Follow Slashdot blog updates by subscribing to our blog RSS feed


Forgot your password?

VMSK/2 Promises 5 Times More Bandwidth 124

ksan writes "Acording to this article in EDN Magazine; VMSK/2, a new modulation technique may improve modem, FM, AM and other types of transmission. They say that its possible to transmit 100 channels of 128kbps MP3 over an FM channel. Anyone can say more about this?"Read below to find out the *major* problems with this article.
This discussion has been archived. No new comments can be posted.

VMSK/2 Promisses 5 Times More Bandwidth

Comments Filter:
  • Cancel those DSL orders! I wonder how the new technology will compare to other technologies, pricewise and feasibility-wise. I hope it can give DSL a run for the money. DSL is a great technology, but this sounds pretty cool. Things just keep getting better. I guess it'll be awhile until the technology will reach consumerville, at least my neighborhood anyway.

  • This is a emf signal, not a sound wave. How fast do you drive?

    Doppler happens with any motion. How do you think the nice traffic officer figures out how fast you're moving? With the precision of this modulation scheme, the little bit of doppler that happens will become significant.
  • by Anonymous Coward
    I work at Alphacom and have talked with several folks working on VMSK/2 and yes it is true it is an improvement. But it's definately not as great of an increase as this article claims.
    After reading this article my impression is the author decided to exaggerate the technology to make his article more exciting. From what I've heard so far, the maximum transfer rates VMSK/2 will achieve is around 256kbit/s in optimal conditions. Most users can expect better then dual channel ISDN, which I think is pretty good for wireless.

    Note: These are my personal views and do not reflect the views of Alphacom Communications.
  • > No one really "owns" the airwaves

    Do you really believe that? If you live in the US, then the airwaves are owned (licensed? bought? stolen?) by a few large media corporations. Infinity, Clear Channel, and AMFM come to mind. Learn more about the situation at [].

    If you really want the airwaves to be free, join us in protest of the NAB (National Association of Broadcasters) on September 20-23 in San Francisco during their convention.

  • I seem to recall that a large part of the secret is phase modulation. (erm. although that may be my star trek fixation talking.)

    I have vague recollections of polar plots (phase and amplitude). Each pulse is a point in this graph. Depending on how good your circuitry is, you can discriminate finer and finer regions. Ig you have 8 regions, you send 3 bits per pulse.

    So if you believe that as a very rough sketch, you can extrapolate that better phase/aplitude detection allows you to discrimnate 2^19 regions. Hrm that sounds like a lot. Maybe not; assume 8 bits amplitude, that leaves you 2^11 bits phase, which you can do at 2048 hz sample rate.

    How they then go to 2^90 regions... that I leave up to the experts. or better bullshit artists.

    Does that make sense to anyone?
  • My understanding of the tech is that it is very similar to HDSL. The basic theory (of DSL) is that the only part of a wave that really matters is the peak or trough. The rest of the wave is just wasted bandwidth. To increase the data rate the DSL modem will produce a wave up to the peak/trough then stop that wave and go right into the next wave. The effect it produces on an o-scope looks like high-speed chirps.

    The next stratagey they use is to increace the combination of bianary digits that a wave can represent. Instead of letting a peak/trough represent a bianry 1/0, they use a "bullseye" to let the peak/trough send multiple bits.

    Since /. dislikes whitespace, i'll give you an example using amplitude/phase readouts

    0 volts amplitude and 0 degrees of phase shift would be binary 0000
    .1v and 0d is 0001b
    .2v and 0d is 0010b
    .1v and 1d is 0011b

    to help visualize this think about each wave as a seperate entity and not part of the carrier and then place the voltages and phases on a sheet of graph paper.

    By allowing each wave to represent a byte/word vice a single bit, you pump up the datarate. This is why we refer to modems in kb/s and not baud anymore.

    If this is what they are doing, they are just modifying the current algorithms that everyone uses for standard modems.
  • by Anonymous Coward
    I read the EDN article yesterday and it has a number of factual errors. For example, it mentions that a 56kbps modem can deliver 19 bits/Hz which is just not true. The author got that number by dividing 56kbps by the 3kHz "bandwidth" of a voice line. Of course, 56kbps modems only do analog on your local loop and use more than 3kHz of bandwidth. That pair going to the central office can support much more than voice bandwidth (which is why DSL works so well) and the 56kbps modems use that fact as well as a characterization of the D/A and A/D converters at the CO. The real number is closer to 14 bps/Hz. Beyond your local loop the signal is digital all the way.

    The amount of information that you can transmit through a channel is dictated by the Shannon limit:

    D B * log2(SNR + 1)

    Where D is the data rate in bits per second, B is the bandwidth in Hz and SNR is the signal to noise ratio of the channel (assuming Gaussian noise). In order to get 90 bps/Hz, you would need a channel SNR of over 270 decibels! Exponentials are ugly that way. For those not in the know, that's HUGE.

    There's no avoiding noise in commuications channels: thermal noise abounds and in a radio link is not going to go below the 3 Kelvin cosmic microwave background radiation (in completely empty space). In order to get the 270dB signal to noise ratio with a 3 Kelvin noise floor, you would have to transmit with 42 kilowatts of power per Hertz of bandwidth. The whole thing doesn't seem very efficient to me. To compare, at 14 bps/Hz you can get away with an SNR of about 42dB, which would only require you to transmit about 10^-17 watts of power per Hz to compete with the cosmic microwave background.

    Of course in a real application, your noise floor will be much more than the 3 Kelvin CMBR...

  • Here's [] an interesting page which describes VMSK technology in a semi-technical manner.
  • I understand that nudging the duty-cycle would cause spreading in the frequency domain (at least, I think I do) but didn't the article say that the signal is then run through some narrow bandpass filters or something? So then wouldn't the nudging sort of degrade to a phase shift or something?

    My point was that this filtering would lose information. Your available bandwidth influences how fine and how fast a phase shift you can detect.

    Good question, though.
  • I don't really understand this stuff (not an EE), so I'm asking this not too say that you are wrong but for clarification.

    I understand that nudging the duty-cycle would cause spreading in the frequency domain (at least, I think I do) but didn't the article say that the signal is then run through some narrow bandpass filters or something? So then wouldn't the nudging sort of degrade to a phase shift or something? Oh, man...freakin' electronical things, hurtin' my brainez.


  • And then there are those people who live in areas with no other bandwidth alternitive except $600/month 128k ISDN..
  • This is probably a really really dumb question (to someone) but every time I hear about some new advance in wireless networking (which assumes 2-way communications) I start wondering about this:

    How much wireless stuff can be crammed into a given volume of physical space before the crosstalk and interference disrupts everything?

    I visualize the problem as this: radio is really just another form of light, so each wireless device is the electromagnetic equivelent of a 3-watt light bulb. The color of the light is its frequency (or "channel", if you prefer.)

    Imagine the average suburban neighborhood, but where all the houses are made of glass (to allow the light metaphor to work - radio passes through walls) Each house now has a couple of thousand of these colored lights stuck in it (all my appliences, etc. are networked), shining all over the place. The neighbourhood is awash in light! How are these devices supposed to function in this kind of environment?

