Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
Technology

A Wireless Revolution From The Garage 123

Posted by Hemos from the rags-to-riches-and-riches-to-rags-riches- dept.
Saige writes "There's an interesting article in this month's Discover magazine about a lone inventor who's managed to develop a new method of wireless transmission that he was told was impossible - using "time-coded ultra-wideband electromagnetic pulses" - something that apparently can offer vast amounts of spectrum, allowing ways around the billion-dollar bidding going on now, with the bonus of better transmission through walls, higher speed/capacity, and higher resolution."
This discussion has been archived. No new comments can be posted.

A Wireless Revolution From The Garage More

A Wireless Revolution From The Garage

Comments Filter:

  • Re:Great story of invention (Score:1)

    by Anonymous Coward
    Too bad most of them are crackpots and never develop a solution because you can't beat the laws of physics.

    So the next time your professor tells you something can't be done, you've better listen to him!

  • Re:(+-) 10^-12 seconds!!! (Score:1)

    by Anonymous Coward
    Atomic clocks just need a few decaying cesium atoms. Probably not very dangerous.

  • Re:Location, Location, Location (Score:1)

    by Anonymous Coward
    >The biggest wow-factor is that the protocol allows you to determine location.

    Wow! They invented something, let's call it, um, "radio detection and ranging", hey, cool, that abbreviates to RADAR...

    And, btw, you can do precision position measurements with narrow bandwidth, continuous wave signals, too. People have been using triangulation for centuries if not millenia, and interferometry is at least one century old.

    The nice thing is that all these seemingly different methods are actually equivalent once you look closer at them.

  • bah! (Score:1)

    by Anonymous Coward
    it was tesla's invention, Marconi is just one of the enterprising bastreds to rip it off. U.S. Supreme Court ruled Tesla's to be the sole valid patent in June 1943, only months after Tesla's death. marconi sux http://www.dalmatia.net/croatia/mcadams/nikola_tes la.htm http://www.pbs.org/tesla/ll/ll_whoradio.html rsk@prodigy.net

  • Re:No free lunch (Score:1)

    by Anonymous Coward
    I'm surprised that you can get any receiver with a bandwidth larger than 100GHz, so much for the THz, or inverse picosecond. Note that picosecond pulses are fairly easy to produce, but you'll have trouble building any receiver with enough bandwidth to receive them in a moderately faithful way.

    Then the transmission over significant distances is a problem, the atmosphere will not act like vacuum: some frequencies will be strongly attenuated by oxygen and water vapor in the atmosphere. Check any recent textbook on radioastronomy and you'll see why. So your nice pulse will be very distorted and lengthened depending on the distance between emitter and receiver, humidity, phase of the moon, etc...

  • Re:Great story of invention (Score:2)

    by Anonymous Coward
    IMAO: In My Assinine Opinion?!?

    Dude, don't be so hard on yourself.

  • Re:Spread Spectrum Technology (Score:2)

    by Anonymous Coward
    > They're trying to get FCC approval since it doesn't cause (enough) harmful interference to take down communications with existing equipment.

    This is a wrong assumption. It doesn't matter if you describe/encode information in the time domain, the frequency domain, or any arbitray other domain - the total transmission capacity of your channel is the same. If you add broadband spread-spectrum signals, you do take away capacity, and you _DO_ interfere with the more traditional ways of RF transmission. Spread spectrum signals look like additional noise to narrow-band receivers; the more spread spectrum signals you have, the more noise you get. The more noise you get, the worse the S/N of your narrowband link gets, which degrades the capacity of your narrow band channel.

    There's nothing revolutionary or even new about this "ultra wide band" stuff. Wether you describe an RF signal as a superposition of sinusoidals, pulses, wavelets, or any other linearly independent set of base functions simply does not matter.

  • Re:VMSK, anyone? (Score:5)

    by Anonymous Coward on Friday May 04, 2001 @10:18AM (#245197)
    It isn't VMSK. VMSK was the digital equivilant of that tired old april fool's joke of reducing FM deviation until the signal had essentially no measurable bandwidth. Shannon's Limit shows why it's not workable.

    This stuff is real, but the hype and spin is nothing short of amazing. Folks, get an electrical engineering degree and you'll learn that it is hardly revolutionary.

    Look up "Costas Loop" on your favorite search engine and you'll see how he stays synchronized. No, you don't need an atomic frequency standard. And by the way, this design as been around since at least the late 1950's.

    I first learned about pulse phase modulation from an ARRL publication in 1978 while studying for an Extra Class license. The idea of spreading it further was simply a matter of changing the timing in a pseudo random fashion. BFD.

    In any case, UWB is not much of an issue UNLESS you give it to the general public to take everywhere. Before you know it, Radio Astrononomy gets hosed. Next, a few ham radio operators complain. Then a few incidents in airliners regarding Portable Electronic Devices start making GPS go nuts, and before you know it, police radios stop working within a few hundred feet of Starbucks when they start offering wirless internet portals...

    That's right, you can be compliant with 47 CFR 15 and still mess with licensed radio services. It's called the near/far problem with spread spectrum. Look it up.

    If we could somehow all turn off all of our narrow-band radio gear and magically turn on our computer enhanced digital communicators, we'd all probably love it. But that's not practical. We will have narrowband communications for a very long time to come.

    UWB devices will have to avoid trashing GPS frequencies (it's a spread spectrum system too) without introducing filters that introduce excessive group delay distortion. They'll also have to figure out how to get good dynamic range and sensitivity with a very wide band front end that lets in some remarkably powerful signals.

    Most computer professionals know so little about the transmission media they use for their LANs and WANs that it's comical. They would love to dispense with FCC licensing but they have no concept of what they're getting in to. Folks, unlicensed operation means you must ceace and desist when a licensed operation asks you to. It also means you must accept any interference you may receive. And people want to build a wireless infrastructure on THAT?

    We need the FCC. We need narrowband channels. Just because they're not doing their job poorly doesn't mean it doesn't need doing.

  • Re:Spread Spectrum Technology (Score:5)

    by tzanger ( 1575 ) on Friday May 04, 2001 @09:33AM (#245199) Homepage

    erhaps I don't fully understand this invention, but to me it sure looks like snake oil.

    You don't understand the technology.

    Check out the Ultra WideBand Working Group [uwb.org], time domain (mentioned in the article) and dozens of other sites. The pulses are what makes this thing work and the spectral splattering is exactly intended. They're trying to get FCC approval since it doesn't cause (enough) harmful interference to take down communications with existing equipment.

    It's way cool stuff. Check the UWB link provided; they want to use this for positioning, through-wall "radar" and communications. It has very serious potential. If I can find the EDN issue which gave an in-depth study of this I'll post back.

  • One intriguing antenna design, which was dropped a LOOONG time ago, was the square arial. (There was a lot documented on it, in the 1920's, in Wireless World, but there's not been a whole lot since.)

