Multi-Input Timing Recovery Over Packet Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Timing recovery in packet networks is hindered by packet delay variations, resulting in high jitter and wander in the recovered reference clock, which is not acceptable for many applications, especially legacy services, due to reliance on a single stream of timing packets for synchronization.
Innovation Solution
A method that exploits space diversity in timing packets from the same clock source arriving through different network paths, involving the calculation and filtering of time delays to minimize packet delay variations, weighting and combining these delays based on residual error and sample rate, and using a digital controlled oscillator for clock recovery.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single stream of timing packets is used for clock synchronization, then the system complexity is low, but the recovered reference clock exhibits high jitter and wander due to packet delay variations
Solution Approach 1:
The patent combines multiple timing packet streams from different network paths into a unified synchronization system. By merging the delay measurements from multiple streams and applying weighted combination based on residual error and sample rate, the system achieves improved synchronization quality with reduced jitter and wander while maintaining manageable complexity through systematic integration.
Solution Approach 2:
The patent transitions from a single-dimension approach (one timing stream) to a multi-dimensional approach by utilizing timing packets arriving through different network paths. This dimensional expansion in the form of multiple independent streams provides diversity that mitigates the impact of packet delay variations on synchronization quality.
2Reliability
If multiple packet streams from different network paths are used for timing recovery, then robustness to packet impairments and fault tolerance improve, but the device complexity increases due to filtering and combining multiple streams
Solution Approach 1:
The patent dynamically adjusts the weighting parameters for each timing stream based on their residual error and sample rate characteristics. By changing these parameters adaptively, the system optimizes the combination of multiple streams to maximize robustness while minimizing the computational complexity through parameter-based prioritization rather than equal processing of all streams.
Solution Approach 2:
The patent creates multiple copies of the timing measurement process through parallel packet streams from different network paths. Each stream provides an independent copy of the delay measurement, and through filtering and weighted combination, the system extracts a more reliable synchronization signal that is robust to impairments in any single stream.
3Reliability
If timing packets are filtered to minimize packet delay variations, then the jitter and wander in recovered clock reduce, but the processing time and computational load increase
Solution Approach 1:
The patent applies partial filtering by selectively processing multiple timing streams with different weighting factors rather than uniformly filtering all data. The filtering operation focuses computational resources on the most reliable streams (those with lower residual error), achieving effective jitter and wander reduction while minimizing unnecessary processing of lower-quality streams.
Data Source
AI summary
In a method of recovering timing information over packet networks, a receiver receives a plurality of packet streams over different paths from the same source. The raw delays experienced by the timing packets for each stream are filtered to provide a filtered delay for each stream. The filtered delays are weighted based on the quality of each stream, and the weighted filtered delays are then combined to form an aggregate delay estimate. Frequency adjustments for a local clock at the receiver are derived from the aggregate delay estimate.