    How well does the aether scale?
  • by dbateman ( 150302 ) on Tuesday August 22, 2000 @08:02AM (#837309)
    I've read the article and even visited the referenced websites and read the documentation there too. On the page [] they stated that if they modulate their carrier with a modulation depth of less that 0.25radians they effectively have a pure spectral line. I can't this as anything but a bit a black magic.

    We can write the modulated signal as

    v(t) = A * exp(j * m * (d(t) - 0.5))

    where d(t) is a binary data signal. If m=Pi/2 then we have a BPSK style modulation. Make m small and we get a VMSK signal. This small value of m is what gives them a proportionally smaller occupation of the spectrum for the bit rate.

    NoW consider the constellation of this modulation in the complex plane, we have two points representing the bits 0 and 1. For BPSK they are seperated by Pi radians, while with VMSK they are much closer together. VMSK is therefore clearly more vunerable to noise than a BPSK scheme. At the above website they clearly state that the noise resistance (C/I) is significantly better with VMSK!!!

    It seems clear that there is a basic fault in their reasoning, and they can't beat Shannon in this manner. D.

  • Unfortunately, you're forgetting that FM is broadcast, and you don't have the power to broadcast the signal from your laptop (receive only)

    Sure you do. Cellphones broadcast relatively strong FM signals, so why can't laptops? Of course it would require a complex network of antennas (perhaps use existing cell towers?).

  • FM radio has a 200 kHz bandwidth. You need some seperation between channels so assume that each FM channel has a useable bandwidth of 150 kHz (as per the article). As for FM audio it only uses about 56 kHz. There is a subband of about 20 kHz that most FM stations lease out to transmit things like shopping mail music. There is a whole lot more bandwidth, but today's FM stations are not equipped to use it.

    FM stereo uses 0 to 18 kHz for Left+Right, has a stereo pilot signal at 19 kHz, and has Left-Right centered around 38 kHz (20 to 56 kHz). I believe the Left-Right audio is double sideband modulated and the stereo pilot signal is frequency doubled to be used as its carrier. I can't remember where the subband is, but it somewhere above 56 kHz. I believe that the subband is either AM or double side band modulated (ie AM carrier surpressed). Yes the FM radio spec wastes a lot of bandwidth.

    This is all from memory. Please correct me if I am wrong.

  • Could it be deployed in a cellular-like fashion, so when I drive cross-country, I could access my mp3s on my machine back home?

    Great idea, but due to the possibility of someone intercepting the data, you'd have the RIAA on your ass :\

    Could this be deployed in a car-to-car fashion?

    Again, an interesting idea to most (including myself), but probably not to the RIAA. Incidentally, will this be a "tune-in-and-listen-to-what-I'm-listening-to" sort of thing or will it be a "listen-to-whatever-I-have-that-you-want" sort of thing? The second one could get rather complicated, especially with the driver trying to simultaneously drive and page through a list of MP3s trying to decide which one to listen to. It's hard enough trying to find a radio station to listen to :P

  • They claim to be able to fit 100 128K bit/sec streams into an FM radio channel. The bandwidth of an FM channel is 100 KHz, not 200 KHz as stated in the article. Nevertheless, we'll give them the benefit of the doubt.

    The information-carrying capacity (C) of a channel is a function of its bandwidth B and signal (S) to noise (N) ratio. The formula (this is elementary information theory) is

    C=B log2(1+S/N)

    We can plug in the given values for B (200 KHz) and C (12.8M bit/sec) and determine the required signal to noise ratio:

    12.8e6=2e5 log2(1+S/N)


    S/N=2^64 - 1

    Each bit of signal-to-noise ratio corresponds to 3 decibels (6 dB if you're talking about power, which I won't, just to be charitable), so the required S/N ratio is around 190 dB.

    Is this achievable? In a word, no. It's not even close. They're off by more than 100 dB.

    When these guys claim that `when properly implemented and used under the right conditions, the digital modulation scheme reportedly delivers 90 bps/Hz-over the air' they're blowing smoke. They're saying they need a 270 dB S/N ratio. (That's larger than my number because I gave them some extra slack.) If they could get it, their scheme would work, but they can't, so it won't.

  • There is one main reason why radio networking will not really catch on for big businesses, and that is because of security. With a regular (ethernet for instance) network, there is one fundamental thing a potential 'hacker'(or whatever bad movies are calling them nowadays) needs, which is physical access to the network. Sounds stupid, but its the first obstacle someone has to get past to gain access to the network, which can be quite difficult (believe me :-)). In theory, you could just sit outside the building and cause havoc on the system, and a paranoid clueless sysadmin will probably be shit scared of this.

    I doubt it will take off.

  • I've looked for any decent links on this and most of the pages read like alien abduction sites, all fluff and hype. I found only one university with info on it and it was in China. Anyone care to translate? jt2000/0002/000225.htm []
  • dr. dopler's strange effect works for waves in general, not just the audio band.
  • 70 dBm is 70 dB (=10000000 times) above 1 mW = 10 W.
    70 dBW is 70dB above 1 W = 10kW.

    I think the math is OK ...
  • There's clearly something wrong here (see amusing calculations below). So, we go look at their web page. Wait a minute! Their pictures [] use square waves, not sine waves. They're actually broadcasting on all frequencies, and are misrepresenting their bandwidth. Problem solved. The real data rates may be what they claim -- they just use way more bandwidth than they claim.

    The critical test, which they do not report, is to send and receive data through a narrow bandpass filter that eliminates the high-frequency components of the square-wave signal.

    Amusing calculations:

    The photon shot noise of a 1W, 100 MHz signal implies a maximum SNR of about 10^12.

    SNR = Sqrt[Power/h*BW], h=6*10^-34 SI

    A more reasonable microwatt radio signal has a maximum SNR of 10^9 and, with 100 MHz bandwidth, a throughput limit of 6 Mbps.

    To get the SNR of 10^27 they claim as maximum (see below) you'd need a star-powered transmitter. And our sun probably isn't big enough.

    To get a SNR of 10^19 you need something like 10^13 watts of receiver power. A nuclear plant wouldn't work. So, clearly either we've misinterpreted the research, they're full of crap, or they're on to something new.

  • See for yourself []. My opinion? If it sounds too god to be true....
  • as an executive member of my college radio station, I'd argue the fact that there isn't any good music. If anything, there is MORE good music now than ever before. Thanks to technology and decreasing costs of really good equipment, almost everyone can make an album that sounds as good as one produced by WB or the other major labels.

    Therefor, the question isn't that there isn't good music, just that no commerical top 20(which all commercial radio stations are now, by the way...) radio station is willing to play them. They are only willing to play something "safe," something that falls clearly within the genre, something that sounds similiar to another band that has proven sucessful.

    If you want to hear the good music, now you have to turn to college radio. One I would suggest is the U of Minnesota's Radio K--they broadcast via realaudio, and you can check them out here []

    mike kohout

  • &gt 70 dBm is 70 dB (=10000000 times) above 1 mW = 10 W.
    &gt 70 dBW is 70dB above 1 W = 10kW.

