Forgot your password?
typodupeerror
Network Communications Networking The Military Wireless Networking Technology

DARPA Begins Work On 100Gbps Wireless Tech With 120-mile Range 83

Posted by samzenpus
from the greased-lightning dept.
MrSeb writes "DARPA has begun development of a wireless communications link that is capable of 100 gigabits per second over a range of 200 kilometers (124mi). Officially dubbed '100 Gb/s RF Backbone' (or 100G for short), the program will provide the U.S. military with networks that are around 50 times faster than its current wireless links. In essence, DARPA wants to give deployed soldiers the same kind of connectivity as a high-bandwidth, low-latency fiber-optic network. In the case of Afghanistan, for example, the U.S. might have a high-speed fiber link to Turkey — but the remaining 1,000 miles to Afghanistan most likely consists of low-bandwidth, high-latency links. It's difficult (and potentially insecure) to control UAVs or send/receive intelligence over these networks, and so the U.S. military instead builds its own wireless network using Common Data Link. CDL maxes out at around 250Mbps, so 100Gbps would be quite a speed boost. DARPA clearly states that the 100G program is for US military use — but it's hard to ignore the repercussions it might have on commercial networks, too. 100Gbps wireless backhaul links between cell towers, rather than costly and cumbersome fiber links, would make it much easier and cheaper to roll out additional mobile coverage. Likewise, 100Gbps wireless links might be the ideal way to provide backhaul links to rural communities that are still stuck with dial-up internet access. Who knows, we might even one day have 100Gbps wireless links to our ISP."
This discussion has been archived. No new comments can be posted.

DARPA Begins Work On 100Gbps Wireless Tech With 120-mile Range

Comments Filter:
  • by Anonymous Coward

    With this kind of bandwidth, fleets of tele-operated ground vehicles will become reality. Today there isn't enough bandwidth today to send back video, location, and other sensor info to intelligently navigate more than a vehicle or two. This will save many lives. Bravo DARPA!

    • Today there isn't enough bandwidth today to send back video, location, and other sensor info to intelligently navigate more than a vehicle or two.

      Really? You mean, nodes can't act as repeaters for one another?

      What a silly notion.

  • I am guessing that this only works because a huge amount of radio spectrum bandwidth is allocated per user. There probably is no actual method of scaling this up for general-purpose usage. The last line of the OP seems beyond speculative.

    • by afidel (530433)

      To achieve that kind of bandwidth the signal is probably going to have to be very narrow and so you can achieve greater throughput using spacial distribution as well and channel distribution. Some of the higher end WiFi systems already do this kind of bandwidth improvement by using beam steering technology to logically switch between multiple clients, the same kind of technology is also used in satellite systems where the same frequencies are reused many times between the ground and transponder.

      • by Gibgezr (2025238)

        Spacial distribution, like "channel hopping" to avoid interference? That helps with moving transmitters/receivers, but not within a static local area. In the end, for any given area covered by a transmitter, the frequency availability will be the hard cap on the shared bandwidth for that area. If 10k people in a given coverage area all want to download large files at 100 Gbps, all the trickery in the world won't increase that cap.

        Somewhere, someone must have a simple rule-of-thumb for this sucker, like how

        • by afidel (530433)

          No spacial distribution like using beams 1 degree wide to turn a 100Gbps per 40MHz technology into a 36,000Gbps per 40MHz technology over the area covered by a synthetic array. Of course it doesn't quite scale like that because you have to spread out what part of the 40MHz your clients use so that clients in close proximity aren't using the same frequencies at the same time, but it's fairly amazing just how much bandwidth you can fit into an area using beam steering versus using a simple monopole. Currently

    • by jonadab (583620)
      Actually, I'm guessing it could work for commercial usage if all the links were site-to-site links achieved with some kind of directional antennae (perhaps using conic-section backplates), so that they mostly don't pollute the airwaves all around them very much. In other words, an ISP (or any medium-to-large company) could set up a directional antenna in $smalltown and aim it at their other directional antenna in $wellconnectedlocation, creating a high-speed link between the two sites, without running fibe
  • 120 mile range? (Score:4, Insightful)

    by rossdee (243626) on Monday December 17, 2012 @01:21PM (#42315193)

    How? Is it airborne or something? You are not going to get any straight line reception at that range due to curvature of the earth, even in the plains.

    • Why, they dont need to have straight line of sight from any single drone because they plan to blot the sky out with drones so they have floating network of wireless ad-hoc access points!
      • Why, they dont need to have straight line of sight from any single drone because they plan to blot the sky out with drones so they have floating network of wireless ad-hoc access points!

        ...Then we will browse in the shade?

        /ducks

    • How? Is it airborne or something? You are not going to get any straight line reception at that range due to curvature of the earth, even in the plains.

      It's assumed that the antenna would be mounted on something like, I don't know, say a tower, or building...