    The square arial works on the idea of having two dipoles that are linked via the two cross-pieces. Since you design antennas, you can probably go into more detail on this concept, but as I understand it, it's essentially a self-amplifying arial.

  • This scheme sounds strangely too familiar to
    VMSK. (a fraud perpetrated by Hal Walker).

    There's an excellent treatment of it by Phil Karn (of KA9Q fame) at http://people.qualcomm.com/karn/papers/vmsk/index. html
  • *My* car already starts everytime I download Natalie Portman pictures!

    Sorry about yours...

    Vroom vvroooom!

    t_t_b
    --
    I think not; therefore I ain't®

  • Perhaps it uses some sort of pseudo random number generator. If sender and receiver use the same generator, pulses could be sent and the receiver would know when to expect them. Now how one would go about syncing this up and making sure that the receiver doesn't start following another transmitter is another story.
  • It also modifies the precise intervals at which the pulses are released to create a uniquely randomized pattern that can be recognized only by a similarly coded receiver

    Sounds like this makes it so two devices can only talk to each other. From the statemets made that this is secure, it can't be as simple as turning a radio dial to find a frequency.
    ---------------------------------------
    The art of flying is throwing yourself at the ground...

  • If you want to resolve picosecond pulses, you need electronics that can pass frequencies up to 1 THz.

    Not really. In fact, there is probably no frequency higher than 200MHz on the chip, but it can control the phase of the signal with very fine resolution.

    Here is how this might work:

    Start with a ring oscillator (basically an odd number of NOT gates in a ring) and lock it onto a reference frequency with a phase-locked-loop. Use a multiplexer to tap the signal at different locations along the loop. This gives you the same signal at different phases, but the resolution isn't quite down to tens of picoseconds yet. The next step is to use another multiplexer and a series of carefully tuned delay elements to further improve the resolution. You now have a pulse generator that can modulate the position of individual pulses with a resolution of tens of picoseconds by a digital control word.

    This can all be manufactured with standard CMOS processes on a chip.

    It may take millions of dollars to develop this technology and make the process repeatable enough, but once you have it, making many copies of the circuit is cheap.

    The low power usage results from the fact that an ultrawideband signal does not suffer from fading - if there are no sines then different reflections of the same signal can't arrive out of phase and cancel out each other. Most wireless systems take a very wide margin to accomodate for this type of fading.

    -
  • VMSK sounds completely different. Timedomain's technology is made up of wideband pulses. The key is that the receiver has to be clocked in to received the signal. If it is not aligned, you don't hear anything.
  • Not only that, it was posted to /. on april 10th 1999. See http://slashdot.org/articles/99/04/10/1920203.shtm l [slashdot.org]

  • Re:Spread Spectrum Technology (Score:3)

    by Detritus ( 11846 ) on Friday May 04, 2001 @09:40AM (#245208) Homepage
    You can't have a short pulse without all those harmonics. That is what makes it ultra-wideband. You also need a transmission line and antenna that wont smear the pulse. It would have to be wideband with good group delay characteristics. The advantages of using a very short pulse would be frequency reuse. You could have a large number of transmitters and receivers simultaneously (actually it is a form of TDMA) using the frequency band.

    As a firm believer in TANSTAAFL, I wonder what kind of interference this system will generate. The EM spectrum already suffers from severe noise pollution in many areas. Think about that when you run your computer with the case off.

  • So it's like digital CW.
  • The transmitter and receiver can be synchronized if the receiver knows "when" the next bit will come. But, the speed of light (which is the speed of the radio signal) needs to be constant, and it is NOT. Remember, the speed of light IN A VACUUM is constant; the speed changes when the light travels through media of different densities, such as the gasses in the atmosphere. I believe there may be problems using this technology for long-distance communications.

    --
    This sort of thing has cropped up before. And it has always been due to human error.

  • You're forgetting that Win95's limitations are a lot less weakly defined than Nature's Law(TM).

    Try break Shannon's Theorem. Go on!

    Go make a Perpetuum Mobile. Now?

    Now make a lossless compression algorithm than compresses any file by 1 bit or more. Well?

    Oh, how about a completely unbreakable encryption algorithm with a reusable finite-length key. Any succes?

    If you claim to have succeeded with that, you're mistaken, or God Himself.

  • did I say "less weakly"?

    Oops typo.

    s/less weakly/more weakly/

  • Did you read 'reusable finite-length key'?

    "iteration period"

    That's where it breaks down to needing an infinite key length for unbreakable encryption.

    2**64 is still a finite amount of data to encrypt with that finite-length key.

  • Very well said HuskyDog.

    Just for people with a more CS oriented background: Consider Shannon's information theorem, given particular signal and noise levels and bandwidth, there is a fixed amount of entropy (=bits of information) that you can put through it.

    Mr Larry Fullerton simply spreads out hist signal across a very wide bandwidth, stealing a little bit of entropy from each frequency. But since the spread is so wide it is only a very very little bit of entropy, such that current frequency-domain transmission systems will not notice a degradation. And since large parts of the spectrum that Larry's signals cover are still unused, he is 'stealing' entropy from a lot of frequencies where nobody is even using it, let alone be affected by the raised noise floor. Hence, he will be able to broadcast a lot of entropy (bits per second) without hampering any of the existing broadcast systems.

    The bandwidth of Larry's system is still limited by the pulse slope of his chips and by signal distortion by his antenna's, hence the bandwidth his system covers is not infinite, hence at any bit rate, he will always raise the noise floor all over the spectrum when he transmits. And I'm sure that when he increases the bitrate of his transmitter enough, he is also increasing the amount of transmitted power without increasing the bandwidth of the tranmission, hence he is increasing the noise floor that the other communication systems will experience.

    So, concluding: there will be a practical limit on how many bits you can transmit with that system, but I'm sure it can make the effectiveness of how we use our available air bandwidth a lot better.

    I agree with HuskyDog's next to last statement though: A spread-spectrum technique with a sufficiently wide spreading achieves the same effect and should be capable of achieving the same level of efficiency. It could be though that Larry's chips will be smaller (read:cheaper) than a spread-spectrum chip trying to get the same bandwidth.

    And then there is HuskyDog's point about the wide bandwidth antenna's... One of the reasons why occupied bandwidth per channel is limited is the fact that antenna's inherently have a maximum gain dependent on the structure of the antenna (segment sizes etc) and that the more wideband an antenna should be, the larger it gets and/or the lower the signal strength you get from it.

  • "Not to mention that keeping two physically seperated clocks in lock-step with sub-picosecond accuracy is not exactly easy."

    That's exactly what killed my experimentation with spread spectrum. Keeping the timebases locked *that* tightly together, at *those* frequencies, well, that was *waaaaaay* beyond anything my meager budget would permit.

