Cross Domain Synchronization in Communication Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In communication networks, the instability of timing trees due to misalignment between timing distribution and data/control/management distribution, and potential master/slave role reversals during network faults or timing source failures, lead to performance oscillations and unstable synchronization.
Innovation Solution
Implementing a cross-domain synchronization mechanism that separates timing distribution and node faults through clock domain separation, allowing for independent selection of master clocks within each domain and remote timing sensing to maintain alignment and stability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a PTP network uses a master/slave timing tree for clock synchronization, then time/frequency distribution is achieved across the network, but misalignment between timing distribution and data/control/management distribution occurs, leading to instability
Solution Approach 1:
The network is divided into multiple timing domains, each with its own independent master clock and timing tree. This segmentation isolates timing distribution from data/control/management distribution, allowing each domain to maintain stable synchronization independently without causing network-wide instability when faults occur.
Solution Approach 2:
Boundary clocks are introduced as intermediary nodes between different timing domains. These boundary clocks translate and synchronize timing signals across domain boundaries, enabling independent timing domains to work together while maintaining alignment between timing distribution and data/control/management distribution.
2Reliability
If the timing tree is rebuilt due to network fault or timing source failure, then synchronization is restored, but master/slave roles are reversed on some network nodes, causing performance oscillation
Solution Approach 1:
By segmenting the network into independent timing domains with domain-unique master clocks, the patent prevents cascading master/slave role reversals. When a fault occurs in one domain, only that domain's timing tree is affected, while other domains maintain their original master/slave assignments, eliminating performance oscillation.
Solution Approach 2:
The patent pre-establishes independent master clocks in each timing domain before faults occur. This preliminary configuration ensures that when failures happen, the timing tree reconstruction within each domain does not trigger unwanted master/slave role reversals, as each domain already has its designated master clock ready to maintain stability.
3Ease of operation
If timing distribution and data/control/management distribution are aligned in a single domain, then simplified network management is achieved, but synchronization performance degrades due to instability from role reversals
Solution Approach 1:
The patent segments the network into multiple timing domains, each with independent master clocks and timing trees. This allows timing distribution to be aligned with data/control/management distribution within each domain while preventing the instability and role reversals that would degrade synchronization performance across the entire network.
Solution Approach 2:
The patent introduces a new dimensional organization by creating hierarchical timing domains. This multi-dimensional structure allows timing distribution to follow data/control/management distribution at the domain level while maintaining independent stability through domain boundaries, effectively resolving the conflict between management simplicity and synchronization performance.
Data Source
Figure 1
Figure 2(a)~2(b)
Figure 3
AI summary
A method for communications is proposed. The method may comprise receiving, by a first network node, a report of clock quality of a second network node from a third network node. A clock of the first network node is selected as a master clock for synchronization in a first timing domain, a clock of the second network node is selected as a master clock for synchronization in a second timing domain, and the third network node is attached to at least the first timing domain and the second timing domain. Based at least in part on the received report, it may be determined whether to synchronize the first timing domain to the second timing domain. In response to the determination of synchronizing the first timing domain to the second timing domain, the first network node can obtain timing information of the second network node from the third network node. The method may further comprise tuning the clock of the first network node to synchronize the first timing domain to the second timing domain, based at least in part on the timing information of the second network node.