The current deployment of reconfigurable optical add-drop multiplexers (ROADMs) into optical networks is driving a need to monitor more than just channel wavelength, power and OSNR.  ROADMs potentially enable faster setting up and testing of new paths in the network, dynamic protection paths, and enhanced fault management.  However, realizing this potential requires a much clearer picture of the impairments inside the network, and their distribution.  

• Monitor performance and margin throughout the network, at the physical layer
• Identify the cause of increasing BER, even with multiple impairments
• Locate the source of the impairment within the network
• Optimise network performance by providing feedback to impairment compensation schemes

How we do it

The laboratory eye diagram is the most familiar tool for measuring signal quality and estimating the underlying causes, but requires clock recovery, which is format and bit-rate dependent, and can be difficult to extract for strongly distorted signals.  Instead, we use asynchronous delay-tap sampling to sample the signal and a delayed copy of it in order to produce eyeD pictures. 

The eyeD pictures are an incredibly rich representation of the signal, and have specific features that correspond to different impairments.  We can also vary the delay between the taps, giving a whole additional dimension of information beyond eye diagrams.

By applying some really smart machine learning-based techniques to analyze the portraits, we can determine both the signal quality, and give numerical measurements of the underlying impairments.

Download a one page summary of our eyeD technology for monitoring multi-impairments