    I suppose the new gizmo has figured out a way to overcome this, other than maybe also transmitting a carrier signal at its timebase frequency. I'm still rather distrustful of it. I grok that the reason it's "wideband" is because it uses a pulse. What I don't grok is why it *has* to be "wideband", and why it can't just be done with phase-shift modulation of a carrier.

    Also, at the terahertz range, won't that be limited to VERY short line-of-sight? Granted, "sight" might not mean "A building is blocking my signal", but I'd have to say that the "signal" will be *waaay* down in the mud already, given that you *must* also pick up, on a non-resonant antenna, the third, fifth, seventh, ninth and so on. Amplifiers that work at the THz range are VERY non-trivial, and it'd have to be designed so that there is *zero* tolerance allowed for signal delay on all those harmonics, else your "pulse" would not look like a pulse so much as like a splatter.

  • Re:Spread Spectrum Technology (Score:3)

    by MsWillow ( 17812 ) on Friday May 04, 2001 @09:23AM (#245216) Homepage Journal
    Well, maybe it is a form of spread spectrum. Each "trillionth" of a second, the transmitter sends a "pulse". It is either slightly ahead of, or slightly trailing, the timer signal. As it's a "pulse", it *is* a sinusoidal wavefront, comprised of the fundamental, and the third, fifth, seventh (and so on) harmonic.

    It looks doable, with a phase-locked loop to tell you whether your signal was leading the timebase (a digital "1") or trailing it (a digital "0"). What I don't understand is why he has to use a "pulse" at all? Why not just phase-0shift modulate a carrier signal?

    All the pulses do is splatter harmonics every which way, providing no real benefit. It's not like you receive a "pulse", per se, what you receive is a whole huge pile of very brief signals on a gazillion different frequencies.

    Perhaps I don't fully understand this invention, but to me it sure looks like snake oil. Has he ever looked at an FFT of his signal? I think not. I also suspect that, while he may be bright, the inventor hasn't actually studied rf theory.

    It's not a hard mistake to make. I did the same myself, years ago, before learning more about fourrier transforms and what they tell us.
  • The advantage of LOS technology is that there is no shared spectrum, and therefore no legitimate need to regulate.

    A guy I know named Cooper Lee used to run a company where this little fact was the sole reason for their profit margin.

    The company that he and his grandfather [tmex.com] put together, called Tmex [tmex.com], run telecommunications services between the US and Mexico over ATM using Astroterra's (now known as OpticalAccess) [opticalaccess.com] LOS products [opticalaccess.com]. The beauty of their solution is that by doing this they can hop the border between the US and Mexico without having to pay costly interconnect charges to Telmex [telmex.com] (the Mexican telephone monopoly) and whatever US carrier their client is using.

    Rather, they just tie in to the local lines in the US and Mexico. They also skirt all the pesky regulations of having to obtain the same frequency spectrum in two countries at one (as would be the case with microwave and other RF technologies).

    All this with just a box on top of a Howard Johnson's hotel in the US, and a tower in Mexico 5 miles apart.

  • As cool as this sounds.. I recall reading this exact stuff at least 3 years ago.
    It's not new news.

  • Ommissions and SETI (Score:3)

    by z00t ( 24587 ) on Friday May 04, 2001 @09:46AM (#245219)
    Conspicuously absent in the article is mention of the inherent security of this communications scheme. The low power of the signal combined with a pseudo random sequencer makes this signal virtually indestiguishable from noise. One needs to be within a couple of *feet* to detect a transmitter.

    I know the Pentagon/Nat'l security apparatus aren't especially keen to have devices based on this enter the civilian domain very soon. The FCC will surely play a part in regulating the timetable.

    As a curious aside, if alien civilizations are using this type of communication or any other which strives to make its signal indistinguishable from noise to anyone but the intended receiver, SETI will never pick them up.
  • Well, it does say he's been working on it for decades, so some of us have heard of it. Assorted info in the field can be found in aGoogle search "wideband+pulses" [google.com].
  • Except for the fact that this particular band already suffers from interference produced by electric motors at about the power level. As such, the noise floor has already been raised significantly. Except distance deals well with this now. So, based on similar power levels it should not cause more interference than running a Razor.

    This does mean the power level of such devices would have to be strictly controlled. But it means there could be no reasonable expectation of unusual interference to pre-existing services.

    Even a large array of them within a small area should not create significantly more interference than a manufacturing plant with a large number of eletrical motors.

  • Great. (Score:4)

    by revscat ( 35618 ) on Friday May 04, 2001 @10:40AM (#245222) Journal
    Now we're all gonna have to dip into our pockets so that SETI can add search capabilities for *this*. Man, I wish the aliens would just go rap on G. Dubya's noggin and say "Howdy" instead making us waste all this energy on screensavers that probably are gonna result in jack squat.

    Aliens on the brain... And my ass hurts. Hmm.

    - Rev.

  • Evil Uses (Score:3)

    by rarose ( 36450 ) <<rob> <at> <robamy.com>> on Friday May 04, 2001 @09:29AM (#245223)
    Note the company's website promotes the idea of police using this stuff for through-wall radar. Remember that thermal imagers have led to people getting arrested because their garage gave off more heat than the cops liked when they drove down the street...
    Next people will be getting arrested because the cops saw them leaning over their coffee table too frequently... or maybe in Georgia the cops will use the radar to see some oral-to-body-part contact that happens to be illegal down in the South.
    This is a bad thing.
  • Not a signal geek myself, but...

    Isn't part of the point of this approach that frequency allocation would be unnecessary if everyone used this type of signal? And that the "dirtying" of "other people's" spectrum is not noticable? IOW, the only people who should give a shit are those whose "property" will be devalued by abundance? I'll shed a tear for Time Warner, honest.

    Of course, if I read the article correctly, the transmitter and receiver have to be coded to the same random "seed" - the same one that keeps the signal from producing harmful interference - which means that this technology is immediately useful for things like local wireless nets, walkie-talkies, etc. That's a long way from the Ultimate Wireless Solution.

    Boss of nothin. Big deal.
    Son, go get daddy's hard plastic eyes.

  • After reading this, to me it's just old and tried Spread Spectrum technology. The spreading seems just to be wider than we are used to.

    I have troubles appreciating any real benefits from this technology.

    //Pingo

  • oohh... I'm soooo excited!!!!!!

    Please, I can't believe that /. would print this... cold fusion we're going to save the world!
  • What about fractal antennas? I thought those fixed a lot of the problems associated with transmitting/recieving wideband. I haven't heard anything about them recently, but I'm curious what you might have to say about them.
  • Yeah, it sounds basically like low-power digital frequency modulation (FM).

