Fibre optic communication subsystem test procedures - Part 2-9: Digital systems - Optical signal-to-noise ratio measurement for dense wavelength-division multiplexed systems
This part of IEC 61280 provides a parameter definition and a test method for obtaining optical signal-to-noise ratio (OSNR) using apparatus that measures the optical spectrum at a multichannel interface. Because noise measurement is made on an optical spectrum analyzer, the measured noise does not include source relative intensity noise (RIN) or receiver noise. Three implementations for an optical spectrum analyser (OSA) are discussed: a diffractiongrating-based OSA, a Michelson interferometer-based OSA, and a Fabry-Perot-based OSA. Performance characteristics of the OSA that affect OSNR measurement accuracy are provided. A typical optical spectrum at a multichannel interface is shown in Figure 1. Important characteristics are as follows. • The channels are placed nominally on the grid defined by ITU Recommendation G.694.1.[4] • Individual channels may be non-existent because it is a network designed with optical add/drop demultiplexers or because particular channels are out of service. • Both channel power and noise power are a function of wavelength. For calculating the OSNR, the most appropriate noise power value is that at the channel wavelength. However, with a direct spectral measurement, the noise power at the channel wavelength is included in the signal power and is difficult to extract. An estimate of the channel noise power can be made by interpolating the noise power value between channels. The accuracy of estimating the noise power at the signal wavelength by interpolating the noise power at an offset wavelength can be significantly reduced when the signal spectrum extends into the gap between the signals and when components such as add-drop multiplexers along the transmission span modify the spectral shape of the noise. These effects are discussed in further detail in Annex B, and can make the method of this document unusable for some situations. In such cases, where signal and noise cannot be sufficiently separated spectrally, it is necessary to use more complex separation methods, like polarization or time-domain extinction, or to determine signal quality with a different parameter, such as RIN. This is beyond the scope of the current document.