NATO Plans To Build Satellite Links As Backups To Undersea Cables (tomshardware.com) 65
Tom's Hardware reports that NATO is developing an advanced system to address the growing number of undersea cable disruptions observed in recent years. Known as HEIST (Hybrid Space-Submarine Architecture Ensuring Infosec of Telecommunications), the project is designed to significantly enhance the resilience of undersea communication networks. HEIST will enable damage detection with an accuracy of one meter, facilitate rapid data rerouting through satellite networks when disruptions occur, and establish open-source protocols to foster global collaboration. From the report: Satellites are the primary backups to undersea cables, but their bandwidth is far behind physical connections. For example, Google's latest fiber-optic lines can hit 340 terabits per second. In contrast, the frequency used by most satellites -- 12 to 18GHz -- can only handle about 5 gigabits per second or about 0.0015% of the maximum throughput of Google's fiber connection.
Work is underway to upgrade satellites from radio transmissions to lasers, increasing the speed by about 40 times to 200 Gbps. Starlink already uses this technology to communicate between its satellites, while Amazon is also developing it for its own Project Kuiper. However, it still faces challenges, like poor visibility and targeting precision between the satellite and ground station.
Because this is a major NATO project, the alliance plans to open-source part of the process. Making it public would allow anyone interested to find holes and make many iterations. Gregory Falco, the NATO Country Director for HEIST, believes that this is the fastest way for the project to achieve its goals and help prevent any catastrophic loss of data transmission in case of deliberate attacks against these underwater infrastructures in international waters.
Work is underway to upgrade satellites from radio transmissions to lasers, increasing the speed by about 40 times to 200 Gbps. Starlink already uses this technology to communicate between its satellites, while Amazon is also developing it for its own Project Kuiper. However, it still faces challenges, like poor visibility and targeting precision between the satellite and ground station.
Because this is a major NATO project, the alliance plans to open-source part of the process. Making it public would allow anyone interested to find holes and make many iterations. Gregory Falco, the NATO Country Director for HEIST, believes that this is the fastest way for the project to achieve its goals and help prevent any catastrophic loss of data transmission in case of deliberate attacks against these underwater infrastructures in international waters.
Latency, not bandwidth, is the problem (Score:1)
Ignoring that "HEIST" suggests this post is 3 months too early...
The problem with satcom is latency more than bandwidth. Lasers and radio are different in that you can put more data into a laser datastream than you can a radio datastream... but if you're talking goesync orbit than it's the same latency. Too much for gaming, live video, audio, or anything other than web surfing and one-way downloads. Still, even TCP ACKs take a long time reducing usable data transfer rates. Sats won't fix any of that...
Re: (Score:2)
Indeed, about the only useful increase in measuring where the break is would be actual coordinates, as in if a ship is dragging its anchor, you might know the distance to each break, but not where the cables have been moved to.
That said, I don't think we need to worry about power. Underwater power lines are underwater power lines, and communications is communications. Most communications lines only carry power to help power the very equipment used to do things like boost the signals.
My biggest problem wit
Re: (Score:3)
Backup circuit makes latency moot (Score:1)
Draw a triangle, It has two legs and a base. It is the definition of a triangle that the total length of both legs will always exceed the length of the base.
The legs are 1. Ground station up to LEO sat and 2. LEO sat to other ground station. The base is the fiber-optic distance between the two ground stations.
Now with LEO satellites that triangle is actually a trapezoid because we need to add the sat to sat distance as well
Starlink orbit are around 550km, so that trapezoid is 550+inter-sat-distance+550km
Re:Latency, not bandwidth, is the problem (Score:4, Informative)
Latency via LEO satellites is lower than via fiber cable. Light travels faster in vacuum than in glass.
Well somewhat yes, but context is needed.
Let's do a simple calculation, assuming we want to send a signal over a distance of half the circumference of the earth. Ignoring electronics delays, congestion, and the short hop from ground to sat and back again, we get that a signal takes a time equal to half the circumference of the divided by the speed of light to travel that distance: 20,000,000 m / 300,000,000 m/s = 0.0667 s, or 66.7 ms. If that distance were glass fiber-optic cable, then the speed is slower by a factor equal to the refractive index of the cable material. That varies depending on the type of glass, but 1.5 is typical. That would mean the signal would take 66.7 * 1.5 = 100 ms to travel the same distance in a fiber-optic cable.
So, you add a latency of about 33 ms using a terrestrial fiber-optic cable over that distance. But that doesn't consider other factors, such as electronics delays (which affect both technologies) and network congestion (which can be more severe for satellites because of lower bandwidth.)
Altogether, the latency gain you get from satellites by not using a high-refractive-index medium is not that big a deal when you consider other factors.
Re: Latency, not bandwidth, is the problem (Score:1)
ironically, you could save 33% off the latency by drilling a straight line. at least for the longest routes, the gains of making a hole between, say, Tokyo and san Francisco are a bit less.