      • by CaptainNerdCave (982411) on Monday December 17, 2012 @02:53PM (#42316143)

        The problem that was already addressed is the curving of earth, because it can be overcome with height. Let's sustain that increasing the altitude of your dishes will allow greater distance without the sphere's shape interfering, you still have all of the factors associated with those heights: weather, cost of getting there, service, general maintenance.

        Maintenance: How easy is it to remove ice? Snow? What about the cost of maintaining the tower?

        Service: What do you do when you can't communicate with the unit, and you've ruled out everything except the cable between the unit and it's nearest point of contact?

        Cost: This is a broader issue than maintenance, because it allows for not owning the tower/building. Tower space is premium, building roof-tops are premium, labor to install, service, or repair is EXTRA premium. Not only do you need guys willing to climb 200+ feet, but they need to be technically capable. http://www.midweststeeplejacks.com/ [midweststeeplejacks.com] charges no less than $250/hr.

        Weather: Why don't you see point-to-point connections on towers that are 200ft up on towers? Because the bandwidth requires very high frequencies, and those frequencies are very susceptible to any movement caused by wind. I've seen a gentle breeze (on the ground) turn a wireless link from -45 dbi to -60. Let's not forget rain and snow.

        The only good ways to mount an antenna or dish at a height, and ensure reliability, are with a very large antenna (think something with 3 or 4 legs and covering at least 400 feet^2), or a building.

        • by jimbouse (2425428)

          The problem that was already addressed is the curving of earth, because it can be overcome with height. Let's sustain that increasing the altitude of your dishes will allow greater distance without the sphere's shape interfering, you still have all of the factors associated with those heights: weather, cost of getting there, service, general maintenance.

          Maintenance: How easy is it to remove ice? Snow? What about the cost of maintaining the tower?

          Service: What do you do when you can't communicate with the unit, and you've ruled out everything except the cable between the unit and it's nearest point of contact?

          Cost: This is a broader issue than maintenance, because it allows for not owning the tower/building. Tower space is premium, building roof-tops are premium, labor to install, service, or repair is EXTRA premium. Not only do you need guys willing to climb 200+ feet, but they need to be technically capable. http://www.midweststeeplejacks.com/ [midweststeeplejacks.com] charges no less than $250/hr.

          Weather: Why don't you see point-to-point connections on towers that are 200ft up on towers? Because the bandwidth requires very high frequencies, and those frequencies are very susceptible to any movement caused by wind. I've seen a gentle breeze (on the ground) turn a wireless link from -45 dbi to -60. Let's not forget rain and snow.

          The only good ways to mount an antenna or dish at a height, and ensure reliability, are with a very large antenna (think something with 3 or 4 legs and covering at least 400 feet^2), or a building.

          You sound like someone who has never looked at a communication tower, much less installed and used equipment on one.

          I run a WISP and have equipment dangling hundreds of feet in the air. With proper planning, amazing results can be achieved. Weather is a factor over 6Ghz but, once again, this is not a problem with proper planning.

          • Actually, I work for a WISP. I've done my share of climbing 60-100ft antennae to install or repair the equipment we've got up there, and I've experienced the tower sway at that low height. There are a few other locations where there are things mounted around 250 or so, but I haven't tended anything that high up on a flimsy structure. Lots of things are hanging out on top of buildings that are 120-300ft, but most of those installations have enough structure to remain pretty rigid.

            I agree that proper plann

      • by jonadab (583620)
        I think to achieve a 120-mile range, you're going to need towers at both ends. A quick back-of-the-envelope calculation suggests that if one endpoint is on the ground, the other would have to be elevated about three miles, at which point fiscally speaking you may just about as well put it in low earth orbit.

        However, I suspect the "120 mile" figure is probably what they figure the equipment would support GIVEN line of sight as a base assumption. In practice, dealing with topography and whatnot, you're prob
    • I had already heard of airborne wireless base stations being deployed for the U.S. military (possibly reported here a few years back?), so this actually doesn't sound as crazy as you think. As I recall, they have some blimps or zeppelins geared up to do this sort of thing. If not, however, who says you need line of sight? I mostly stick to software, so I'm about as far from an expert as you can get when it comes to radios, but even I know that AM radios can be picked up at these distances, particularly at n

      • by mcgrew (92797) *

        I mostly stick to software, so I'm about as far from an expert as you can get when it comes to radios, but even I know that AM radios can be picked up at these distances

        It depends on the frequencies. Short wave and AM radio can be heard around the world because the waves bounce off the stratosphere. The term for this is ":atmospheric skip". Other frequencies, such as TV and FM radio, are limited to line of sight, with the earth's curvature being the limiting factor -- the waves don't skip, they keep going i

    • Re:120 mile range? (Score:4, Informative)

      by Eevee (535658) on Monday December 17, 2012 @03:00PM (#42316235)
      Well, if Slashdot ever bothered linking to the original article [darpa.mil], you'd see:

      The goal is to create a 100 Gb/s data link that achieves a range greater than 200 kilometers between airborne assets and a range greater than 100 kilometers between an airborne asset (at 60,000 feet) and the ground.