    Gee, where was I when I wrote that ?
    Yes, 70dBm is 10kW, but it only makes the case more clear. By using VMSK/2 modulation you use a lot of power to conserve bandwidth. This can be advantageous in some situations, but clearly *not* when you need high baud rates
  • FM-radio channel has generally a 200khz bandwidth(in u.s.) reserved to it. It is going to use lot less, but this is to prevent interference between stations.

    Yeah, I didn't count that, since it's not actually carrying data. In fact, the FCC doesn't like to license stations less than 500kHz apart. 'Course, New York and L.A. are big markets...

    Stereo pilot is at 19khz if i'm not terribly mistaken.

    Yeah... I've been thinking more of MTS stereo on TV, which is basically just the same thing as FM stereo with a different pilot frequency.

    All frequencies higher than 18khz or so are filtered out(so that they don't mess with stereo carrier). 40khz bandwidth is a good estimate..

    Well, remember, they're not filtered out, they're modulated into the additive and difference signals. The (L+R) signal that you hear on a mono FM radio, then the pilot which turns on the FM stereo circuits, then the (L-R) difference signal which is used to create the stereo image. All that space is available under such a scheme.

    Increased bandwidth does have a bigger impact, however. Would probably want to use frequency bands over 1ghz where a bandwidth of 500khz-1mhz is obtainable..

    Why, when you can basically run a dozen 56k modems slingshotted through the same given bandwidth? Use the bandwidth efficiently, and you don't actually need the 500-1,000kHz bandwidth.

    Besides, UHF/microwave sucks. It's way too position dependant. I don't want a Walkman that has to be moved three feet to the left before I'm able to receive the Howard Stern show again. Since UHF/microwave tend to bounce off lots of things, reflections and self-cancelling signals are going to be a constant pain in the ass.

  • Hemos wrote:
    VMSK/2 Promisses 5 Times More Bandwidth

    ...And 50% more "s"s.
  • Reading the article it sounds like this is a very good technology, however what will be the cost if this is adopted. MP3 data over the air would be nice but what kind of equipment would I need to buy to play them. Thinking of how much the consumer would need to buy to play them, and also how much would stations need to pay to broadcast these signals.

    A second thing that I was concerned with is how much would interference affect these signals. The article mentions interference should be low since it is such a narrow channel. I was thinking of a weak signal and how would it affect the data. Would the loss of data make the signal unusable. Plus in areas where overlapping stations on the same frequency, how would the data react.

    It seems like they are testing this cellular communications and given the authors thoughts on the problems of FM and AM interference, it looks like they plan to use this with cellular service. Maybe radio from your cell phone? It looks this will be a pay service to subscribe, like radio cable. I wonder if it will catch on any faster than that?

  • Spoken word was replaced by books, books were replaced by radio, radio was replaced by video. Two conclusions: TV is evolutionary, radio is dead.

    MTV stopped playing music years ago because there simply isn't enough good music to fill the airwaves. Britney Spears, N'Sync, and everyone else that will be on the FOX Teen Choice Awards tonight are proof that there is something wrong. Regardless of what percentage they make up in the general population, teenage girls seem to have a HUGE influence on the music that is chosen to be "popular". Popular music hasn't appealed to the masses in decades.

    You can have all the bandwidth in the world -- if nobody's interested in what's on, it's all a big waste of time.


  • There's an interesting article in this week's New Scientist about actually making use of multipath reflections to enable more bandwidth to be squeezed out of a given transmission band by simply increasing the number of transmitters and receivers (well, OK, not simply, but you get the gist). Combined with an advanced modulation technique like this...
  • They say that its possible to transmit 100 channels of 128kbps MP3 over an FM channel. Anyone can say more about this?

    Well, the bandwidth of an ordinary PSTN telephone line (not DSL!) is only from about 300Hz to about 3kHz. And in that bandwidth, the practical transmission limit, the current state of the art, seems to be stuck at 56k.

    You can't stream a 128kbps MP3 at 56kbps. Not in real time, as radio implies.

    What's the bandwidth of a conventional FM radio station? 20kHz or so carrier deviation for mono. To light up the stereo light, the stereo pilot must be found, and that runs at about 21kHz carrier deviation, if I recall. A full FM stereo signal takes a carrier wave an modulates it about 44kHz either way, so 40kHz bandwidth is probably a practical conservative estimate and has easy enough numbers to work around.

    Since a 56k modem uses quadrature amplitude modulation on ?4? simultaneous carriers in a total bandwidth of 2.7kHz (3000Hz top end - 300 Hz bottom end), then how many carriers could you stuff into a 40kHz wide data channel?

    I think it's exponentially more.

    This sounds very exciting.

  • For the benefit of us poor slobs who had the unutterably bad taste to attend college 20 years ago, when the best modulation methods going were things like Huffman codes and trellis codes, the notion of sending 19 bps/Hz (much less 90 bps/Hz) is a bit mysterious. Can one of you folks with a modern education post a few URLs where I could see how they are managing to accomplish this? Thanks, from the annoying old fart in the next cube.
  • by JebOfTheForest ( 207893 ) on Tuesday August 22, 2000 @07:13AM (#837329)
    This Shannon guy has hurt technology a lot more than Gates ever could. That freakin' limit has held us back for decades! He's hurt technology more than anyone since Nyquist, or maybe Einstein.

    We need more scientists like Moore, who ensured that chips would continue to get faster and cheaper, and they have. That's progress. That's a good guy.

    All Shannon, Nyquist, and Einstein have done is limit the rates of communication we can attain and bloat our harddrives with 2F-sampled signals (Mp3z, pr0n). What jerks. Much worse than Gates.


  • Instead of broadcasting just MP3's or any other form of digital media over the airwaves would it be feesable to get rid of them all together and use the internet for everything? If this kind of bandwidth is possible could we not change the FM/UHF/VHF bands to internet only communication, and the broadcasters would use the internet instead to broadcast their media. Because the broadcasters would use the internet they would be able to reach everyone in the whole world and not just the people in thier area. Of course this would allow anyone to be a broadcaster which the current broadcasters would likely be against.

    With the broadband that is starting to roll out, houses with fibre or other broadband technologies would be able to use it as the connection, or for people out in the middle of nowhere they could use the wireless technology. Car radios would be internet clients that connect to all the radio stations, TV's would also do the same thing for TV broadcasts, etc.

