Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
Check out the new SourceForge HTML5 internet speed test! No Flash necessary and runs on all devices. ×
The Internet

Wireless Networking Research at Berkeley 165

zootallure writes "An interesting article about a self-configuring, wireless networking project going on at Berkeley's Wireless Research Center. Apparently, these Berkeley guys are convinced that they're going to leave Bluetooth and 802.11 in the dust."
This discussion has been archived. No new comments can be posted.

Wireless Networking Research at Berkeley

Comments Filter:
  • Although obviously a generic press release, but cute that they don't mention security [shmoo.com] concerns [netstumbler.com].
    • Re:Security? (Score:4, Interesting)

      by Gogo Dodo ( 129808 ) on Thursday March 28, 2002 @03:25AM (#3239617)
      PicoRadio [berkeley.edu] doesn't appear to be based on 802.11x at all so it doesn't have the same issues. Now it could have other issues, but I imagine that with all of the publicity that 802.11x garnered, that real security is being considered.
      • by dnoyeb ( 547705 )
        Why should there be ANY security on wireless? There is no security on wired data. That is to say, 802.11 provides no less security than ethernet. Except for the fact that its more easily overheard.

        Security is provided by your application, not your hardware layer. Thats how it should be. Else, once your hardware layer is cracked, ALL your data will be open.
  • by Anonymous Coward
    But the trick is to make one that's useful across a broad spectrum of uses.

    If it weren't for all the TCP/IP crap, network transmissions would be pretty fast. The transmissions would also be pretty useless.
    • Actually, if you read the article, high speed is not the purpose of this research project. Rather, a self-configuring, micropower distributed network which appears to be geared towards sensor applications. In fact, each sensor requires so little power that ambient radiation is sufficient to run the thing!

      It's likely that the true killer applications of this technology have not yet been proposed yet. For example, a network of biological implants might be possible that take measurements from several points and then perform some useful computations. The wireless communication and micropower features would be very attractive (provided that the body doesn't attenuate the operating frequency too much..)

  • a device "no bigger than a shirt button" capable of "tracking items or people" that "can harvest all the energy it needs from its environment."
  • by Anonymous Coward on Thursday March 28, 2002 @03:02AM (#3239496)
    Just look at wired networks. That's where wireless is going (but it's not gonna reach it, read on). The enemy of wireless is the 1/r^2 law. One must try to reduce r by using multiple base stations and complex protocols so that the r in each "hop" is as small as possible (saves energy, reduces interference). When r is 0 then it's called a wired network, which is the most efficient design, but which is unfortunately *not* wireless. But the design will look the same. So keep reducing those r's, and look at wired design for inspiration.
    • Someone needs to tell this guy that wireless is not wired. Why should wireless provide what wired does? AT every stage of the internet there is a most successful transport medium. You dont see gigabit networks in local business or home networks do you? Does that mean the Gigibit networks are not successful?
    • 1/r^2 only applies to an access node. Once you
      are trying to go point to point you use antennas.
      If it's a dish your loss is entirely do to the media your passing through: Water vapor, tree's, etc.

      The real bottleneck is that in air you can only use limited bandwidth (in the EE sense.) On a wire you get a huge swath of the EM spectrum, and then you get another huge swath on the wire you put next to it. Or, you get a small portion of a really high frequency EM when you install fiber, and again on the fiber you lay next to it.

      Wireless is another LAN & MAN solution, not a backbone solution. Though I'm pretty sure there are microwave links being used to cross Nevada, and why not? Light travels 50% faster in air than in fiber... (EM in clean copper only travels at 2/3 speed too.)
      • Wasn't NASA recently bragging how one of their satellites recently designed (launched?) had a directional high gain antenna that was made out of hundreds of tiny nondirectional antennas. To achieve a directional signal, each antenna broadcasts at a slight time skew to achieve constructive interference in the direction you wished to broadcast (and presumably, in the opposite direction as well).

        In theory, this should trickle down to consumer devices in 5 years or so. A portable device should be able to analyze the direction of incoming signals from the base station (by trangulation on its microantenna array) to figure out which way to broadcast the response.
  • by arivanov ( 12034 ) on Thursday March 28, 2002 @03:02AM (#3239500) Homepage
    ultra low power and ultra low bit rate

    Either the pressdrone have misheard or this is a specific project. Something like the X10 but on the air. It is quite cool for controlling devices and collecting data and stuff but it is a different niche. It is not competing with 802.11b(a) as suggested in the post

    • You're exactly right. The post doesn't even mention power, which is the whole point of this technology. This actually sounds cool. Do Slashdot's editors do this on purpose, or are they just being sloppy?