Re: (Score:2)
Get back to me with the cost of drilling such a hole and protecting a cable running through it, and then we'll talk. ;-D
Re: Latency, not bandwidth, is the problem (Score:2)
Re: (Score:2)
The problem with satcom is latency more than bandwidth. Lasers and radio are different in that you can put more data into a laser datastream than you can a radio datastream... but if you're talking goesync orbit than it's the same latency. Too much for gaming, live video, audio, or anything other than web surfing and one-way downloads. Still, even TCP ACKs take a long time reducing usable data transfer rates. Sats won't fix any of that... and even StarLink isn't perfect.
Relationship between bandwidth and latency is a function of congestion window size. There are PEPs and various hacks to conceal these kinds of problems.
Article 5 (Score:1)
Be strong NATO and invoke Article 5 is a country (ie Russia) damages any more infrastructure...Currently you are ignoring everything and all damage and hoping Pootini does not do anything else but he does so you ignore it again hoping Pootini does nothing more but he does.....
Re: Article 5 (Score:5, Insightful)
Re: (Score:2)
Perhaps the West needs to go after Vlad himself...
Re: Article 5 (Score:4, Funny)
Re: (Score:1)
Re: (Score:2)
Re: (Score:2)
This is ridiculous. Trump and Putin would never fight. In the first place, Trump's a cowardly bitch. Then, they're somewhat friends. Finally, Trump is subservient to Putin, so if Putin said, "Lay down," Trump would.
Wouldn't this rather be combat by proxy... Some washed up 80's pro-wrestler for Trump and well, probably the same for Putin but more Slavic.
Re: (Score:2)
Dolf and Stallone can do a reprise.
Re: Article 5 (Score:2)
Re: (Score:2)
A lot of people write comments (here and elsewhere) about just "taking out" Putin. What would this really change? Who is going to run Russia after such an event? How would their policy change? (Especially if it were known to be due to an assassination via a foreign power). I suspect the end result would be a worse situation than the one that exists now. Wouldn't this just invite Russia to start taking out Western leaders? They likely have the means to do an awful lot of damage if the gloves come off.
I am no
Re: (Score:2)
A lot of people write comments (here and elsewhere) about just "taking out" Putin. What would this really change? Who is going to run Russia after such an event? How would their policy change? (Especially if it were known to be due to an assassination via a foreign power). I suspect the end result would be a worse situation than the one that exists now. Wouldn't this just invite Russia to start taking out Western leaders? They likely have the means to do an awful lot of damage if the gloves come off.
I am not defending Putin here, just pointing out that just killing him wouldn't likely solve very much - and quite possibly make things worse.
We'd probably do like we keep doing in the middle east. Prop up some new dictator who will become the ultimate evil when he breaks his leash in a decade or so and starts going his own way. Why not? It works well for creating very profitable wars in the middle east. Why not throw that idea into action elsewhere?
Re: Article 5 (Score:2)
Re: (Score:2)
even babies do not want to be Russian anymore
I realize that you probably didn't mean to be funny, but I'm still LMAO. Thanks for the laugh, and Happy New Year!
Re:Article 5 (Score:5, Informative)
Be strong NATO and invoke Article 5
We're not gonna start WW3 over a cable.
A proportionate response is to do what Finland did last week: Seize and impound the ship caught dragging its anchor.
Re:Article 5 (Score:4, Interesting)
Re: (Score:2)
It's an extremely bad idea to try to deny one country the use of the seas. Lots of countries would love to do that to grab more territory and disrupt neighbours. Global trade would be badly affected, prices would shoot up.
Re: (Score:2)
We're not gonna start WW3 over a cable.
A proportionate response is to do what Finland did last week: Seize and impound the ship caught dragging its anchor.
The mere invocation of article 5 is separate from any response that might be agreed to as a result.
It is perfectly possible to invoke article 5 and everyone agrees to do nothing, issue a sternly worded warning, carry out similar acts of sabotage or push rapid dragons out of cargo doors.
Re: (Score:3)
Invoking article 5 gives you a meeting with the NATO members that can then decide if any action should be taken and if so what. That action is likely going to be proportional. So, they would maybe take out a cable of Russia or apply some sanction. Pretty much what is happening now, just without the article 5 meeting.
Actual war is only going to happen if one of the sides actually wants a war. Which seems pretty unlikely now.
More money to Musk and Bezos? (Score:2)
Let me guess. SpaceX and Blue Origin are gonna get all the contracts.
Re: More money to Musk and Bezos? (Score:3)
Re: (Score:2)
Let me guess. SpaceX and Blue Origin are gonna get all the contracts.
What else is there? Arianne is expensive and Boeing is bust.
There are other non-US launch agencies around (besides Ariane.) Of course, they are at various stages of viability and lift capacity.
oops (Score:3)
A Russian built weather satellite just spun out of control and "accidentally" crashed into the NATO satellite.