    I don't know if the bandwidth would be able to handle all of this, but if you could have 100 streams over each frequency I think we'd be okay, especially if they were low power which would allow the frequencies to be used over and over again within a certain distance. I'm sure someone out there will be able to figure it out.
  • by Anonymous Coward
    This refers to the channel capacity. This is the data rate supported by 1 Hz. of channel b/w. Remember that you can "violate" the nyquist rate for a digital signal if you use more than one "symbol," ala QAM, QPSK, etc. Therefore, it's possible to send more than 0.5 hz of informational b/w for every 1 hz. of channel b/w. btw: This is most likely a scam -- as a graduate degreed electrical engineer, with emphasis in information theory (which, strangely enough, is not mentioned in any of the sites claiming the miracle of this "new" modulation technique), I find many of the claims to be without substance. First off, eliminating redundancy from information to be transmitted is nothing new -- most non-lossy compression algorithms operate on this principle (for example, LZW compression used extensively on computers). The problem with eliminating redundancy is that you need to add some "controlled" redundancy back in to the information stream, in order to be able to detect and correct errors (hence convolutional, ECC, CRC-types of coding techniques for instance). I hear no discussion of the error performance of "VMSK" modulation. I also hear claims of "no drop in SNR as bandwidth is reduced." Sorry, guys, but if you're sending any "useful" information on the channel, this is going to require exponentially increasing power. This sounds an awful lot like "perpetual motion machine defeats bulk entropy. Film at 11." --- tjc
  • by Matt_Bennett ( 79107 ) on Tuesday August 22, 2000 @06:35AM (#837332) Homepage Journal
    From my reading of this, this modulation scheme relies heavily on timing- which would really get screwed up if you happened to be moving, because of doppler shifts. Keeping accurate track of timing is very, very hard when moving. Also, they don't really address interference (intentional or incidental) which will make it that much harder to reach those bandwidths.

    VMSK/2 seems to be mostly an academic modulation right now- they need to send it through much more rigorous testing in the real world, with real world components.

    I'm skeptical when people talk about overturning Shannon's limit- people would love to disprove it, but it holds up time and time again.
  • by inburito ( 89603 ) on Tuesday August 22, 2000 @07:14AM (#837333)
    FM-radio channel has generally a 200khz bandwidth(in u.s.) reserved to it. It is going to use lot less, but this is to prevent interference between stations. Stereo pilot is at 19khz if i'm not terribly mistaken. All frequencies higher than 18khz or so are filtered out(so that they don't mess with stereo carrier). 40khz bandwidth is a good estimate..

    Transmission rate is still dependant on the information theory(bandwidth and the more power you put out, the better transmission rates you can expect, square power and rate doubles, double bandwidth and rate doubles). Could still improve the transmission rate of telephone line by improving the s/n-ratio(putting out more power) but benefits would be small compared to the cost..

    FM-broadcasts are much more powerful than your average telephone conversation and s/n-ratio could be made better thus improving transmission rates. Increased bandwidth does have a bigger impact, however. Would probably want to use frequency bands over 1ghz where a bandwidth of 500khz-1mhz is obtainable..

    You can stuff as many carriers you want into a bandwidth but limiting factor is going to be the modulating frequency which is going to appear on the both sides of the carrier(could suppress one). you don't want interference..

  • Well actually since it's per Hz and Hz is per second, it would be 90 bits per second * second, or just 90 bits. But it's more informative to leave it at 90 bits per second per Hz.
  • I don't agree with you that Radio is dead. While it is true that most people don't sit at home listening to the radio anymore as they did in the early days of radio, what about while people are for instance driving? If radio was dead, would car makers still put radios in every car? Would there still be as many radio stations as their are now? I don't think radio is anywhere near dead.
  • by bjb ( 3050 ) on Tuesday August 22, 2000 @08:18AM (#837336) Homepage Journal
    I'm an audiophile. I am quite picky about sound quality and, to some extent, video quality. Basically, I get the heebie-jeebies when I see or hear digital distortion as a result of the technology. For example, DVD and Digital Cable TV, while techincally superior in its delivery capabilities and compactness, sucks in my opinion - digital artifacts are quite visible in "shading" amongst other things. Now, with MP3s so out in the open, people are embracing 128Kbps MP3s as "CD quality". Personally, I think this is a statement made by someone who is no where near the category of Audiophile.

    With this technology, boasting that they could deliver 100 128Kbps MP3 channels probably means that they are probably planning on doing such a thing, maybe even squeezing the compression a little more to get a few more channels out of it.

    While this is all great from a technological standpoint (and probably a business one at that), I see a trend of lower quality broadcasting coming about. While I understand that radio transmission is typically less than perfect (static, power wires, etc), digital artifacts of compression come through even on the clearest of signals. And on a side note, with digital cable, sometimes it almost looks like I'm watching a video screen with an 8-bit color depth.

    I guess my rant is that the people on the delivery side of television and radio are letting their quality standards decrease. They can pack more content into the wires/airwaves by using compression, and this leads to lower quality decoded signals. For the average person, this is not an issue (and you're probably smirking right now). For the audiophile, this is quite a disturbing trend.

    My two cents; no refunds.


  • Thinking of how much the consumer would need to buy to play them...

    That's why we still broadcast TV as chroma/luma, so all the old 1950's B&W units still work. It's the same reason FM stereo is transmitted L+R plus a 1/2 frequency impulse for an additional L-R (or R-L) component. Just filter below the stereo impulse and you get mono for all non-stereo FM receivers.

  • That would be very cool. Of course, you KNOW that the signal will be sent content-scrambled, and that only "licensed" receivers will be able to descramble the signal. Then someone will crack one of the players, and get taken to court by the Digital Radio Copy Control Association.

    I guess that's how the whole dang human comedy keeps perpetuating itself. ok bye.


  • TV uses a timing pulse. Just put a sync pulse on a particular frequency...
  • Multilevel Marketing Organisation.
    Typically ML organisations are more interested
    in building a network of "sales" agents and
    taking their money than in delivering anything
    like they promise.

    Be skeptical.

    See Alphacom Corporate page []
    for their spin.

  • Yes, It would be great, but in other peoples bandwidth challenged countries (like mine ... england), the government owns the air waves. The only form of transmission that you could use is cb stuff. Doh!
    Funny, in my bandwidth challenged country (Scotland), there's already a company that does wireless broadband internet access (check out Atlantic Telecom []) I think they're rolling out in Manchester soon, too. They don't seem to have a problem getting the frequencies to do that with.
  • by geirt ( 55254 ) on Tuesday August 22, 2000 @08:21AM (#837342)
    Nice theory, but this is impossible to implement:

    Do the math:

    Shannon's channel capacity theorem:

    C= B * log2 ( 1 + SNR )

    C = capacity in bps
    B = bandwidth
    SNR = signal to noise ratio

    Solving for SNR in dB ( = 10*log10(SNR) ) gives:

    SNR_db = 10 * log10 (( 2 ^ (C/B)) - 1)

    With C = 12.8 Mbit/s and B = 200 kHz you get SNR = 192 dB !