    Digital transmission is pretty interesting, and the low power is certainly interesting, but I don't see how this can be used for the vast amount of applications they are proposing. And I also can't see how they can say it won't interfere with conventional (analog) radio signals.
  • One day someone just might, and they'll throw away these theories and say "I dont know how anyone could have believed xyz". The point is that Microsoft had a false assumption, that PE files were more resistant to infection, this was proven to be false and the theory of viruses being impossible on the win95 platform was invalidated. Soon after they said it was only possible to write a resident virus because memory protection is slack on win95 platforms, and then said it was impossible to write a resident winnt virus. They were wrong again. why? Because their assumption that a resident virus must run in ring 0 was false. Per process residency viruses run completely in ring 3 and are as effective in practice as a system wide virus. Once again, the assumption is undermined and the theory folds. If you would like we could give some examples in physics, but I think the domain I have chosen demonstrates the issue just as effectively. Never say never.

  • They told me it was impossible to write a virus for win95. We did it before the first beta was released. Impossible is highly overrated.
  • If that really is Larry, he's lost some hair since the last I saw him. And grown a beard.
    ---
  • Must be different. I'm talking about Andre

    I meant that if that really is Larry Fullerton's photo with the article, he's changed a lot since I've seen him last. But then, I've lost a bit off the top myself since he's seen me.
    ---
  • any day now as I grow carnivorous plants in my bedroom. Its a eccentric hobby of mine.
  • Actually, the NTIA assessment of this stuff is that, Yes, it may work, and, based on actual and theoretical evaluations, its potential for harmful interference is significant above certain threshholds for power and "chipping rate."

    The ultrawideband nature of this approach appears to offer a splendid opportunity to raise the noise floor across a large region of RF spectrum, to the point that little things, like GPS navigation and cell phones as we know them today will be adversely affected.

    You're right: It's a neat development. But has it been appropriately evaluated to allow its implementation in concert with other, extant narrowband and conventional spreading technologies? I think not.
  • Their prototypes are not nearly as ridiculous sounding as the press articles I have seen, and aren't really that great except for their extremly low power consumption:

    The power consumption could be the kicker! As so many others on this item have posted, the Shannon limit is just that, a limit. You can use more of the spectrum to get more bandwidth, but then you start hitting other walls, like being unable to build antennae with the characteristics you need. So the search for more bandwidth is constrained--there's only so much out there.

    But have you seen the cost of wireless communication in terms of power? Power requirements for wireless transmission technologies in use now are on the same order of magnitude as computation; I don't care whether it's narrow-band or wide-band. There's lots of sneaky things people are doing to lower the power consumption of CPUs. However, the return on those sorts of optimizations is getting less and less important because of the power requirements of the rest of the system. Consider that the power requirements of the CPU, display, and hard drive are all the same order of magnitude (6W to 4W to 2W in at least one system [cmu.edu]).

    But on top of that, in the next, say, eighteen or twenty-four months, communication is projected to be more than three quarters of the power budget of a handheld wireless device. (And I can't find the goddamn link for this one.) That's where a wireless transmission technology with low, low power requirements could really make it in the market.

    To get that pie-in-the-sky handheld Roblimo was dreaming about the other day, the power consumption of these sorts of systems has to come way down. Targetting the power use of communication is an important first step.


    Jonathan David Pearce

  • Aw, hell, I found the link. It's the same paper that I linked to before. Here it is again [cmu.edu].


    Jonathan David Pearce

  • Well, it is possible what he is doing is equivelant to a single-sideband radio, using the phase locked oscillators to reconstruct the signal. This mights justify their low power claims. Doesn't really sound like it to me, though. I should read the whitepaper on their website.

    But the fact remains--to resolve 1ps pulses, you need electronics that can handle 1THz. That doesn't seem so likely to me :)

    Not to mention that keeping two physically seperated clocks in lock-step with sub-picosecond accuracy is not exactly easy.
  • They didn't say it would eliminate the bandwidth problem, they said it would eliminate the spectrum (the fact that we're running out of it) problem.


    Well, the reason we are running out of spectrum is because high-bandwidth applications take a lot of spectrum.


    Now part of the problem is that you always lose some of the spectrum when you divide it into channels finely. Upon further reading on their website, it looks to me like they aren't really going for particularly high bandwidth, but just being able to have many, many channels. After reading a large chunk of their whitepaper, I am skeptical that it is as scalable as they say, but it comes off as a lot more plausable than the articles linked to.


    As a bit of historical context, every few years someone comes up with a brand new revolutionary way of dividing up the spectrum into channels that is incredibly efficient. Two that come to mind are CDMA and FHSS (used by cell phones and some wireless lans, respectively). In each case, once they make it to actual production and deployment it turns out they vastly overestimated the number of channels that can work simultaneously without causing severe degredation. I doubt this will be any different.


    My best guess for the deadly factor is multi-pathing. While the pulsed nature of the radio should prevent destructive interference that causes cell phone fadeout in doors, it will probably substantially reduce the number of channels they can use before they start causing problems.

  • Re:Spread Spectrum Technology (Score:4)

    by norton_I ( 64015 ) <hobbes@utrek.dhs.org> on Friday May 04, 2001 @11:24AM (#245239)
    I am reading the information on timedomain's website, which includes a frequency spectrum analysis, and what they have actually demonstrated. First of all, their Their prototypes are not nearly as ridiculous sounding as the press articles I have seen, and aren't really that great except for their extremly low power consumption:
    • A full duplex 1.3 GHz system with an average output power of 250 microWatts, and a variable data rate of either 39 kbps or 156 kbps. The radio has been tested to beyond 16 kilometers (10 miles).
    • A full duplex 1.7 GHz walkie-talkie with an average output power of 2 milliWatts, a data rate of 32 kbps and a range of 900 meters. The unit was also capable of measuring the distance between radios with an accuracy of 3 cm (0.1 ft).
    • A simplex 2.0 GHz data link with an effective average output power of 50 microWatts, a data rate of 5 Mbps at bit error rate (BER) of 0 with no forward error correction (FEC) and a range of 10 meters (32 ft) through two walls inside an office building.
    Also, their pulses are not remotely square, but gausian monocycles of the form: V=t*e^(-t^2). In addition to low power, they claim that they can do really dense channelization by using different clock sequences -- very similar to the way FHSS works, only by varying the time base, rather than frequency hopping.

    Another plausable advantage is that since they don't use continious waves, multi-pathing isn't a big problem. The wave packets from the two paths are completely distinguishable, and therefore do not interfere. However, this makes each path look like a seperate transmitter on a different channel. So you sacrifice total bandwidth (by reducing the number of channels availbable) in exchange for reducing fade-out from point of destructive interference.

    In any case, anyone interested should check out this whitepaper [timedomain.com] more info. I doubt this is a scalable as they claim, but they do have some interesting ideas, and the single-chip positioning and radar sounds cool, too.

  • Not to be a naysayer... (Score:5)

    by norton_I ( 64015 ) <hobbes@utrek.dhs.org> on Friday May 04, 2001 @09:28AM (#245240)
    Not to be a naysayer, but this really isn't that remarkable. In particular, it basically has exactly the same constraints as "normal" frequency domain signals. In particular, it has to take up the same sized region of the spectrum as conventional broadcasting, and all of your electronic components have to work at the same high frequency to resolve data.

