While optical fibre is a fantastic means of transportation of data it does have its limitations. There are a range of optical parameters that limit how far a given transmission system can transmit a signal before optical-electrical-optical regeneration is required. As transmission systems and fibres have developed over the last 25 years, we have also seen those limitations shift amongst the fibre parameters.
Let’s take quick look at the history that led to the present situation. In doing so we follow the evolution of the ITU standards.
Still referred to as standard transmission fibre and still making up for the absolute bulk of optical fibre cable installations. The fibre has a relatively high Chromatic Dispersion and in the early days of manufacturing no particular eforts were made to avoid Polarization Mode Dispersion. The main speed/length limiting parameter up to the 10 Gb/s systems was attenuation.
As Chromatic Dispersion was an issue even at single channel transmission the G.653 was an excellent answer with its zero-dispersion wavelength at 1550 nm. This fibre had a very short technological lifetime and manufacturing has been discontinued long ago.
With the introduction of Dense Wavelength Division Multiplexing (DWDM) zero-dispersion does not work well. Four Wave Mixing amongst the transmission channels became a new limiter. So Chromatic Dispersion was reintroduced in the fibre although at much less numbers than in the G.652 fibre.
Chromatic Dispersion could be repaired by means of Dispersion Compensating Modules – in essence a short spool of fibre with a very steep reversed Chromatic Dispersion. Polarization Mode Dispersion could not be repaired and a lot of diferent initiatives were taken to limit it both in fibre and cable production. These Dispersion Compensating Modules are also installed in great numbers in G.652 networks. While they work splendidly, they have three drawbacks: added price, added attenuation and added PMD.
Key optical parameters of G.655 were Chromatic Dispersion and dispersion slope as well as Polarization Mode dispersion. It remained the optimal choice until RAMAN based amplifier systems came around.
The RAMAN based amplifier pumps light into the fibre at approximately 100 nm below the wavelength it shall amplify. Since the G.655 fibres have a Zero Dispersion Wavelength between the pump and the transmission, Four Wave Mixing occurred frequently between the two limiting the transmission range. The G.656 was the answer. Not only does it place its Zero Dispersion Wavelength below the pump light, some manufacturers choose a lower Mode Field Diameter, which counter intuitively – in fact is an advantage for RAMAN based systems.
The observant reader will notice that I address G.654 after G.656. Indeed the G.654 was invented and standardized by ITU after G.653. The terminology in the naming of the standards follows chronology. However, the G.654 fibre was originally developed especially for transoceanic cables. The design rules of these cables are significantly diferent from terrestrial applications for a range of reasons like availability of electrical power, number of submarine repeater stations, dispersion management etc.
And since we discuss terrestrial cables here the G.654 is a new fibre. This fibre is characterized by having a very large efective area.
With the introduction of coherent transmission there was an actual disruption in the long haul / high speed optical world. Earlier Chromatic Dispersion and Polarization Dispersion were the major speed limiters in the optical network. With coherent transmission systems taking care of these previously troublesome optical distortions, today’s speed limiter becomes the amount of optical power that can be inserted into the fibre without hitting the noise floor limit.
The illustration is based upon 4 fibres and their individual performance.