    To have a SNR = 192 dB, the signal has to be 1.8 * 10^19 times stronger than the noise ! A receiver with 200kHz bandwidth will typically have a noise floor at -120 dBm, so you need more than 70 dBm received signal strength. 70 dBm is 10 W !! And that is the signal strength at the receiving antenna, so the transmitter would have to be in the gigawatt range, to reach short distances.
    This means that VMSK/2 can be used, but you can't reach 12.8 Mbit/s without a nuclear powered transmitter. You can get a decent bit rate with VMSK/2 on battery powered equipment, but you have to design for a few kbit/s, not 12.8 Mbit/s. Nice theory, but ....
  • maybe starting something similiar to a shoutcast/icecast server and recieving that, then you can listen to your mp3's and so can anyone else that may want to!
  • A receiver sensitivity can be given by

    Receiver Sensitivity (dBm) = Required C/I (dB) + Implementation Loss (dB) + KT (dBm/Hz) + 10log(Bandwidth Hz) + Receiver Noise Figure (dB)

    Just the KT term from Boltzmanns constant gives you a value of -174dBm/Hz. Assuming their 10KHz bandwidth, required C/I of 4.9dB as they state at their website and ignoring implementation losses and receiever impartments, we therefore have an ideal receiever sensivity of -129.1dBm. After taking into account losses in filters before the LNA and implementation loss and the noise figure of the receiver is is more likely to be -120dBm.

    Taking your 270dB S/N gives a received power requirement of 150dBm. Take a path loss from the receiver to transmit of a very lower 30dB and we need to transmit 180dbm or 1e15 watts of RF!!!! Star wars eat your heart out!!!!


  • Doppler happens with any motion. How do you think the nice traffic officer figures out how fast you're moving?
    Correct in respect to Doppler but wrong in its application. A radar gun actually does not measure anything but Delta D over Delta T (change of distance over change of time). Granted it does this quickly but it does not use the doppler effect to measure the speed. For doppler to really become an issue I would think that you would really need to be traveling around 1/10000th the speed of light or even 1/100000th for something like this.
  • Yeah i agree, but I find it with things like DVD's especially. I find that because I know about mpeg'ing and how it works, I can see defective parts of the image. I feel that I have spoilt DVD's for my self because i know so damn much! doh! When i ask other people about it they say the image quality's great and they don't know what im on about.

    I think this is one of those few cases where I acually wish I were 'happily dumb'.

  • What? Not a single Asian among the authors?
    It can't possibly be true, they got their math wrong for sure. :)
  • For example, DVD and Digital Cable TV, while techincally superior in its delivery capabilities and compactness, sucks in my opinion - digital artifacts are quite visible in "shading" amongst other things.

    I agree completely. I prefer a Betacam or 3/4" professional videotape, and failing that a good home Beta machine or analog cable feed.

    The Beta machines don't screw up the chrominance quite the same way even the best modern VHS machines do, and neither one of them has the jerky and artificial rendering of a DVD.

    Note that the DVD has *not* been embraced as a professional format: most TV stations are shipping around digital Betacam (almost no compression), analog Betacam/Betacam SP, or burning their own analog video discs.

    Now, with MP3s so out in the open, people are embracing 128Kbps MP3s as "CD quality". Personally, I think this is a statement made by someone who is no where near the category of Audiophile.

    Okay. Here's a bit of an issue.

    I worked for years as a professional sound technician. I've done sound for Garth Brooks, Harry Belafonte and the Three Tenors.

    I've had self-proclaimed audiophiles (who double as lawyers and dentists during the day) ask my why it is that I don't have Monster Cable connecting my stack of Crown amplifiers to the TurboSound bins suspended from the stadium roof. Or, why I don't use vinyl. Or why I don't use sonic CD stabilizers (the green rings) or other similar crap that people who don't know anything like to use.

    Now, not knowing whether or not to class you as that kind of audiophile, or the kind I am, I will defend the MP3 this way: they do sound pretty damned good when they get above 160kbps. In fact, I can't consistantly tell them apart, above 160kbps.

    But, I agree, the majority of people seem to think that the 128kbps MP3 is CD-quality. I'd suggest that the 160kbps MP3 is indistinguishable from CD-quality. But remember, you do live in a world that has embraced VHS over Beta, and where people still buy pre-recorded 1-7/8IPS cassette tapes. Quality is not what people want.

    As for me, I really want to do some more R/D on my old SoundBlaster 16. I'd love to be able to get the D/A converters out of the RF field inside the computer's case. (I've already managed to extricate the analog buffer amplifiers.) Newer sound cards are very hard to hack this way.

    My listening setup is a pair of Acoustic Research AR-4s connected to a Sound A-5000 solid-state amplifier. The preamp stages of the Sound A-5000 have been replaced with a 12AT7 preamp board that I designed. Surface-mounted, ground-plane, fine quality components, DC heaters for the tubes, etc. S/N over 98dB on the amplifier. THD is less than my equipment can measure. And, as any real audiophile knows, the only thing on par with a pair of AR-4s is a pair of Celestion Ditton 44s.


  • Good for pirate radio though. Would shortwave be high enough quality?
  • either that or copyright/patent their full names, so they are paralyzed in signing checks, legal documents, etc. Probably cheaper. One-click shopping indeed. IBM patented the indent at the beginning of a paragraphs, I think.


  • I'm as suspicious of Alphacom as the next guy, but I don't agree with the posts that basically amount to, "Shannon's law says it's not possible. Don't even think about it." What if everyone thought that way?

    Physicists are showing it may be possible to send some information faster than the speed of light, which is impossible, mathematically proven I believe.

    An old "law" of physics says that it's impossible to set up magnets passively, with no outside supports, that levitate. Recently, scientists unveiled a magnetic bearing that does just this.

    Didn't engineers say it was physically impossible to have a feature size of less than something like twice what we are using now? Granted, that wasn't a mathematical proof, but they seemed pretty confident about it.

    When Einstein was working on relativity, I'm sure people thought of all this dilation of time as being mathematically impossible, though I don't know what proof that would be violating.

    Quantum physics is chock-full of logical impossibilities.

    The point is, sometimes there are non-obvious ways to circumvent "laws" of nature. I'm not saying Alphacom has done this or anything, I'm just saying that we shouldn't brush off the possibility by saying that it would violate Shannon's law if they did.


  • Even if this is technologically sound, how easy would it be to implement something like this? Digital radio was supposed to be the "next best thing" and for some reason has failed miserably. Why? Because everyone has the same kind of radio, and not everyone can upgrade on a moment's notice. There won't be any people on the new system, so stations will stay on the old system (ratings, remember?). Since there won't be any stations on the new system, everyone will stay on the old and have no reason to upgrade.
  • I like! Now, how stable can we make an FM signal to a computer? This could damn well be another method of providing high-bandwidth internet access, and a good one at that.

    Because No one really "owns" the airwaves, like one does a cable, you could get a lot of competition amongst providers for customers.

    I can see the Laptop/Handheld owners screaming now. I have a palm, and the thought of cheap high bandwidth wireless access for it amuses me.
  • Nice to see this "confession" from someone in the know. It's still not enough, though. He says that the increase is not as big as the article claims when in fact there *cannot* be any increase in performance compared to state-of-the-art techniques like QAM.

    The implementation of VMSK/2, however, is significantly simpler. VMSK/2 can be implemented with an extremely low gate count without quadrature A/D converters, complex digital signal processing and equalization. So if its performance is not too suboptimal it might still be interesting. It would have been much more interesting 10 years ago, though. Today you can put half a million transistors on a chip without blinking.

