Alphabet's Project Taara Laser Tech Beamed 700TB of Data Across Nearly 5km (theverge.com) 34
An anonymous reader shares a report: In January, Google's parent company, Alphabet, shut down Project Loon, an initiative exploring using stratospheric helium balloons to distribute wireless internet (an attempt to use solar-powered drones folded in 2017). However, some technology developed as a part of the Loon project remained in development, specifically the Free Space Optical Communications (FSOC) links that were originally meant to connect the high flying balloons -- and now that technology is actively in use providing a high-speed broadband link for people in Africa.
Sort of like fiber optic cables without the cable, FSOC can create a 20Gbps+ broadband link from two points that have a clear line of sight, and Alphabet's moonshot lab X has built up Project Taara to give it a shot. They started by setting up links in India a few years ago as well as a few pilots in Kenya, and today X revealed what it has achieved by using its wireless optical link to connect service across the Congo River from Brazzaville in the Republic of Congo and Kinshasa in the Democratic Republic of Congo. In 20 days, Project Taara lead Baris Erkmen says the link transmitted nearly 700TB of data, augmenting fiber connections used by local telecom partner Econet and its subsidiaries.
Sort of like fiber optic cables without the cable, FSOC can create a 20Gbps+ broadband link from two points that have a clear line of sight, and Alphabet's moonshot lab X has built up Project Taara to give it a shot. They started by setting up links in India a few years ago as well as a few pilots in Kenya, and today X revealed what it has achieved by using its wireless optical link to connect service across the Congo River from Brazzaville in the Republic of Congo and Kinshasa in the Democratic Republic of Congo. In 20 days, Project Taara lead Baris Erkmen says the link transmitted nearly 700TB of data, augmenting fiber connections used by local telecom partner Econet and its subsidiaries.
clear line of sight? so wind / rain / dust / snow (Score:3)
clear line of sight? so wind / rain / dust / snow / etc can block it? and in an big rain you can have rain fade outages?
Re: clear line of sight? so wind / rain / dust / s (Score:5, Funny)
If you use a powerful enough laser, those things don't matter.
Pity anyone who gets in the way, though.
Re: (Score:2)
You jest, but FTA: They credit its resilience in the face of haze, light rain, birds, and other obstacles to the ability to adjust laser power on the fly, as well as improved pointing and tracking.
Now IDK how much you have to increase the laser power to punch through birds ...
Re: clear line of sight? so wind / rain / dust / s (Score:4, Funny)
You jest, but FTA: They credit its resilience in the face of haze, light rain, birds, and other obstacles to the ability to adjust laser power on the fly, as well as improved pointing and tracking.
Now IDK how much you have to increase the laser power to punch through birds ...
They must have an auto chicken nugget setting.
Re: (Score:3)
if a bird with smaller than your beam width or if birds passing through briefly cause fewer dropped bits than you can correct, then it seems like it would work fine under these circumstances.
Re: (Score:3)
Laser beams need not be tiny. They just have to be collimated. In theory, you could have a laser beam that's three feet wide, which would be almost entirely immune to destructive interference from wildlife even without using enough power to set the bird on fire.
Or you could just crank the power up and have drones that pick up the carcasses and bring them directly to Tyson for processing or whatever. Either way. But the concern there would be airplanes. :-D
Re: (Score:2)
If you use a powerful enough laser, those things don't matter.
Pity anyone who gets in the way, though.
They are keeping their specs secret, but If you zoom in in the mast you can read the sticker:
“Do not look into laser with remaining cranial cavity”
Re: (Score:3)
clear line of sight? so wind / rain / dust / snow / etc can block it? and in an big rain you can have rain fade outages?
Well the system operates in two modes: "Line of Sight" mode and "Make Line of Sight" mode. The algorithm is pretty simple: Upon failure to transmit a packet, increase power and try again.
Forget Data (Score:2)
Forget data transfer. Can you use it to cook food or even just boil water?
Re: (Score:2)
Evidently you can use it to skewer birds, at least.
Re: clear line of sight? so wind / rain / dust / s (Score:4, Funny)
Upon failure to transmit a packet, increase power and try again.
So Time-to-Live is a double entendre in this system?
Re: (Score:2)
If only the Earth was flat!