Re: oops (Score:2)
Re: (Score:3)
I think you're dramatically overestimating just how much satellites have control over their orbital dynamics.
War: Forget about every oversea online service. (Score:5, Insightful)
Because when the next war heats-up, you will be lucky if your text only e-mail manages to reach another continent.
All those undersea cables AND all those big communications satellites will be gone.
The mini satellites the likes of Starlink may survive longer due to their small size and numbers.
Hence, no more YT or other video site.
And say goodbye to your rented games a.k.a. Steam "purchases" if you have to replace/upgrade your PC.
Re: War: Forget about every oversea online service (Score:4, Funny)
That is how I grew up.
Re: War: Forget about every oversea online servic (Score:2)
Re: (Score:2)
Luxury! We had to send smoke signals by burning our toys.
Re: (Score:2)
You and I will be fine.
The sweet summer children with their online addiction? Not so much.
Today's companies and business, with their ever increasing reliance on online services? Even less.
Re: (Score:2)
The sweet summer children with their online addiction? Not so much. Today's companies and business, with their ever increasing reliance on online services? Even less.
One can only hope.
Re: (Score:2)
No YouTube, no steam games, no oversea emails... the horror. The horror!
...
That is how I grew up.
Erm... things like Steam, Youtube, et al. are more likely to survive a global communications crisis because their CDNs are distributed. I don't download steam games from the US, I have a variety of sources in the UK to choose from, if not the EU. The US will have Canadian and Latin/South American sources as well as US based ones that will be accessible. Keeping them sync'd globally will be an issue but it wont be "oh noes, teh fall of the intertubes".
It makes zero sense to stream from the EU if you're in
Re: (Score:2)
AND - No PornHub - the horror! :-)
Re: (Score:3)
Obviously straight "I need to talk to this server in London and this one in Berlin from this one in New York and this one in LA" scenarios would go...poorly...in the event of widespread cable cuts and the allocation of relatively paltry satellite links to military purposes and pays-more-than-you-for-their-link enterprise customers; but my impression is that, outside of some really underdeveloped areas, the bulk of traffic that is
Re: (Score:2)
No BS video/audio sites, no BS games, no BS overseas calls? All you need to do is ensure it'll take down facebook and twitter and I'll start the damn war myself.
Sort of a tangent, but (Score:3)
What's with these so-called acronyms where the letters don't actually line up the words? It's not the daily Scramble puzzle! It drives me nuts...
It's like when an imbecile from UW CSE came up with a project he insisted on calling "FRITTER" even though the closest he could do with the actual words was RFITTER.
Re: (Score:2)
It's not scrambled. There's no standard for which parts of an acronym becomes capitalised beyond being the first letters of words.
In this case it's Hybrid space-submarine architecture Ensuring InfoSec of Telecommunications. With Infosec being a short for of Information Security hence making sense to include the S.
Re: (Score:2)
(Go look up "scheisse".)
Plans to? (Score:3)
Re: (Score:1)
The technology didn't exist back then.
It doesn't really exist now. This is just a backup for critical military and government stuff, the bandwidth is too low for general internet traffic.
Re: (Score:2)
The technology didn't exist back then.
It doesn't really exist now. This is just a backup for critical military and government stuff, the bandwidth is too low for general internet traffic.
Decades ago the backup was radio relays and the old dispatch system (fast navy ships that carried physical letters)... I suspect they still are, which is why the US and UK maintain such a huge number of overseas military bases but having satellite is a good move forward.
Hmm, wouldn't it be cheaper... (Score:2)
... to just lay so many undersea cables that it becomes infeasible to damage the network in a meaningful matter.
Satellites are, after all, very expensive when you want LEO (for latency) and high bandwidths. You'd need lots of satellites, probably more than Starlink has.
Also while satellites can provide communications you still want to have other lines like power lines.
Re: (Score:3)
Given how much costs an undersea cable versus how cheap it is to cut them (a boat + anchor), I can't see how would that be possible.
Unnecessary and probably unwise (Score:2)
How long until we see a Starlink contract? (Score:2)
NATO project - price? timeline? (Score:2)
Consider what Starlink is doing. On a commercial budget, they are putting up thousands of satellites in LEO, which can provide pretty decent throughput for individual connections over a huge area.
NATO wants a backup for undersea cables. They don't need to cover the earth, they need to provide basically point-to-point connections. They also want low latency, so geosynchronous satellites won't cut it. So: loft an entire constellation of satellites, just for this? That doesn't make a lot of sense.
If they wan
Very Long Lines (Score:1)
Don't forget the friggin' sharks (Score:2)
Buried terrestrial fiber (Score:2)
Make the permitting friendly and run more fiber over land. Satellites are vulnerable and will be targeted in open warfare, conventions or treaties notwithstanding (and they won't.)
If only (Score:2)
Re: (Score:2)
Capacity issues to replace wired connections.
Fixing symptoms not the root cause (Score:1)