      • That says something far more interesting than any of the rest.

        If this idea actually works on a drone, that means it's both lower power and physically compact. No giant dishes, no enormous power budgets. That makes it far more likely that it would be useful in a consumer context.

        Something tower-based with typical military power requirements wouldn't nearly as interesting. Sure, it could work, but...

        "What was that sizzling sound?"
        "Line noise."
        "Why is there a dead bird lying on the ground outside?"
        "Like I s

    • by dywolf (2673597)

      Um...you dont need Line of Sight. It helps, but its not required.
      HAM operators have been talking around the globe for years.
      I pick up radio stations (AM and FM) from Atlanta in Macon, and from Tulsa in OKC regularly, and I'm definitely over the horizon in both cases, no matter how tall those broadcast antenna are.

      The key part of the transmitter is power, and it's the easiest way to extend the range.
      But the reciever matters too: the key step in a radio is the quality and design of the circuitry. My KIA car r

  • by argStyopa (232550) on Monday December 17, 2012 @01:24PM (#42315215) Journal

    And if you manage high-bandwidth 125-mi range, the next step is obvious - a constellation of LEO (200-500mi altitude) satellites serving as a nearly-untouchable* backbone for the theater-WAN.

    *ok not for peer-level opponents, but I'm pretty certain that a peer-level conflict
    a) will not be based on UAVs for long (my biggest concern about UAV-dependence of our forces), and
    b) will be over one way or another pretty fast if it's not going to turn SO nasty that any conventional force tech will be nearly irrelevant anyway (the not-so-comforting corollary that would invalidate my concerns above)

    • Actually, the next step is for the FCC to ban it under some vague and previously unknown test protocol as causing inteference to devices receiving the signal at -108dB. See also: Every attempt made to bring faster wifi to the masses so far.

      • Don't complain about the FCC, they don't make the rules. You should direct your invective toward Shannon, it's all his fault.
      • by afidel (530433)

        If you're talking about the Clearwire 2.4GHz trial, that was correctly halted when it was proven that 80+% of the commercial GPS receivers in the test area were effectively jammed. It shouldn't necessarily be the case that licensed users are responsible for negligible interference on adjacent bands, but the Clearwire trial was provisional on them being able to show that they could leverage those bands without affecting critical national infrastructure and they failed. That the failure was due to the failing

  • by evil_aaronm (671521) on Monday December 17, 2012 @01:30PM (#42315259)
    As a tool of our military, wouldn't this be rife for jamming by our enemies? Or is jamming avoidance part of the technology?
    • by LWATCDR (28044)

      Yes, yes, and yes.
      One possible solution would be using a directional antenna. A tight beam to the "WAP". Also Jammers have a short life time. Jammers tend to put out a lot of RF. Even missiles like the AIM-120 have a home on jammer mode. If this seems like an issue then the HARM AGM-88 will be updated to home on those jammers. The US also now has a lot of old AIM-7 SARA missiles that could also be updated to be an anti jammer missile. What it comes down to is do you want a fast network that may more may no

  • Who knows, we might even one day have 250 Mbps wireless links to our ISP.

    There, fixed that for ya.

  • they also need to focus on beaming energy around, esp. that distance. By being able to beam energy around, they make it possible to provide remote support to forward lines esp. with tanks, FOBs, and even ships at sea.
  • by Browzer (17971) on Monday December 17, 2012 @02:11PM (#42315689)

    This is actually a DARPA help wanted ad. And from description of the project sounds like a good job opportunity for some slashdoters.

    here is the ad:
    http://www.darpa.mil/NewsEvents/Releases/2012/12/14.aspx [darpa.mil]

    and here is the proposers' day conference:
    https://www.fbo.gov/index?s=opportunity&mode=form&id=e21984e31d49c3780966a53983daa4f6&tab=core&tabmode=list&= [fbo.gov]

  • by zerodl (817292)
    "Who knows, we might even one day have 100Gbps wireless links to our ISP." And be throttled to less than dial-up like every wireless isp out there. Mine sure does.
    • by snadrus (930168)
      With wireless links of that capacity, the competition for ISPs would be through the roof. For that matter, if the regulation allowed it, darknets may start forming (think Open wifi that doesn't connect to the regular internet). That's the best case, if setting-up a router made you a peer on a network of peers (all independently owned), there would be no ISP.
  • This is similar to an idea I cover in a blog post about future cities and how those cities may be connected.
    https://gautiertalksideas.wordpress.com/2012/09/23/economic-revitalization-of-an-advanced-civilization [wordpress.com]

The clearest way into the Universe is through a forest wilderness. -- John Muir

Working...