  • Match this with that power-line broadband, and just think of the potential... heck, we could stream MP3s faster than the speed of light - they could even be ripped and distributed before the artist records them!

  • I wonder how this will affect the coming of Digital Radio, as well as the FCC's new standards to ease the starting of small community oriented radio stations...
  • Most of your concerns are being addressed at this moment by a company name Ibiquity Digital []. I'm not sure why the editor god's don't think it's relevant information. I guess some people have never heard of Lucent Digtial Radio or USA Digital Radio (the two companies which recently joined to create Ibiquity). Designs and working prototypes have been proposed/tested/etc which provide for a clean transition from analog to digital radio via an IBOC (In Band On Channel) design. Oh well, I guess it would be better to just argue academic points instead of acknowledging the reasonable/existing solutions.

    (Note - I happen to work for Ibiquity, so do some searches and feel free to come to your own conclusion.)

    A man destined to have no karma,
  • I'd have to believe that, if confirmed and tested to be viable, it could open up broader uses of existing wireless allocations, as well as help focus some others (e.g. UHF TV frequencies).

    As an avid ham, I've been watching each new commercial wireless application drool over expanded frequency space. Having the ability to expand the efficiency of existing allocations seems like a godsend.

    Tis a shame it's already patented - open source radio, anyone?

  • by Tower ( 37395 )
    Sounds like a breakthrough
    Higher bandwidth radio
    Shortwave Pirates Smile

  • I don't know about where you come from, but I was taught (by my father, and ex Telco engineer) PSTN has a bandwidth of 300 to 3300Hz, a total of 3KHz.

    As for 56k being the limit on that, not really true. That Eris' apple of bandwidth is only achieved by taking advantage of the fact telephone networks are digital to "the last mile".

    They use the 64Kbps channel as 8KHz x 7bits (they drop the lower bit for signalling and other reasons, aparently). Oh, look! 8KHz! But didn't we say 3KHz bandwidth? They're taking advantage of the fact PSTN aint what it used to be.

    Yeh, I know. This isn't really on topic for VMSK/2 (which looks interesting, but time will tell), but I hate to see disinformation. (o8

  • Now we need MP3 radios.

    Founder's Camp []

  • by joshwa ( 24288 ) on Tuesday August 22, 2000 @06:15AM (#837362) Homepage Journal
    Am I the only one that read that as "VMS 2K?" Now there's a scary thought...

  • Very Minimal Shift Key

    Heres [] the link to the "technical information." I still don't think it's anything but a scam. It reeks of it, just look at the ads and scrolling text.

    My shift key is small enough. :P


  • I Think the operative word here is *may*

    I bet this is vapor ware and nothing will ever come of it.
  • This really sounds like a good idea, like getting standard phone lines to have like 256kbit/s as standard, but would it work in practice with a lot of users on same telephone central??
  • by HoovrBass ( 30071 ) on Tuesday August 22, 2000 @06:35AM (#837366)
    I'm HIGHLY doubtful, to say the least, and the article does nothing to give any credence to the claims. A quick search of Compendex yields this article by the research report authors. Our library doesn't carry the publication so I can't check it out. If anybody else has access, I'd really be interested in a review. What is Applied Microwave and Wireless? Is it peer-reviewed?

    Koukourlis, C.S.
    Pliatsikas, J.C.
    Sahalos, J.N.
    Walker, H.R.
    Spectrally efficient biphase modulation FOUND IN:
    Applied Microwave and Wireless v 10 n 4 May 1998. p 74, 76-81

    Publ. year:
    Phase modulated biphase codes which are transmitted single sideband-suppressed carrier at RF frequencies require much less bandwidth without any significant increase in circuit complexity. These codes have an important advantage over other bandwidth efficient modulation methods in that they do not lose bit energy with increasing bandwidth efficiency (compression). Actual measurements confirm this characteristic. 10 Refs.
  • I noticed that too, but I was told not to mention the Dark One, or the sign of the Dragon may be placed on my door... [sorry]
  • Must be marketing,
    Only a certain bit rate,
    My pr0n remains slow.
  • by Christopher Thomas ( 11717 ) on Tuesday August 22, 2000 @06:35AM (#837369)
    From the article:

    You can think of VMSK/2 as a form of duty-cycle modulation (Figure 1). Think of a "square" wave whose total period does not vary but which, depending on whether a given bit interval contains a 1 or a 0, spends slightly more or slightly less than half the period in the high state.

    Problem - this kind of nudging of the duty cycle causes spreading in the frequency domain. In fact, it is these additional frequencies that encode the change in the duty cycle.

    If you try to transmit a signal modulated using this technique through a very narrow channel centered about the carrier frequency, you will lose a lot of the duty cycle information, and your data signal will degrade a *lot*.

    I am skeptical of this getting much more effective use of bandwidth than conventional encoding schemes. The best I can see them getting is a modest gain if this technique is less sensitive to common types of noise (which has yet to be demonstrated).
  • The trouble with the Palm is that you'll still have to get your data compressed at Palm.Net or something like it. I've used a VII and it can be ridiculously slow. Like 5 minutes to download 2k (which had been compressed).

    Unfortunately wireless apps over any distance are so slow that they're barely usable. The Palm VII only gets 8Kb/s.

  • I agree, although I felt the journal article was just about right for my two years rusted EMC and Communications stuff.

    The main thing that I picked up on was the sheer lack of spacing between the transitions of the final data and the original clock. As I read it, this means that the clock recovery circuitry is going to have to be spot on every single cycle.

    Admittedly things may well have changed since I last saw them, but PLLs were notoriously unstable when I played with them. I wonder how they're going to make the receivers sufficiently portable to make this worthwhile in a mobile situation.

    What we probably need is the original StrongARM design crew from DEC and give them something like a DSP32 to play with ;-)
  • PS: I daren't imagine what the constellation diagram would look like for such a creation.
  • the digital modulation scheme reportedly delivers 90 bps/Hz

    Correct me if I'm wrong but doesn't Hz mean "per second". Does this mean that we now have 90 bits per second per second? After 1 second, 90 bits transferred, after 2 seconds 360 bits... Wow, this could be big.

  • Where have all the 20 channel 128kbit mp3 stream FM channels been?
  • by mesocyclone ( 80188 ) on Tuesday August 22, 2000 @07:25AM (#837375) Homepage Journal
    You can achieve very high spectral efficiencies if you use very high power. To send 2 bits/hz, just send the plain old signal single sideband AM (the bandwidth of a 1kbps baseband signal is 500hz). To send 4 bits/hz, you could (as an example), use four different voltage levels. For5 bits/Hz, use 8 levels, etc. Furthermore, you can modulate phase and amplitude independently. Doubling the data rate again.

    There are a myriad of modulation schemes (and related coding schemes) for achieving spectral efficiency. Basically, beyond the simple stuff (filter off the extra sideband, use phase AND amplitude), they achieve that efficiency by encoding data in more subtle aspects of the signal (read: more noise sensitive). This VMSK/2 scheme appears to be one which generates smaller sidebands by modulating the signal less. As such, it requires higher power to achieve it's spectral efficienty (ignore the claims of lower power - that's *per carrier* in the signal, but they use more carriers).

