The diagrams of the spectrum bands are wild for me (coming from the RF world) - in that world, a 2GHz channel that I'd used in some systems was considered ridiculously huge, but here in fibre the 'small' channels are 50GHz!
People really don't get the enormity of the difference - when there were policy debates in my country about rolling our new fixed line infrastructure there were literally people saying "but won't all homes and businesses just be able to use wireless in the future?"
My armchair guess is that because traffic hitting the cable is already serialized in some way that larger channels make sense? Of course, those large channels could also be multiplexed in some way and most long-range lines run DWDM/OTN, so I'm just as likely to be talking out of my ass.
I think the point is that fiber can have wayyyyyy more total bandwidth than any wireless technology, so you can afford to make your channels much larger (with more bandwidth). The only reason to make a channel smaller is to make room for a different channel.
I am surprised there are repeaters involved. Is this because of the imperfections of the surface of the fibreglass tubes that cause decay of precision of the reflection over long distances(a visual noise)?
Even the best optical fiber transceivers and glass are limited (practically) to about 100km; repeaters are typically placed every 60-70km. The technology for delivering power to the repeaters is fascinating. They inject 5,000-10,000VDC at one end and each repeater shunts off a tiny amount of current to power the amplifier. All of this is embedded in the cable itself before being loaded onto the cable ship.
The history behind TAT-1, the first transatlantic telephone cable, and the repeaters used, is fascinating. Bell Labs designed the repeaters. The repeaters used vacuum tubes for amplification and were designed for extreme reliability. The flexible repeaters were integrated into the cable like modern cables.
The tubes were tested to an extremely high standard. Only a small fraction of the manufactured tubes were selected after testing: Bell Labs designed a test regime over 18 years to detect minute flaws in manufactured tubes
The cable and its 306 tubes operated for 22 years with no failures.
To add more details to other replies you received, the primary factors are Rayleigh scattering and impurities absorbing light energy, at 1550nm (where this loss is least pronounced) the number that usually gets thrown around is 0.2dB/km in attenuation. That adds up to needing those repeaters at the intervals we have them.
I think it's mostly that the fiber isn't a perfectly transparent medium, over tens of kilometers attenuation adds up. As said in https://news.ycombinator.com/item?id=45159639 these are just to boost power, they don't reform the signal.
Optical repeaters are 1R repeaters, I.e. they regenerate power. Inside the repeater "boxes" (they are actually cylinders) there is an optical amplifier. For typically these are Erbium doped fiber amplifiers (EDFA). I other words a piece of fibre doped with Erbium (a rare earth). The amplifiers are pumped with laser diodes (typically 1-4 per EDFA) at 980 nm and 1480 nm wavelength. By pumping the doped fibre with these wavelength you provide high gain to the telecom channels which are usually in the optical C-band (~1525-1565nm). This way you can reamplify signals over a large bandwidth (~4 THz) without having to do detection and retransmission (which would be unscalable). Repeaters are typically spaced at 60-80 km in submarine, with a "transparent" design (the gain compensates for the transmission loss of the 60km fibre).
Power delivery to the laser diodes is done through the metal jacket of the cable. The whole submarine cable is essentially a very long DC transmission line. Which is a fascinating topic in itself, E.g. What is ground in such a line, it will differ by 1000s of Volts between continents.
> This way you can reamplify signals over a large bandwidth (~4 THz) without having to do detection and retransmission (which would be unscalable).
This trick also means the cable doesn't care about the rest of the technology. If it was a retransmitter then we'd need to replace the entire cable if we change from 100Gbps over Protocol #39 to 200 Gbps over Protocol #40 because every retransmitter needs to be equipped for the new protocol, but the optical amplifier doesn't care why these photons turned up, what they mean - when provided with power it just ensures proportionately more photons like them come out of the amplifier.
Because they're not actually the same photons weird quantum tricks that would work on bench scale, where it was literally the same photon at the receiver as when you transmitted, will not work, but any conventional signalling within quite broad limits is OK. Researchers at the University where I studied as an undergraduate developed EDFA.
Fun fact: Pirelli, the tire company, used to be big in submarine cable repeaters and related products. The ones I saw at telecom shows were painted Pirelli yellow. That part of Pirelli was sold to private equity.
Readit News
People really don't get the enormity of the difference - when there were policy debates in my country about rolling our new fixed line infrastructure there were literally people saying "but won't all homes and businesses just be able to use wireless in the future?"
1 - https://en.wikipedia.org/wiki/MAREA
Given it's a larger market, I would have thought there would be more direct runs landing on the US coast instead of an 'intermediary' point in Canada.
Deleted Comment
The tubes were tested to an extremely high standard. Only a small fraction of the manufactured tubes were selected after testing: Bell Labs designed a test regime over 18 years to detect minute flaws in manufactured tubes
The cable and its 306 tubes operated for 22 years with no failures.
Power delivery to the laser diodes is done through the metal jacket of the cable. The whole submarine cable is essentially a very long DC transmission line. Which is a fascinating topic in itself, E.g. What is ground in such a line, it will differ by 1000s of Volts between continents.
This trick also means the cable doesn't care about the rest of the technology. If it was a retransmitter then we'd need to replace the entire cable if we change from 100Gbps over Protocol #39 to 200 Gbps over Protocol #40 because every retransmitter needs to be equipped for the new protocol, but the optical amplifier doesn't care why these photons turned up, what they mean - when provided with power it just ensures proportionately more photons like them come out of the amplifier.
Because they're not actually the same photons weird quantum tricks that would work on bench scale, where it was literally the same photon at the receiver as when you transmitted, will not work, but any conventional signalling within quite broad limits is OK. Researchers at the University where I studied as an undergraduate developed EDFA.
I wonder why DC though. Is AC lossy when surrounded by salt water?
Does that translate to free energy for the repeaters?
Deleted Comment
Dead Comment