      According to Rabaey, each node in the picoradio network will draw no more than 100 microwatts of power, three orders of magnitude less than 802.11b's 300 milliwatts. The benefits of such a low power network are obvious: no batteries are needed because each node can harvest all the energy it needs from its environment.
      • This is typical of Slashdot. The editors don't have a clue about wireless, except that it's kewl and articles about it will get page hits for the new uber banners. To hell with research, journalism, or even proper spelling, we need page hits. Everyone thought this site was a pioneering journalistic community effort, but it was really someone's attempt to learn Perl and be r00t on a server.

        Getting back on topic so I don't get karma raped, hams have been experimenting for years with low power operation, which is called QRP. It's amazing what low power can do. There's hams that have communicated thousands of miles with just millwatts of power.

        • Well we all need page hits! Why do you think I have a hyperlink in my signature block - but going back on topic - any robot that far away would have to have virtual autonomy over its control because of the time delay involved in controlling it from Earth. The reason HAM radio can communicate on such low power is because in that waveband there is hardly any background noise to overcome.
  • They should try to get some funding from the XFM people.
    Since they are all paranoid about 802.11.
    • Do you have a link? I'd like to read more about what you say.
    • It's not actually paranoia, the satellite radio folks are trying to keep part 15 devices from interfering with their equipment. Right now they can shut down a part 15 (i.e. ISM / 802.11) system even if it is complying with out of band emmissions specifications. By tightening the out-of-band specifications, they put the responsiblity on the equipment manufacturer and not the end user.

      If anyone is paranoid, it's the 802.11 people. They're predicting the end of wireless ISPs and most don't have a clue what this is about.

  • From the gist of the article, it looks like this application is targeted more towards autonomous control networks; i.e. the example they cited of having these notes talk to each other in a building to control the environmental settings. Basically, these units sound like they'd be ideal in environments where there was limited or no power available, to transmit very small amounts of data in an un-supervised, and really un-maintained fashion, situations where 802.11b/Bluetooth would be overkill. Can anyone else think of other possible uses?
  • by nurightshu ( 517038 ) <rightshu@cox.net> on Thursday March 28, 2002 @03:10AM (#3239539) Homepage Journal

    As I was reading the article, I was thinking that picoradio would make an excellent addition to a modern soldier's loadout. As low-power as they are, the transceivers could be used to share data within the small-unit level without radiating enough energy to alert an enemy.

    Combine these with the Army's Intervehicular Information System (IVIS), and commanders would have real-time, accurate information on the location of not just every tank, APC, and field artillery piece on the battlefield, but also each soldier. It would definitely work to reduce the number of friendly fire incidents that occur in a future conflict.

    Plus, you could connect them to biomonitor equipment that would allow medic teams to both locate and triage injured personnel much faster.

    • Since these things are the size of a shirt button, millions of them could probably be dumped from a C130. They're light enough to not smash into a billion pieces when they hit the ground and their sheer numbers would make it impossible for enemies to effectively wipe them out.
      From the article:

      the nodes would monitor variables like temperature variation, light conditions, humidity factors and building occupancy.

      What would be neat is if they used natural packaging, such as faux rocks, or seeds. Even thistles that could stick to the enemy.
    • by jpmorgan ( 517966 ) on Thursday March 28, 2002 @04:10AM (#3239840) Homepage

      They're called ad-hoc wireless networks and the military is fully aware of their potential - both systems for maintaining reliable communications on a rapidly changing battlefield, and also the potential to drop thousands of small sensors from planes, etc... They first started looking into them in the early '70s, just after the development of the first wired packet-switching networks.

      In 1972 DARPA (the same people that brought you ARPANet, which later grew into the Internet as we know it) created a research project into a packet radio network, a.k.a. PRNet. They didn't get it working until around 1980, but in the end it did work, and was pretty fast too.

      The research was eventualy taken by the Army, Navy and Air Force who all started working on ad-hoc wireless networks tailored for their particular needs (for example, while the Army is mostly interested in fairly short-range applications, which this would be useful for, the Navy and Air Force are interested in algorithms to create reliable connectivity between ships, aircraft and ground stations that are below the horizon from each other through networks of satellites and aircraft).

      Current US military implementations of ad-hoc wireless networks that I'm aware of are the US Army TF XXI's Tactical Internet, the US Navy's ELB (Extending the Littoral Battlespace) ACTD (Advanced Concept Technology Demonstration) and DARPA's GloMo (Global Mobile) Information Systems program. These are all, for the most part, a bit more high-powered and high-speed than the system these guys are working on.