    If you want to resolve picosecond pulses, you need electronics that can pass frequencies up to 1 THz.

    That isn't to say there aren't applications, and I am going to read their whitepapers to see how they get such phenomenally low power usage, even when constrained by the inverse square law, but it isn't a revolutionary technology that is going to eliminate the bandwidth problem.

    Short-range transmitters with closely spaced receivers connected by fibers could solve the bandwidth problem, but A) would require massive investment in infrastructure and B) be no better or worse than Bluetooth type proposals using more conventional radio technology.

  • Re:Spread Spectrum Technology (Score:3)

    by scoove ( 71173 ) on Friday May 04, 2001 @09:40AM (#245241)
    Oh, there are some benefits, such as using frequencies allocated to others.

    Check out the ARRL's response [arrl.org] to the FCC (including references to the Qualcomm report). A primary concern is:

    the broad nature of the interfering signal . . . indicates that any interference would extend to all VHF and UHF amateur bands." That particular report dealt with lab tests to assess the impact of UWB emissions on PCS phones using code division multiple access (CDMA).

    Sure UWB gives you more bandwidth... by using other user's licensed frequencies. It's like bragging about how much faster your PC goes when you stealthly sneak distributive processing apps into other PCs without their authorization.

    *scoove*

  • how about s/less weakly/weaker/
  • its pronounced doo-mass

  • Not exactly new... (Score:5)

    by Matt_Bennett ( 79107 ) on Friday May 04, 2001 @09:45AM (#245244) Homepage Journal
    Ultrawideband is not exactly new, I worked on a UWB radar almost 10 years ago. It has promise, but it also has problems that you don't see with traditional carrier based methods- since you are talking about picosecond resolution, communication with a reciever that is moving with respect to your transmitter is hard- in a picosecond, light travels about .3 mm. Virtually any motion can cause your reciever to start looking at the wrong part of the data- since the data is all time-coded.

    Many have said that you can't detect UWB, but usually, they are looking at it with a traditional carrier based spectrum analyzer- you can't see it, but it is there. It just looks like additional noise. which looks very similar to spread spectum- which is exactly what UWB is, but you're spreading the spectrum very, very wide- an impulse in the time domain is uniform (all frequencies, equal amplitude) in the frequency domain.

    Time domain has been around, and pretty big in the UWB arena for a long time, but most of what they have is vapor- big promises, but they fail to deliver. There are other players in the field, Aetherwire [aetherwire.com] is one of the other biggies of UWB. There has been a huge amount of money poured into UWB, but dreadfully little usable technology.
  • It's not in the box, it's in the band.
  • As for established players, they've got the advantage of being established, if they can't adapt, let 'em sink. Most Republicans feel this way.

    "Most Republicans". Yep, except for the ones who actually hold office. For them, protecting established players is the sole purpose of their existence ... especially when they consider how much money in bribe-- er, "campaign contributions", they've received from those established players.

    Same is true for Democrats too, so don't even bother responding just to point that out. Corruption is bi-partisan.

  • Ever since a professor at the University of Arkansas told him that such [ultra-wideband] pulses could not be used ...
    They're told it can't be done, and they obsess until they develop a solution.
    Or they were told so by an idiot, or they misremembered what they were told, or...

    What did the professor really say? "not practical [using today's technology]"? Or "physically impossible"?

    In a nutshell: TANSTAAFL. This is just spread spectrum using a very wide band. Yeah, he says imperceptible to other users of the spectrum - as long as he's the only one doing it! Get millions of users doing this and every user of spectrum will notice a higher noise floor. Higher noise floor == decreased channel capacity. Read Shannon.

  • Isn't part of the point of this approach that frequency allocation would be unnecessary if everyone used this type of signal? And that the "dirtying" of "other people's" spectrum is not noticable?
    No.

    If we switched our entire usage of spectrum from bandwidth allocations to this kind of ultra wideband usage, it just shifts the regulatory burden. Instead of assigning frequencies (and therefore controlling the number of transmitters) you might assign power levels (and hence control teh number of transmitters).

    The dirtying of other people's spectrum is not noticeable if he's the only one doing it. If this becomes widespread [so to speak!] you bet narrowband users will notice it. The noise floor will be raised, so their channel capacity is lowered.

    GPS signals use codes (pseudo-random sequences) to spread their signal out. A type of aviation beacon - DME, or Distance Measuring Equipment - also does it. By controlling how you use the code to spread out your signal, you can confine your usage to one band of the spectrum, and hence not interfere with users outside the band. GPS uses this so all satellites occupy the same spectrum without interference. DME uses it so many airplanes can use one beacon without interfering with each other.

  • My physics tells me that there is only one wide band spectrum in the atmosphere. There is no frequency band to tone into. ... The question is how to identify your buddy among thousands of wide band users.
    GPS does this - all the satellites use the same carrier frequency, and each signal is below the noise level. So how does the GPS unit receive the signal?

    You don't separate the signals from one another (and from the background noise) by being selective in the frequency domain (ordinary "tuning"); you do it with other means. Without going into a lot of theory, each satellite (or in this scheme, each transmitter) has a specific pseudo-random number (PRN) sequence (this can be generated by feeding a seed into a linear-congruence pseudo-random number generator). You use this sequence at 1 MHz to modulate the phase of the carrier (for civilian GPS, the L1 frequency of 1.57542 GHz) which spreads the signal over a bandwidth of 1 MHz. But before you do this, you modulate the PRN sequence with the message (at 50 Hz). To receive, the receiver knows the PRN sequence but it doesn't know what the propagation delay is; so it keeps shifting its private copy of the PRN sequence and looks for a statistical correlation between it and the received signal. When it finds it, it has "locked on". Now it can demodulate the satellite's PRN sequence to recover the 50 Hz navigation message. This is CDMA - Code Division Multiple Access. The conventional "tuning" approach is frequency division multiple access. You can also use phase, time, or other attributes to divide the signal into multiple channels. In the USA, Sprint PCS mobile phones are CDMA; AT&T mobile phones are TDMA.

    Here's partial correlation [colorado.edu] between the recieved and generated GPS PRN sequences. Here's a little animation of PRN "lock on" [colorado.edu].

    For this Fullerton method, the same sort of thing, except that instead of spreading a carrier into a 1 MHz bandwidth using a PRN sequence to modulate the phase, you use time division access over a much wider bandwidth - effectively 0 - 1THz or so which means the power at any frequency i.e. the chance of interfering with a user at that frequency - is very very low. But it could interfere with GPS.

    Clear? No? That's what I was afraid of! It's hard to understand all this without a little signal processing background.

