I work in the subsea optical transport space, and frustratingly the author falls victim to the same false notion that sharks have ever 'attacked' cables. Don't get me wrong, it's silly, but it...
I work in the subsea optical transport space, and frustratingly the author falls victim to the same false notion that sharks have ever 'attacked' cables. Don't get me wrong, it's silly, but it shows a lack of attention to detail on the subject (one of the top comments on the embedded video debunks this immediately)--one which is genuinely fascinating given how little many people know about just how information is physically exchanged around the world. It's a dumb thing to get worked up over, I'll concede, but if the author is going to educate people on the matter they could at least fact check this bizarre claim. Cables are armored for a myriad of reasons outside of shark bites. The rest of the piece is interesting enough, and I highly recommend folks look into the coherent optical systems that transmit data over these cables (and all long haul terrestrial cables) if they're more curious about how the optical layer of the world's networks really function.
I used to work in a NOC and learned a lot about OTN/fiber. There's one question about undersea fiber that I never learned the answer to, maybe you can help me out. So light levels need to be...
I used to work in a NOC and learned a lot about OTN/fiber. There's one question about undersea fiber that I never learned the answer to, maybe you can help me out.
So light levels need to be amplified after X miles, which is where erbium-doped fiber amplifiers (EDFA) come into play and make a lot of sense for undersea fiber. However, amplifying signal will also amplify the noise, which means after 9(?) EDFAs, the signal itself will need to be regenerated. What is the solution for light regeneration on an undersea fiber bundle? Or am I mistaken in the EDFAs also amplifying noise?
Amplifiers amplify noise, but the signal to noise ratio is worse if you take longer to amp it. The amp increases the power of the signal AND the noise, so if you amplify every 10km, you'll be...
Amplifiers amplify noise, but the signal to noise ratio is worse if you take longer to amp it. The amp increases the power of the signal AND the noise, so if you amplify every 10km, you'll be amplifying mostly signal, with very little noise. Compared to every 100km where the signal amplitude will be much closer to the noise amplitude, so the resulting amplified signal has a lot more noise in it.
Also, coherent transmission helps a lot with signal integrity. And RAMAN amplifiers instead of EDFAs give you really great distances.
I'll take a stab at it, sure! So the amplifiers used in subsea cables are called repeaters, which is a bit of a misnomer as they don't regenerate the signal, but what they do (besides power...
I'll take a stab at it, sure! So the amplifiers used in subsea cables are called repeaters, which is a bit of a misnomer as they don't regenerate the signal, but what they do (besides power monitoring as these systems are far more sensitive than terrestrial) is match the total output power along the cable at regular intervals and along each individual fiber pair (as that can vary from pair to pair depending on potential use of different modulation schemes/transmission modes to support different client types), which allows for very minimal noise gain due to tighter component tolerance and better power monitoring than what is used in standard EDFAs terrestrially.
Additionally, relative to the signal to noise ratio, what is often done for subsea is that for a coherent transponder that might be capable of running at 1.6Tbps, it will be tuned down to maybe 1.4T or 1.3T to allow FEC to better overcome the total noise on the system. This is done in terrestrial applications too for long haul but is extremely prevalent for long subsea reaches.
Raman amplification is only typically used on short subsea links that are closer to terrestrial deployments than the subsea links I'm talking about.
I work in the subsea optical transport space, and frustratingly the author falls victim to the same false notion that sharks have ever 'attacked' cables. Don't get me wrong, it's silly, but it shows a lack of attention to detail on the subject (one of the top comments on the embedded video debunks this immediately)--one which is genuinely fascinating given how little many people know about just how information is physically exchanged around the world. It's a dumb thing to get worked up over, I'll concede, but if the author is going to educate people on the matter they could at least fact check this bizarre claim. Cables are armored for a myriad of reasons outside of shark bites. The rest of the piece is interesting enough, and I highly recommend folks look into the coherent optical systems that transmit data over these cables (and all long haul terrestrial cables) if they're more curious about how the optical layer of the world's networks really function.
I used to work in a NOC and learned a lot about OTN/fiber. There's one question about undersea fiber that I never learned the answer to, maybe you can help me out.
So light levels need to be amplified after X miles, which is where erbium-doped fiber amplifiers (EDFA) come into play and make a lot of sense for undersea fiber. However, amplifying signal will also amplify the noise, which means after 9(?) EDFAs, the signal itself will need to be regenerated. What is the solution for light regeneration on an undersea fiber bundle? Or am I mistaken in the EDFAs also amplifying noise?
Amplifiers amplify noise, but the signal to noise ratio is worse if you take longer to amp it. The amp increases the power of the signal AND the noise, so if you amplify every 10km, you'll be amplifying mostly signal, with very little noise. Compared to every 100km where the signal amplitude will be much closer to the noise amplitude, so the resulting amplified signal has a lot more noise in it.
Also, coherent transmission helps a lot with signal integrity. And RAMAN amplifiers instead of EDFAs give you really great distances.
I'll take a stab at it, sure! So the amplifiers used in subsea cables are called repeaters, which is a bit of a misnomer as they don't regenerate the signal, but what they do (besides power monitoring as these systems are far more sensitive than terrestrial) is match the total output power along the cable at regular intervals and along each individual fiber pair (as that can vary from pair to pair depending on potential use of different modulation schemes/transmission modes to support different client types), which allows for very minimal noise gain due to tighter component tolerance and better power monitoring than what is used in standard EDFAs terrestrially.
Additionally, relative to the signal to noise ratio, what is often done for subsea is that for a coherent transponder that might be capable of running at 1.6Tbps, it will be tuned down to maybe 1.4T or 1.3T to allow FEC to better overcome the total noise on the system. This is done in terrestrial applications too for long haul but is extremely prevalent for long subsea reaches.
Raman amplification is only typically used on short subsea links that are closer to terrestrial deployments than the subsea links I'm talking about.
Next time I have any issues connecting to the internet, I'm blaming sharks.
As we all do!