      In the public world, the IETF MANET working group are also trying to create a standard for ad-hoc wireless networking, but like the military implementations these are also a bit higher-end than picoradio. But if you're looking for something to wipe the floor with IEEE 802.11, IETF MANET is what to watch - but be prepared to wait a little while.

    • I might not know what I am talking about here, but isn't the idea of putting a bunch of wireless sensors on individual infantry soldiers sort of a bad idea?

      What is the possibility of a hostile force getting the very basic technology needed to pick up those transmissions and use them to locate said 'wireless enabled' soldier. Sure, the data itself would be encrypted, but with the location and movement vectors of a soldier who is using the system, you have all the information you need to make fairly dramatic changes in that soldier's biometric monitoring.
  • Interesting research (Score:2, Informative)

    by soundsop ( 228890 )

    I saw Jan Rabaey's talk at the International Solid-State Circuits Conference (ISSCC, the hardest of hardcore circuit conferences). The research is bold and fairly interesting. The slides [berkeley.edu] from the presentation are worth the read. The research might not pan out, but it's definitely worth a shot.

    You can find more technical info about his research on the PicoRadio [berkeley.edu] page.

  • MP3 player included (Score:4, Interesting)

    by villoks ( 27306 ) on Thursday March 28, 2002 @03:27AM (#3239627) Homepage Journal
    Two friends of mine is working in that project. As some of the posters have already noted, it's aiming to be something different than current systems e.g. ultra-cheap transmitters, which uses as less energy as possible. They are not quite yet there (as the article metions), the first prototype is actually quite large and uses lot of energy, but on the other hand only it's aimed to be a "proof-of-concept"

    Their current prototype has also a built-in MP3-decoder chip (really!) so it's possible that RIAA & CO will try to shut down the project with DMCA :-)

    Here's the homepage [berkeley.edu] of the project.

    V.

    • Those MP3's must sound great running at less than 10Kbit/sec. [berkeley.edu]This is definitely too slow to stream, and much slower than 56K modems to transfer. Is the goal to be able to remotely control a MP3 player (that has its own local storage or fast network connection)? The MP3 decoder seems like a waste of space and power to me. These devices seem much better suited for text and sensor communications.
  • The network could then configure itself--without any human intervention....sounds a bit like skynet only these things generate their own energy so we cant even try to pull the plug.
  • Some resources (Score:2, Informative)

    by jpmorgan ( 517966 )

    So this is just a very low-powered ad hoc wireless network, then. *yawn*

    They're talking about creating them to power themselves from their environment, and give examples of generating power from vibrations, or from small solar cells... which makes me wonder whether it would be possible to create picoreadio devices which power themselves from the ambient radiowaves. I know some British scientist built a radio which is entirely self-powered in this way, and it seems to me to be a great way of powering things like these (if their power requirement is low enough). Anyway...

    For those who don't know, an ad hoc wireless network is a wireless network like IEEE 802.11, but entirely self configurating, etc... etc... They're pretty neat things, but there aren't any real implementations outside of the US military, so these guys will really have a first if they get picoradio done soon. They're based on such great acronym^H^H^H^H^H^H^Hprotocols such as ZRP (Zone Routing Procol) or DSDV (Destination Sequenced Distance Vector) and DSR (Dynamic Source Routing).

    There is currently an IETF working group (MANET) trying to develop some standards, but there's a lot of research to be done first, so it could be a while before you see anything. Once they do put something out in a few years time, it'll kick IEEE 802.11's ass. =)

    If you want more info on adhoc networks, you can look at MANET's 'official' webpage here [ietf.org]. That page is pretty useless though, so you should look at their unofficial website here [navy.mil]. It has links to a lot of great resources.

    Ad hoc wireless networks are cool.

    • There IS a system outside the military. It was
      Richochet! Their radio-network was self-organizing!

      Metricom's ORIGINAL product plan called for building a
      system of radios that could be hooked up to Utility meters to
      get rid of meter readers. They needed the network to be self
      organizing so that the radios could be deployed randomly.
      It worked.

  • Out of all the science-fiction in star trek, I would have never guessed that a COM badge would become science-fact.

    Our privacy-hating government would love that. Screw putting all of a person's personal data on a card, just require the use of a COM badge that records your every move, word, which opens doors you are allowed access to, and it also acts as your car key. Whatta world...

    Aside from possible applications, the technology itself is fascinating. Don't look at my pointy ears, buddy.
  • Darn Kids (Score:3, Funny)

    by ByteHog ( 247706 ) <chris AT bytehog DOT com> on Thursday March 28, 2002 @05:00AM (#3239959) Homepage
    Always think they know better. This [linux.no] is a much better protocol for transmiting wirelessly. CPIP (carrier pigeon internet protocol).

