Multi-Window Least Squares Timing Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing timing-over-Packet (ToP) technologies face challenges in accurately synchronizing digital controlled oscillators (DCOs) across asynchronous packet networks due to frequency deviations between local oscillators at the transmitter and receiver, requiring efficient methods to adjust the receiver's DCO center frequency while minimizing data transmission impact.
Innovation Solution
A method using a multi-window least squares approach to compute frequency deviations by processing time-stamped packets, adjusting the receiver's DCO center frequency to synchronize with the transmitter's, employing overlapping data collection windows and periodic updates to minimize error and enhance synchronization speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If timing packets are transmitted at a low rate to minimize impact on data transmission, then data transmission is less affected, but the time required to collect sufficient data for accurate DCO adjustment increases considerably
Solution Approach 1:
The patent divides the data collection process into multiple parallel windows (first window, second window, etc.) that operate simultaneously. Each window collects timing data independently, allowing the system to accumulate sufficient data faster without increasing the overall packet transmission rate. This segmentation resolves the contradiction by enabling parallel data gathering while maintaining low transmission overhead.
Solution Approach 2:
The patent implements preliminary data collection in multiple overlapping windows before making DCO adjustments. By preparing timing data in advance across multiple windows, the system reduces the waiting time for sufficient data accumulation while still transmitting packets at a low rate. The overlapping windows ensure data readiness ahead of adjustment events.
2Productivity
If multiple parallel data collection windows are used to accelerate DCO adjustment, then synchronization speed improves, but system complexity increases
Solution Approach 1:
The patent segments the data collection function into multiple independent windows with dedicated counters and accumulators. Each window operates autonomously, simplifying the control logic within each segment while achieving complex overall functionality through parallel operation. This segmentation enables faster synchronization without proportionally increasing overall system complexity.
Solution Approach 2:
The patent employs periodic window management where windows are activated in sequence and reset after completing their data collection cycle. This periodic operation pattern simplifies the control mechanism compared to continuous complex processing, as each window follows a regular activate-collect-reset cycle, reducing the overall system complexity while maintaining high synchronization speed.
Data Source
AI summary
Timing information is transmitted over a network by sending time-stamped packets between a transmitter and receiver. The time-stamped packets are used to compute an estimate of the frequency deviation between the transmitter clock and the receiver clock. The local receiver clock is periodically updated to match the transmitter clock based on the estimate, which is preferably a least squares estimate. A multiple window approach is employed to increase update speed.