    Note also that increased spectral efficiency is only part of the issue. In the modern cellular world, you need increased efficiency in terms of bits per Hz per square kilometer (i.e. you share the frequencies over an area). A requirement for higher power (which really means a requirement for higher signal-to-noise ratio) reduces the areal sharing that you can achieve.

    Ultimately, you can't beat Shannon's laws. If you can, you can also make perpetual motion machines and free energy (yeah, it's a stretch, but the connection is there).

    Since this company is selling multilevel marketing, I am more than a bit suspicious of any claims. Multilevel marketing schemes are too often fraudulent and based on overblown claims. I am not saying these guys are wrong, just that their approach is suspicious.

    As far as comments on here on FM signal bandwidth... FM stations use a 200kHz wide channel. A stereo signal uses a composite of simple FM for the Left+Right signal, and a subcarrier at 42kHz carrying Left-Right. There is still room left in the spectrum for an additional subcarrier (or more) - which is where you find service such as Muzak. Plain old FM mono is a *spread-spectrum* modulation scheme, in that the RF signal is occupies significantly more bandwidth than the modulating signal.
  • Worse, a merger [] between AMFM Inc. and Clear Channel has been approved... scary, huh? (actually that's kinda old news [late april]... maybe they merged already, I dunno)

    What is VMFK/2 []?

  • Most of the article explained how different this transmission mode was from the standard we use now. Electromagnetics is a really complex and interesting field...

    I can't get to the article (where I'm working this week has a weird proxy set up) but I'm curious -- are they proposing moving closer to what Shannon says the link is capable of, or are they trying to break it?

  • In Germany there is some competition called "Jugend Forscht", which allows young people (school pupils etc.) to present their inventions. This year the 1st price in the "technics" section went to somebody who managed to trick the (academical) jury into believing he was able to expand a GSM network's bandwith to 550kbps, rendering UMTS and the whole 3rd-generation mobile-service useless, just with fiddling around a little bit with the handies. That's not all, two big German news magazines and (at least) one federal TV channel came along and made reports out of it. (see )

    The first time I saw this on TV I did not know whether I should burst in laugh about their stupidity or start to cry about what happened to Germany. However, there is one thing for sure: If I was in the recruitment section of a big company I would hire these guys at once. I suspect the value of such people for marketing purposes is vitally unlimited.

  • by Anonymous Coward
    Alpha Communications is such a scam!
    Notice how on the page they don't even mention what VMSK means? They do on their business scam site: 0710

    If you go to their page you see they avoid discussing their technology and talk all about their business plans.

    I'm surprised these guys aren't advertising snake oil or $4000 a month work at home jobs in the newspapers. Don't be fooled they are scam artists.

    I also saw on this advertisement:

    Business Opportunities
    FOR SALE Increase your current internet connection
    Posted on: 2000-08-11 .
    Increase your current internet connection by thirty times it's current speed! FREE for 5 days Become a distributor (under 10 bucks) Check out the VMSK technology!! Have a GREAT day

    Email :
    Contact: Shelby Carter
    Website: 0710

  • Listen to 128k mp3s with headphones especially when you hear a cymbal playing, it sounds distorted as hell. The minimum would have to be 192k to even be close to FM quality.
  • Its a bit confusing calling that stuff `music`. Is there a better term for the stuff the `music industry` produces so i dont get confused?
  • by edhall ( 10025 ) <> on Tuesday August 22, 2000 @08:38AM (#837382) Homepage

    No, they are actually doing the opposite from that: instead of multiple bits/symbol, they are using one bit/symbol. The bit is initially encoded by a change in duty cycle of a square(-ish) wave. They then modulate a carrier with this signal, remove the carrier and a sideband, finally filtering the resulting sideband with extremely sharp, patented-technology filters.

    I'm real suspicious... the initial square wave would have to be several MHz, and it seems that the subsequent processing would either wind up stripping out the information or result in a multi-MHz bandwidth in the resulting signal. I could be all wrong, but until I saw the math that shows exactly how the bits are encoded and extracted, I'd be skeptical.

  • "Plus in areas where overlapping stations on the same frequency, how would the data react."

    I would imagine that some type of addressing would be used. While the reciever might not be able to send a specific IP address nor would the broadcast station want to send out thousands of channels for every user, but they can still embed a source address in every packet. So you might not tune your radio by frequency like we do now. Mabey by name, like "NYCRocks" or something like that. Then your reciever would only open packets from that sender.

    I have no idea if this is what they plan to do, but it seems like one logical solution. The downside to this type of broadcast is that with it being digital, it could easily become a pay for use system, with only paying subscribers reaping the benefits of such a system. It does promise to save a lot of frequency bandwith however, and if this ever became mainstream, we can all benefit from that.

  • What is Applied Microwave and Wireless? Is it peer-reviewed?
    I've seen it, and remember it as more of an advertiser supported `zine rather than a journal with reviewed articles. The sort of publication to announce your new product in.
  • Correct in respect to Doppler but wrong in its application. A radar gun actually does not measure anything but Delta D over Delta T (change of distance over change of time). Granted it does this quickly but it does not use the doppler effect to measure the speed. For doppler to really become an issue I would think that you would really need to be traveling around 1/10000th the speed of light or even 1/100000th for something like this.

    Sorry, *absolutely* correct in its application. The only sort of police "radar" that uses timing to determine distance is a laser radar gun. The traditional RF police speed radar (known as Doppler radar) mixes a sample of the outgoing signal with a received signal and makes a beat between the two. Generally this beat is in the audio frequency range. By beating the outgoing with the incoming it doesn't really matter how much your main oscillator drifts, because over the microsecond or so that the signal takes to make a round trip it won't have drifted far. This sort of radar (as I have described it) can't distinguish between incoming and outgoing targets, but there are DSP based police doppler radars that have this ability, with a different downconversion technique.

    Time to go back to basic physics- for a speed much less than c the doppler shift of a signal will be a factor of 1/(1 +/- (v/c)) where v is the radial velocity to or from the observer (- when approaching the observer, + when going away).

    For example, with a radar operating at 18GHz, the doppler shift of a car moving at 100 ft/s is about 1.85 kHz. Check out this page [] for more information on doppler. Things do change as you approach the speed of light, but doppler still does matter, and I think it matters in this situation because of the extreme timing accuracy that this modulation requires. If the timing accuracy goes down, so will the bit-rate.
  • The Hz in 90b/s/Hz means 90 bits per second per Hz of bandwidth. So for a channel with a Bandwidth of 20 Khz you get 90*20K = 1.8 Mb/s.
  • As an RF engineer and a broadcaster, I'm used to dealing with similar transmission system in satellites and cable distribution.

    Increasing the symbol rate is not hard to understand. The ideas aren't crazy, it's just a tough idea to implement.

    Like the idea of a fusion powerplant, the idea is fine, but making it happen is another issue.