  • Expect the FCC to step right in and destroy it, if a huge company doesn't claim a 15 month old patent on it first.
  • Why is the above post marked as troll? The poster is right - it's spread spectrum technology. What's new here?
  • Nice sarcasm there. d;-)

    - Steeltoe

  • Does the argument presuppose that the "signal" is a constant stream of information? In the event that one is talking about a real information profile that comes in fits and starts, then surely some of the methodologies outlined are better than the others. I agree with the information per unit time limitation but in the real world might not the methodology of the article produce a better responsiveness for the same bandwidth?

  • Very good comments.

    Since the burst transmission used by this technology covers all frequencies, would it be possible to fit a filter to the transmitter, preventing it from using frequencies below 300 GHz?

    This way, it wouldn't interfere with existing carrier transmissions on existing frequencies. Conventional continuous carrier-based transmission could then be used on all frequencies below 300 GHz (infrared). This new burst-based transmission could be used on frequencies above that, which are mostly unused by conventional transmitters. Then, there would be no interference....


    Super eurobeat from Avex and Konami unite in your DANCE!
  • Not quite. The use of an autocorrelated pulse receiver brings many interesting possibilities to the table beyond what DSSS or FHSS can do. You end up with a hell of a lot more process gain, for one thing.

    I wish I understood how the correlator synchronizes itself to the desired time sequence... that's the part I've been unable to grok in the last two or three articles that have been published on TD technology.
  • Best of all, if approved by the Federal Communications Commission, the technology would not require the huge expense of purchasing the right to broadcast on a licensed frequency


    Actually, that's why the FCC won't approve it. Think of the revenue they would loose.
  • The LLNL micropower impulse radar site is mostly hype. There's been a lot of loose talk out of LLNL about how this is a great, cheap technology. Actually, it doesn't work very well. The wideband receiver is sensitive to narrowband noise. Despite years of hype, the one application that's actually shipping is from Bindicator [bindicator.com], an company which builds the boring, but essential, devices that measure how full a bin, silo, or tank is. The micropower impulse radar thing is great for measuring piles of solids, where you can't use a float. Typically units like this go inside a metal tank or silo, so they're shielded. (Previous technology for doing that job involved units like the Bindicator Mark III Yo-Yo [bindicator.com], which lowers a weight until the cable goes slack, then winds it up again. Noncontact methods are a big win.)

  • Just for the record. (Score:3)

    by Mr_Icon ( 124425 ) on Friday May 04, 2001 @09:12AM (#245258) Homepage

    The Time Domain's website is timedomain.com [timedomain.com] (flash plugin required, tho).

  • The FCC and other government institutions are always going to have a hand in anything that scatters signal at RF.

    My bet is on optical line-of-sight being developed by companies like Terabeam [terabeam.com]. OK, right now it's expensive technology only practical for corporate customers. So were computers 30 years ago.

    The advantage of LOS technology is that there is no shared spectrum, and therefore no legitimate need to regulate. The only people who might want to regulate it are neighborhood associations, for aesthetic reasons. However, most of them have accepted satellite dishes so a well designed optical LOS transceiver should pass also.

    Regardless of who or how, the problem will be solved. One day, people will pay only the cost of the transceiver, and they will get free internet, phone, you name it. I told this to my Dad, who owns stock in the phone company and it made him a bit nervous. I told him not to sell yet, but in 20 to 30 years you might not want to be holding stock in any of the companies that sell bandwidth.

  • What about fractal antennas?

    I don't realy want to get into a long technical discussion about antenna design since it is a big subject. But, in summary, any antenna defined by clear physical dimensions tend to be narrow band (for example a dipole which clearly is primarily defined by a single measureable length). To make a good wideband antenna you need to find a structure defined more by angles. At this point I realy need diagrams. I suggest that you stroll over to Tecom's wideband antennas catalogue [tecom-ind.com] and look at some of the pictures (some are sadly hidden by radomes, but there are some reasonable examples).

    For a simple wideband design like a conical log spiral [emctest.com] the lowest frequency is a function of how big you make the big end (limted by cost and obvious size constraints) and the highest frequency depends on how small you can make the pointed bit (limited by power handling, ruggedness and how good your soldering is).

    Fractal antennas are based on the same premisse that if you can't quite pin down how big your antenna is then is must be wideband, but instead of trying to make an antenna defined purely by angles you have lots of dimensions from big to very small and everything inbetween. There is nothing wrong with this concept and it works. The problem is that it doesn't do anything about the basic laws of physics so the benefits you get over more conventional wideband designs are not very great.

    Fractal antennas are not dead and indeed a mate of mine from Loughborough university is presenting a paper at an ESA conference [estec.esa.nl] on this very subject in just a few weeks time.

  • Wow, I spent ages trying to think of some way of expressing those concepts sufficiently succinctly to get it onto a SlashDot post, and failed.

    Well done!

  • ...in the real world might not the methodology of the article produce a better responsiveness for the same bandwidth?

    Well, it is certainly true that the simple frequency division approach is rather inefficient with "fits and starts" signals. So it is probably true that this pulse scheme would be an improvement in that case. However, TDMA and CDMA also provide this advantage which is one reason why they are used in the very "fits and starts" world of cellphones. It is not clear to me how much of an advantage (if any) an UWB solution would provide over these. Answering that question will require a lot of detailed system analysis.

  • Time for some basic education (Score:5)

    by HuskyDog ( 143220 ) on Friday May 04, 2001 @10:25AM (#245263) Homepage
    The work reported here is certainly interesting, but I simply cannot agree that it opens up a whole new area of previously unused spectrum. Let me explain why.

    The bottom line is that there is a fundamental limit to how much data can pass through a given amount of radio spectrum in a given period of time. Let's say you have 100 MHz of bandwidth available and 100 signal sources each generating enough data to fill 1 MHz of bandwidth. Now, we can divy up the spectrum in several different ways.

    The obvious one is to place each source on a different frequency and since each one is 1 MHz wide we can just fit them all in. This is called frequency division multiplexing and its how FM radio works.

    Next, we could get each signal to buffer 1 second of data and then transmit it at 100 times normal speed for 1/100th of a second. Increasing the data rate increases the bandwidth so each signal is now 100 MHz wide and uses the entire available spectrum, but if we get them to transmit one after the other we get 100 bursts each 1/100th second long and once again the channel is full. This is called time division multi access (TDMA) and it is how your GSM cellphone works.

    Lastly, we can take each signal and mix it with a very high speed pseudo random data stream. The signal gets much wider, but the energy per unit bandwidth goes down. At the other end the received signal is mixed with the same random stream and the original signal re-appears. The maths is complex, but it turns out that if I spread each signal out to fill the 100 MHz then I can just about get them all in without them interfering with each other too much. This is called code division multiple access (CDMA) and is how GPS and 3rd generation cellphones work.

    Now for the important bit. As you can see, we have 3 different ways to get 100 signals into out 100 MHz of bandwidth. We can choose any one of them, but we can't choose all three. If our 100MHz is already full of TDMA signals and we put 100 CDMA signals in the same spectrum then they will interfere.