In what time FFS ? USE USEFUL UNITS (Score:1)
I drank 50 pints of beer.
In a year.
Journalism 101
Re: (Score:2)
RTFS. 700TB in 20 days.
Re:In what time FFS ? USE USEFUL UNITS (Score:5, Informative)
20 days apparently.
Which makes it an average of, er... 400 mbit/s.
I mean that's OK. But you can already buy COTS point to point microwave links that do somewhat better over rather longer links in the rain and fog. They do claim 20gbps, but you can already buy 10gpbs microwave links off the shelf today which have superior range and weather capabilities not to mention the lack of fancy motorised hardware. You can also get a lot converging on one point, by use of directional antennas and choosing the frequency band.
The laser one sounds useful for a balloon where you can massively save on antenna size and therefore weight compared to microwave links, but on the ground it sounds like google bragging about a complex solution to a problem that's already solved better.
Where this would really shine (Score:3)
(geddit?) is for indoor applications.
Think warehouse robots or other mobile equipment in an indoor setting.
There are few outdoor applications I can think of that can't be solved adequately using optical fiber or microwave links. A standard objection is that the latter two either cost money or require regulatory approval. And the standard answer is that fsoc also costs money and depending on jurisdiction may also require regulatory approval. If you're spending money on a facility anyway, you're likely going through permitting and construction costs already.
Re: (Score:2)
Think warehouse robots or other mobile equipment in an indoor setting.
Pro tip .. photons like to travel in straight lines, and lasers produce photons all headed in a single direction.
There are plenty of wireless radio links for indoor (and even outdoor) equipment that don't require the line of sight that a laser solution requires.
And if you wanted to mount this system on a mobile device, you are then going to have to add mechanical devices for aiming both the transmitter and receiver - which adds a ridiculous amount of unnecessary complexity and cost to your setup.
Re: Where this would really shine (Score:2)
Pro tip: rtfa. Pointing a link across a river also requires mechanical gimbaling of both the receive and transmit apertures on both sides. Wind loading, thermal cycling, and vibrations from small seismic events and passing vehicles matter in this sort of application.
Your point about this stuff competing with indoor wireless is very well taken, and it's unlikely this is applicable for roaming equipment. But stuff that needs to be moved a few times a day, for example an inspection cart or some sort of calibra
Re: (Score:2)
Pro tip: rtfa. Pointing a link across a river also requires mechanical gimbaling of both the receive and transmit apertures on both sides. Wind loading, thermal cycling, and vibrations from small seismic events and passing vehicles matter in this sort of application.
There's a huge difference between gimbaling a system to point in basically the same direction all the time, and a system that would need to point in between Pi and 2Pi steradians in order to target mobile equipment.
0ld tech? (Score:2)
Re: (Score:2)
Oh, very old [wikipedia.org] technology.
So this is an improvement (Score:2)
over, for instance, Canobeam [canon.com].
Which apparently not for sale any more, is a huge improvement...
Re: (Score:2)
over, for instance, Canobeam [canon.com].
Which apparently not for sale any more, is a huge improvement...
Aside from it being from Google. This new system has at least 16 times the bandwidth over 5 times the distance of your no-longer-for-sale Canon system
Re: So this is an improvement (Score:2)
Well it's not 'my' Canobeam, but yeah, I installed a few. And it was pretty useful in the right application. More, faster, farther is always good.
Re: More, faster, further... (Score:1)
I think this is faster than what folks at MIT [mit.edu] were doing a dozen years ago over similar distance. NASA might have a thing [nasa.gov] or two [nasa.gov] to say about distance, though.
Re: (Score:2)
Fuckers stealing my idea! (Score:1)
and today X revealed what it has achieved by using its wireless optical link to connect service across the Congo River from Brazzaville in the Republic of Congo and Kinshasa in the Democratic Republic of Congo.
Back in prehistory (1991) I was using a GSM modem in Kinshasa talking to a cell tower in Brazzaville to get a UUCP connection to our offices in Paris France.
Kids of today with their fancy laser shit,
(This, by the way, is a true story. I was in Kinshasa for the start of Gulf War 1, watching it on CNN in the offices of a colonel in Mobutu's secret police with Miko Rwayitare).
It's been done, years ago.. (Score:1)
So, about 77 hours of usable bandwidth in 20 days? (Score:1)