    The hard tech will be needed in the receivers. Making extremely narrowband filters isn't easy and from what I read, the patents are all locked up. Operating at current FM & TV frequencies lead to all kinds of multipath issues that can be solved, but add more to the reciever to deal with.

    Look for impressive demos where the receiver is the size of refrigerator. Sizing down the technology to fit in a dashboard and cost $20 bucks a unit will be a real challenge. Much money will be sunk in attempting to make these systems small and rugged.

    A better use might be in high end transmission systems. One transponder on a satellite give me 36MHz of bandwidth. Split the bird between two sites for duplex operation and imagine the amount of data I can squirt around the world. Satellite costs are already less than fiber rates for the amount of data they can pass. If you do the rough math you are talking about 3.24GIGAbits of transfer. What OC- is that?

    It's been a while since Linear Systems & Communincation Theory, but I don't doubt the Shannon limit can be adapted to this keying system. I'll leave it to a comm theory expert to comment.
  • Shannon's law [] says that the number of bits per second (bps) you can get down a channel per Hertz of bandwidth equals the log (base 2) of the signal-to-noise ratio. In the article they bandy about numbers like 90 bps/Hz. There's nothing impossible about this, as long as your signal power is 2^90 bigger than the power of the background noise. 2^90 = 1237 trillion trillion. So if you're using VPSK/2 to modulate the world's brightest searchlight on a moonless night in a coal mine, you could probably get this signal-to-noise ratio. But with real radio signals in the real world? Not likely.

    This whole topic is some combination of genuinely good technology, hype, exaggerations by imperfect journalists, and fraud. From where I'm sitting, I don't know which of the above causes is the true explanation. But I doubt Shannon's law has been repealed. Alas...

  • I consider myself a very poor (no money) audiophile.

    That just means it's time to get out the big book on electrical engineering and the soldering iron...

    Older, good-quality stereo equipment is a good place to start. Tubes can be good, but while I like tubes for preamps, you want to keep them away from vibration (microphonics); either way, I recommend a good solid state output stage, because tubes are high-impedance devices, they drive speakers through output transformers, and they're neat inductive devices which cause all sorts of equally neat inductive effects in your frequency response and THD plots. MOSFET amps are really nice since they have a very low output impedance, but if the amplifier isn't very heavy, it probably doesn't have a power transformer to match the current that it should be able to drive. Every three watts of rated power should weigh about 1 pound. Therefore, if the amp says it's 100 watts and weighs 10 pouds, it's probably not really capable of doing 100 watts.

    Car stereos and computer speakers aren't rated in watts. They're rated in "w", the official unit for "wishful thinking".

    Heat your soldering iron up and replace all the potentiometers, all the resistors, all the capacitors, all the switches. Get yourself some nice Sovtek, Mullard or vintage American tubes at a music store or order them from Antique Electronics Supply in Tempe AZ. Find new low-noise versions of the transistors in your amp's output stage. It's cheap, takes only a few bucks and a few hours of your time, and you can put together a great system for very little money.

    Oh yeah, and if the old stereo you choose to hack is *too* perfect, don't do it. Some of these amps have collector value or fan clubs that would want to see a good one preserved . Usually, a beater that doesn't work will work after replaceing the stuff I listed above.

    Speakers? I run AR-4s; I'd also run anything else Acoustic Research made in the 1970s. Anything Celestion, anything Klipsche, anything vintage Bose...

    I go digital to my amp (I still need a good noise gate though)and I find that MP3s at the fixed 128Kb rate have some serious digital artifacts,

    Oh yeah. They're great as background music. In the car, for example, you'd never hear the noise over the sound of the car. But I can barely stand to listen to 128kbps in my listening room.

    And ya know, the digital artifacts are somehow more objectionable than the tape hiss or turntable ruble of days gone by. I guess it's just that the noise that compression introduces has no soothing natural equivalent... at least tape hiss can sound like something comforting and normal, like a waterfall or a leaking barbeque tank...

    Now if I could only get a motherboard that didn't introduce it's own digital noise into the mix I'd be happy, but alas I am poor.

    Just remember that, as long as the status of logic lines isn't changed (ie. 1 stays 1, 0 stays 0) as the data enters the digital to analog converter, you're clear. Digitally introduced noise would cause all sorts of other things, since it wouldn't just affect stuff going to your D/A converter - it would affect the contents of your computer's data and address buses, too: you'd be lucky to even get a BSoD, let alone audio that still played.

    Any noise will be picked up in the analog stages. So, keep the analog stages as far as possible from the digital stages. Use chokes, bypass capacitors and large filters on the power feeding the resistor ladder in the D/A stage and the output buffers that follow. Keep the analog section physically away from the digital section, and keep the analog and digital sides independantly shielded to a good, strong ground.

    It could be done with existing stuff... anyone got a schematic for a first generation SoundBlaster 16? (Older cards look easier to hack since they're less integrated.)

  • You know, that's true. I wouldn't want you to think me a loser so I'll go write that letter. I just hope my letter doesn't get lost in all the NAB contributions.

  • Anybody have a translation of this?

  • Correct me if I'm wrong but doesn't Hz mean "per second". Does this mean that we now have 90 bits per second per second?

    The author is using a shorthand reference to Hz of bandwidth.
  • Um, maybe it's just me, but I thought that Shannon's law had been prove n [] within the bounds of mathematics.

    Secondly, the article gives _no_ information that Shannons law has been broken.

    In fact, it hasn't. All Shannons law says is the relationship between the signal to noise ration, and the size of the channel, to how much data you can put through it.

    In this case, Shannons law allows this to happen. But requires powers of something like (quick mental sums) 280dB.

    That's, like, stupidly high.
  • I read the article, and it sounds like snake oil to me. The claim is the transmission of 1 bit/symbol (1 BPS/baud), but the bandwidth is minuscule. This is contradictory. Further, what I was able to glean from the article seems to claim that the keying is done by tiny phase-shifts in the outgoing signal which affect the timing of the zero crossing. A tiny phase shift would be easily masked by noise, causing a timing error and decoding of the wrong bit. This contradicts the claim of high noise immunity.

    It's possible that EDN got this all wrong, and someone has actually found an end-run around a whole lot of difficult stuff in conventional signalling theory. But I wouldn't bet on it, and if you're smart, you won't either. This may be some scheme to get investment money out of a bunch of suckers, and it behooves you to not be one of them.

  • I wonder what kind of range this system would have? Like current radio? Could it be deployed in a cellular-like fashion, so when I drive cross-country, I could access my mp3s on my machine back home? This could be really cool - Not needing to burn MP3s to CD so I can have car mixes, just fire up a radio connection and network over it. What about using multiple channels for faster throughput? e.g. recieve on 101.3, 101.5, and transmit back on 101.7? Could this be deployed in a car-to-car fashion? Like people who have their HAM callsign on their vanity license plate....
  • No, it is not okay to point out spelling mistakes.

    Your post should be moderated as
    -1, offtoppic
    -1, trolle
    -1, flambat
    or, to save karma from meta-moderation, -1, undderrrattedd

    Have a nice day. ^_^

I am more bored than you could ever possibly be. Go back to work.