Optical Network Latency Measurement Using Overhead Bytes
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In optical transport networks, real-time latency measurement is challenging due to the static nature of traditional SONET/SDH networks, and the dynamic changes in topology in self-healing mesh networks, making it impractical to measure latency for all possible paths.
Innovation Solution
A method for real-time latency measurement over optical links involves issuing a latency measurement command between nodes, using timers to calculate latency while filtering out frame skew, and adjusting administrative weights based on updated measurements, which can be integrated with optical control planes to operate in-service without network impact.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If in-service latency measurement is implemented in dynamic mesh networks, then measurement accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces an intermediary measurement mechanism using overhead bytes in existing optical protocols (OTN, SDH, SONET) to carry timing information between nodes. This intermediary approach allows latency measurement without adding complex dedicated measurement equipment, as the overhead bytes serve as mediators that transport timing data through the network infrastructure already in place.
Solution Approach 2:
The measurement system utilizes the network's own existing infrastructure and protocols to perform measurements. By embedding timing information in the overhead bytes of data frames that already traverse the network, the system makes the network self-measuring without requiring external probe packets or separate measurement channels, thereby reducing overall system complexity.
2Loss of information
If real-time latency measurement is performed across all possible paths, then measurement completeness is improved, but measurement time increases
Solution Approach 1:
The patent performs preliminary actions by continuously maintaining timing synchronization and recording time stamps at each node using the network's existing synchronization infrastructure. This preliminary timing information is prepared in advance and stored in the overhead bytes, so when latency measurement is needed, the calculation is immediate rather than requiring new measurement actions for each path query.
Solution Approach 2:
The system implements periodic latency measurements at configured intervals rather than continuously monitoring all paths. The measurement command is issued periodically, and timing information is collected at regular intervals, allowing the network to maintain up-to-date latency data without constant measurement overhead, thus balancing completeness with time efficiency.
3Adaptability or versatility
If packet-based latency measurement is used, then adaptability to dynamic topologies is improved, but network overhead increases
Solution Approach 1:
The patent merges the latency measurement function with the existing data transmission function by embedding timing information in the overhead bytes of data frames. Instead of using separate packet-based measurement protocols that would double the overhead, the measurement data is combined with the data plane traffic, so the same physical infrastructure carries both data and measurement information simultaneously.
Solution Approach 2:
The overhead bytes in optical network frames are given multi-functionality, serving both as protocol control information and as carriers for latency timing data. This universal use of existing protocol elements allows the system to adapt to dynamic topologies without introducing dedicated measurement packets, thereby maintaining topology adaptability while minimizing additional network overhead.
Data Source
AI summary
The present disclosure provides systems and methods for real-time, in-service latency measurements over optical links that may be further integrated within various optical control planes. The present invention may utilize minimal unused overhead to calculate latency of an optical line through a transport network. The present invention utilizes timers at two end-point nodes associated with the optical line, and includes a mechanism to filter out frame skew between the nodes. Advantageously, the present invention provides a highly accurate latency measurement that may calculate latency on links as small as one meter, an in-service algorithm operable without network impact, and may be integrated with an optical control plane to automatically provide administrative weight variables associated with link costs.