    Now, if I have a bandwidth which isn't being fully utilised (suppose for example I have only 10 frequency multilexed signals in my 100 MHz) then I can slip in a few CDMA signals and all will be well, but only a few. If I bung 50 in then the FM signals will start to suffer interference.

    What Mr Fullerton has done is basically produced a modified version of CDMA which is spread over a very wide band. The radio spectrum is still not used very efficiently, so there is enough room over that huge bandwidth for quite a few of his transmitters to be fitted in. Provided that their average power is low and there not too close to your receiver. But you don't get something for nothing. He's not using new spectrum, just old spectrum in a different way. We could achieve exactly the same effect by taking conventional CDMA and just making the random bit stream much faster.

    Finally, there are costs to his approach which is related to the huge bandwidth and that is making an effecient wideband antenna. As someone who designs antennas for a living I can tell you that making a wideband antenna which is efficient means sacrificing gain or size or both.

  • A lot of people have been pointing out that
    the ultrawideband idea doesn't open up
    more spectrum. Therefore it doesn't give
    you better capacity, speed, etc. From what
    I have heard, the main benefit of UWB is
    supposed to be that the transmitters can
    be made dirt cheap.

    The UWB transmitters can be made by a spark gap. They basically spit out a really short pulse (i.e. a spark). This means you don't need fancy linear amplifiers. For conventional transmitters, CDMA, TDMA, FDMA, whatever you need to burn a lot of power do to circuit inefficiencies.

    However, you don't hear Time-Domain talking about his advantage much, so maybe it doesn't exist...

    Another issue people haven't talked about is channel estimation. If you blast your signal over an enormous range of the specturm it is going to get distored. Generally you compensate for this distortion with an equalizer. However, you have to somehow learn what the channel is in order to equalize it. Now if you are transmitting over a reasonably narrow range this is possible. It might be more difficult if you are transmitting over an enormous amount of spectrum.

    I'm not saying UWB won't work. I'm just pointing out that many of its potential benefits and drawbacks don't seem to be addressed in the article or in the papers I've read. Does anynone know of references which discuss these issues?
  • There for minute i thought you were describing .NET!
  • Basically you have to hunt around for it a bit.

    Its a bit like tuning a radio except you are adjusting the phase not the frequency.

    Typically, you have the receiver have an oscillator that is running just a bit slower or faster (just a bit).

    That will mean that the phase of the receiver will gradually slide relative to the transmitter. When they finally line up, the filter will light up, then you can adjust the oscillator to match the transmitter speed.
  • IBM is one of Time Domain's chip suppliers. Here's the news release [ibm.com] from IBM.
  • Last fall I spent some time looking into getting some UWB hardware for testing, and it was an interesting experience. I called Time Domain [timedomain.com] several times and was routed to a different person's voice mail each time. None of them returned my calls. At one point, when requesting to be transferred to sales, I was asked why I wanted to talk to someone in sales (apparently it wasn't obvious). When I responded that I wanted information on product availability, I was told that there weren't any products available. I guess they weren't too concerned about the possibility that someone might be interested in products that they might somehow have available in the future...

    Aether Wire [aetherwire.com] on the other hand was the exact opposite in my experience. Not only was I able to talk to someone, but that someone was one of the head engineers. At the time they were planning on having prototype kits available for testing this summer, but I don't know if that date has changed. Aether Wire is focusing on positioning and tracking applications of UWB, and these products could have many important uses for warehouses, hospitals, emergency services, and the military. That is, if the products ever exist and live up to everyone's claims.

  • Since the /. Effect is Alive and Well... (Score:3)

    by whizzard ( 177251 ) on Friday May 04, 2001 @09:21AM (#245269) Homepage
    ...here [google.com]'s the google cache'd link.
  • Instead, you enter the "key" which represents the pattern of the time coded signal and your receiver will then be able to decode the transmission being sent. That, or I don't know what I'm talking about. ;)

  • I'm honestly wondering what applications this might have if, instead of transmitting over the "air" it were to be used as a communications protocol over fiber, coax, etc.

    Could this be a means to come up with a super high capacity firewire, usb, SCSI, etc.? Multiple "devices" could listen on the same "cable" for the data intended for it.

    (Of course, with its intended application for ultra-high resolution of objects through walls (even better than radar), its most likely first profitable application will be for pr0n.)

  • ...this has a huge potential for wireless communications. Because a chip is needed on both ends of the signal, it's possible with this to send and receive digital data securely between two devices (laptop computers, mobile phones, etc). Since up to 40 million digital pulses can be transmitted per second, this sort of secure communication would be essentially uncrackable.

  • You see, it's this type of exploration of the strange that results in something quite useful...

    What I am curious about is the size of the chips and board and case required for the transmitter and receiver. The article mentions using this to find lost keys (by attaching a transmitter I assume) so I take it that it would be quite small.

    Either way, it's hats off to the dude - I always hated it when my teachers told me something was impossible - time to dust off that old time machine in the basement...

  • when you have to deal with a femptosecond clock? (or worse..)

    A lot of production hassles will have to be worked out before these kind of devices become easily mass produced. EEs go through great pains to avoid building anything that will require setting exact frequencies or tuning servo or PLL-style feedback loops. Doing such things in quantity with current mass-produced component tolerances will be difficult. Not that it cannot be worked out, but it adds to the cost.

    Second, this sort of tech works really good when you are the only one on the block. Get everybody and all their electronic fodlops to have one of these, and the noise floor goes up. Which means the reliability goes down dramatically.

  • Meant to post this to the discussion on genetics... came back and clicked on the wrong thread grrr!

    Forgive me, and mod the above down.

    (sob!) My precious Karma... =P

    -Kasreyn
  • It doesn't matter how you share the spectrum - using narrow frequency bands or using uniquely shaped pulses. Amount of information that can be transferred over the same part of the spectrum will be the same. You can try to transfer more, but all you will see is more interference and higher Bit Error Rate.

    Code division/ time division (the invention is an extreme case of time division) actually has some advantages over frequency division, but only under moderate to above moderate saturation of the spectrum. Under high saturation, the data rate and error rate degrade even faster than with frequency division.

    Sorry, no miracles here. This area is studied very well.

    Also worth mentioning, that pulse transmitters will cause a lot of interference with existing systems. They will only live well with each other, but not with narrowband systems. They will give a lot of annoying clicks in your FM radio.

  • Location, Location, Location (Score:5)

    by RangerSpeedBumpp ( 250645 ) on Friday May 04, 2001 @09:21AM (#245284) Homepage
    I can't get through to Discover's website, but if this is the Aetherwire company, it's pretty amazing stuff. The biggest wow-factor is that the protocol allows you to determine location. Since all the pulses are very precisely timed, you can tell where each transponder is located by the small time shifts that result from distances. (If you think this is farfetched, realize that police LIDAR works in exactly the same fashion.) The network can orient itself in 3D space, and route information based on the physical topology that the hardware discovers itself. Pretty amazing stuff.

    http://www.aetherwire.com/ [aetherwire.com]

  • mlp - micropower impulse radar (Score:4)

    by morcheeba ( 260908 ) on Friday May 04, 2001 @09:44AM (#245285) Journal
    I've been watching this subject for a couple of years now... The first link I found was at Lawrence Livermore National Laboratory's MIR site [llnl.gov], where they have a very similar setup.

