a master clock.