    Not only does each node cost a heck of a lot less, but they're self-replicating!
  • One of the reasons BlueTooth caught on is the catchy name.
    Unfortunaltely I don't think 'picoradio' has the same buzz. I used to work on a wireless system called DECT (you yanks won't of heard of it, but it's big in Euroland), but even that's not catchy enough.
    Let's give these chaps a better name for their radio...how about GreenFace or RedSpot:-)
  • by gimlix2 ( 451817 ) on Thursday March 28, 2002 @06:00AM (#3240111)
    According to Rabaey, each node in the picoradio network will draw no more than 100 microwatts of power, three orders of magnitude less than 802.11b's 300 milliwatts. The benefits of such a low power network are obvious: no batteries are needed because each node can harvest all the energy it needs from its environment.

    As a current student in Professor Rabaey's classes, I can say that his ideas are pretty damn cool when he explains it in his own way.

    Sure, it's neat that his nodes will need no energy because it "harvests" energy from it's environment.

    Development is actually going on so that the "harvesting" actually comes from the natural vibration of a wall! [dailycal.org] [site: dailycal.org]

    So what do you get, a bunch of folks each developing their own thing:
    nodes the size of a button that you just stick on the wall and it just works.

    Pin-and-Play anyone?

    (infomercial voice)Set it and forget it!
  • by dtmos ( 447842 ) on Thursday March 28, 2002 @07:15AM (#3240270)

    There are many other research programs, both academic and industrial, on wireless ad hoc networks, going back at least to the 1978 DARPA-sponsored Distributed Sensor Nets Workshop at Carnegie-Mellon University. Most of the work has been funded by DARPA, by the low-power wireless integrated microsensors (LWIM) project of the mid-1990s and now by the SensIT project [darpa.mil]. (Their projects page [darpa.mil] lists more than 25 academic research programs on these networks, complete with links.)

    The University of California at Los Angeles, often working in collaboration with the Rockwell Science Center, has had a Wireless Integrated Network Sensors (WINS) project [ucla.edu] since 1993. UCLA also supports the similar-but-different "Smart Dust" program [berkeley.edu], which also employs ultra-low-power networking, but uses optical communication between network nodes.

    Professor Anantha Chandrakasan at the Massachusetts Institute of Technology is the Principal Investigator of the uAMPS (microAMPS) project [mit.edu].

    On the commercial side, these networks are being developed by Ember [ember.com], graviton [graviton.com], Wherenet [wherenet.com], and Motorola, just to name a few.

    The ZigBee industry consortium [zigbee.org] is the marketing and compliance arm of the IEEE 802.15.4 draft standard [ieee802.org], in a relationship similar to that between WECA (with the "Wi-Fi" brand) and IEEE 802.11b. This draft standard for ultra-low-power, ultra-low-cost wireless networking, now under development, should be finished this winter.

  • From the article:
    "It's almost like querying a database," says Rabaey. "If I send a request into the network saying, 'Give me the temperature in the kitchen,' it propagates through the network until it meets a node that says, 'I'm in the kitchen, and it's 70 degrees.'"

    Reminds you of a P2P network, doesn't it? (just to throw in another buzzword.... :-)
    • Sure.. But I'm certainly not going to configure a few hundred P2P clients for this thing to do all the stuff I'd want it to around my home.

      Temperature, humidity, etc. in each corner of each room of the home. Should also monitor music and light levels adjusting accordingly to where I am.

      Ie. I walk closer to a point light, the light should dim slightly. Walk away and it gets brighter to ensure that I always have a consistent level :)

      Thats alot of configuration for an X10 style setup.
    • Kinda reminds me of how ants communicate: One ant detects danger and sends a signal that is relayed. A2A network?
  • Some of the descriptions of the button-sized nodes sound straight out of Vernor Vinge's novel A Deepness in the Sky" [amazon.com]. Of course, in the novel they were more like grains of dust.

    similar slash story: Nano sized chips [slashdot.org]
    another related story: Smart Dust [slashdot.org]
  • Using this Low Power network you could implant a chip into a person or a group and pass data between each other as passive repeaters. Heck it could possible be an active repeater by using the ambient power by your nervous system. The ultimate living P2P system.

    This could in 30-40 years grow into something between SE Lain and Ghost in the Shell.
    • Put these in cars. If your house is too far from the nearest "tower," a car between you and the tower could relay a couple of packets, and then the next car could relay a few more. . . .
  • I always thought that cars could be used for wireless cells. Just think bandwidth would improve during rush hour! whhhoOOoHhoooo!

    Nick

Hard work never killed anybody, but why take a chance? -- Charlie McCarthy

Working...