    Here are a couple of applications:
    Cheap radar. These things, since they are spread spectrum, don't interfere with each other and are ideal for watching reflected signals (since you are the sender and know the chirping pattern you used to transmit with)
    Complicated imaging. They had some pictures of a larynx -- instead of doing speech recogintion on sound waves, they were doing it by watching the actual parts of the body that move to form the sounds!
    Even more complicated imaging. They had a 3-d radar system to detect reinforcement rods in concrete. Pretty neat

    The above site also has the FCC rules regulating the transmission (since these are single pulses, not repetitive waves, the FCC isn't sure what to do with them) and LLNL's response. One curious thing about ultra-wide radar is that the frequency response of the antennas themselves are the limiting factor on what frequencies actually get transmitted -- so aftermarket antennas might not be so easy to use since they are a vital part of the circuit.

  • Old news (Score:3)

    by leverich ( 264611 ) on Friday May 04, 2001 @09:04AM (#245287)
    This is pretty old news. Posted April 27, 2000:

    http://www.space.com/businesstechnology/technology /pulse_technology_000427.html [space.com]
  • Note the company's website promotes the idea of police using this stuff for through-wall radar. Remember that thermal imagers have led to people getting arrested because their garage gave off more heat than the cops liked when they drove down the street...

    Name a case. The devices have been used to find suspect houses that were then watched and when they had concrete evidence the arrest was made. No one has been arrested for having a hot garage. FUD

    The idea is disturbing enough without resorting to mis-information.

  • Things usually don't appear in Discover first. Discover takes stories that have appeared in the specialist, technical press and presents them for the masses. Scientific American does the same thing, only for more elite masses. 2600 does the same thing, only for more 37337 masses.

  • The geek in me (no-code tech) can't wait to try it myself. Where's the technical details?

    The early adopter in me thinks this is wonderful, and can't wait to buy it. Where's the global-access web-surfing PDAs?

    The pragmatist in me realizes we won't see this for many years. How long have we waited for Bluetooth?

    The Democrat in me thinks this could bring widespread, afordable telecommunications to the masses, and should be encouraged. What's the problem, FCC, why won't you approve this?

    The free-market capitalist in me says this could disrupt the established players in the business and should be watched closely. Who should I buy? Who should I short?

    The Republican in me says this would disrupt the established players in many businesses and must be stopped. Why isn't the FCC working to ban this subversive activity?

    The cynic in me notes who's in charge now [fcc.gov]. Guess who will win?

  • The real value of this technology is military use. These signals have an EXTREMELY low probablility of detection/intercept. I don't know that I see any great commercial advantages here (maybe wireless LANs in urban areas where "ghosting" is a problem), but this stuff will probably be appearing in military gear in the near future.

  • No free lunch (Score:4)

    by dlleigh ( 313922 ) on Friday May 04, 2001 @10:07AM (#245296)
    There's no free lunch with spread spectrum signals. They do not magically provide more information carrying capacity, lower probability of intercept or defense against intentional jamming.

    All modulation techniques use a set of basis vectors to transmit information. The traditional non-spread sprectrum methods use the frequency, amplitude or phase of a sinusoid as a basis. Spread spectrum methods use "weirder" bases, such as pseudo-random codes. The only real differences between the SS methods and the traditional ones are that the SS bases are highly unlikely to look like "common" interfering signals (carriers, impulse noise, etc.) and that they tend to have a wider bandwidth as well.

    "Ultra wideband" is just another spread spectrum technique that uses the time position of pulses as its basis set.

    There is only so much bandwidth out there and the amount of data you can send through it is bounded by the Shannon limit which is a function of bandwidth and noise floor. Any signal you transmit, no matter what modulation technique it uses, will interfere with other signals, at the very least by raising the noise floor, and thus lowering the Shannon limit for that section of the spectrum. It may be possible to demonstrate that a few ultra-wideband signals don't interfere with exisitng transmissions, but that certainly won't be the case for a large number of ultra-wideband signals.

    The one advantage of ultra-widband is that it has very good time resolution because of the short pulses, and since the speed of light is constant, this equates to very good spatial resolution. Lawrence Livermore Labs was demonstrating something called "micropower impulse radar" (MIR) based on these techniques a number of years ago. This was the gadget that lets you "see" through walls. The range was very short, but it did work. This looks like the best application of ultra-wideband.

    I hope the FCC isn't fooled by snake-oil claims of non-interference and unlimited bandwidth. They should not approve ultra-wideband for normal communications use.

  • Great story of invention (Score:5)

    by Chakat ( 320875 ) on Friday May 04, 2001 @09:16AM (#245299) Homepage
    Ever since a professor at the University of Arkansas told him that such [ultra-wideband] pulses could not be used to transmit a clear signal from one antenna to another, Fullerton had been obsessed with proving they could.

    God I love people like this. They're told it can't be done, and they obsess until they develop a solution. These are usually the greatest inventions, IMAO, because they shake up the world, and hopefully make people think a bit farther outside of the box.

  • It doesn't matter how you divide the airwaves in terms of time or frequency, the total bandwidth is constant. Transmissions in the time domain may look like they utilize bandwidth better, but that's an illusion. Time domain transmissions simply spread their data through currently "unused" frequency bands, but those bands are allocated and unused for a reason, and it isn't acceptable to sell and deploy large numbers of transmitters that deliberately add noise to those bands. It's like the bank employee that gets rich by stealing a penny from every account. Existing spread spectrum technologies, on the other hand, aim for taking maximum advantage of the bands they have been allocated without stepping on other allocations.

    If time domain methods caught on widely, we'd have to partition the space of sequences into different "bands" so that particular users are guaranteed a certain amount of bandwidth. In the end, we'd have replaced our frequency based system with a time domain based system and we'd greatly increase the cost and complexity of even the simplest transmitters. That's a nice deal for shareholders of companies pushing that technology, but it isn't for everybody else.

    Maybe eventually, it does make sense to move to time domain systems. But let's make that step deliberately and without believing in the existence of some magical, hitherto undiscovered bandwidth.

  • ...and 25 or 6 to 4?

  • "Picosecond timing is the key to Larry Fullerton's radio pulse technology..."

    Ok, now we need atomic clocks in our cell phones, no danger there.. hehe.
  • Yeah, The FCC will snatch this up as soon as anyone perfects it ! Isnt it lame, but oh so true ? -- Dan DanCclark.com [dancclark.com]

Slashdot Top Deals

Hackers are just a migratory lifeform with a tropism for computers.

Close