Half a croissant, on a plate, with a sign in front of it saying '50c'
h a l f b a k e r y
non-lame halfbakery tagline

idea: add, search, annotate, link, view, overview, recent, by name, random

meta: news, help, about, links, report a problem

account: browse anonymously, or get an account and write.

user:
pass:
register,


                             

Please log in.
Before you can vote, you need to register. Please log in or create an account.

real-time spectrum auctions

bid for bandwidth
 
(0)
  [vote for,
against]

Rather than having the FCC (or whatever local regulatory authority) auction off RF spectrum in large chunks in advance, devise an in-band signaling protocol (which all RF devices are legally required to abide by) which allows anyone to bid on a second-by-second basis for the right to transmit in a particular frequency band within whatever the local range of reception/interference is.

Why? That way deploying RF technology and services in "licensed bands" isn't an all-or-nothing deal requiring massive capital expenses up front (and therefore massive revenue to pay it back). Small companies or individuals can simply pay as they go. Service providers and equipment manufacturers build rules into their equipment to come up with bidding strategy, possibly under the user's control (if they really want their cellphone call to go through, they can choose to pay more and bid higher for the airwaves). Applications can arise which utilize otherwise unused (and therefore cheap or free) spectrum for bulk transfer or store-and-forward messaging. Lots of good things are possible.

(Ultrawideband and unlicensed spectrum aren't a cure-all; bandwidth is still finite, and people still step on each others' toes unless they can come to an agreement. This is especially true in the lower areas of the spectrum with more penetration ability, as opposed to the microwave regions currently available for unlicensed use.)

This has probably been proposed elsewhere by some techno-Libertarian dreamer -- but where?

egnor, Nov 22 2002

An article touching on "open spectrum" or bandwidth commons http://story.news.y...te/the_new_spectrum
[DrCurry]

Spectrum allocation study http://www-swiss.ai...student-papers.html
From an MIT/Harvard Law class I took [DrCurry, Oct 04 2004, last modified Oct 21 2004]

Spectrum allocation study http://www-swiss.ai...student-papers.html
From an MIT/Harvard Law class I took [beland, Oct 04 2004]

Spectrum allocation presentation http://www-swiss.ai...opic5-spectrum.html
Transcript of an oral presentation of the same study [beland, Oct 04 2004, last modified Oct 21 2004]

[link]






       Welcome back, you've been missed, regularly
In the long run, do you think Bid For Bandwidth may have the effect of driving up prices?
thumbwax, Nov 22 2002
  

       Well, I doubt it would drive up prices so much as make them a lot more volatile. Right now the effect of everyone turning on their cell phone at once is half the people don't get through; with B-f-B the effect would be that making a call would get really expensive (for the caller, for the service provider, for whoever the charges get passed onto).   

       I think it would be more efficient in the end, though it would hardly be a trivial thing to implement.   

       It's possible that by opening up new opportunities for communication technology, it would end up creating more demand which would drive prices up, but I think that's a good problem to have. It would also self-regulate, since the high prices would encourage people to find creative ways to reduce bandwidth usage and shift traffic to underutilized spectrum.
egnor, Nov 22 2002
  

       A lot of people have been talking lately about removing the arbitrary divisions and opening up the spectrum up to one and all. This obviously entails real-time negotiations over who uses what, but on a much faster basis than second by second. Can any bidding system accommodate millisecond frequency changing?
DrCurry, Nov 23 2002
  

       They're probably thinking about spread spectrum techniques, which don't usually involve negotiation (though collision avoidance is possible), and certainly no monetary precedence.
egnor, Nov 23 2002
  

       Who sets the base price?
Who collects the payments?
What do they charge for that service?
What stops me from using a bunch of transmitters to tie up all available bandwidth - forcing you to negotiate with *me* for access?
(_!_) Isn't bandwidth technically unlimited? That is, limited only by the ability of the transmitter/receiver to resolve the frequency of operation?
phoenix, Nov 25 2002
  

       The base price is $0.
The government collects the payments.
They charge nothing for the service.
  

       The same thing that stops you from bidding on half of eBay -- if nobody bids against you, you "win" and have to pay a bunch of money. In this case it's even worse because what you've "won" is not a good you can resell but airtime that you've just squandered. Perhaps you can try to resell that airtime at a profit but if so that means someone is paying more than you bid and they'd be better off just bidding against you directly.   

       Bandwidth within a particular frequency range is far from technically unlimited. Information theory, etc..
egnor, Nov 26 2002
  

       At them moment we pay for "blocks" of bandwidth on a time-domain basis (3.6 kHz voice channel over cellular radio, 64 Kbits ISDN line) so in a way this is not much different. It's moving towards the idea behind GPRS where you pay by the volume of data transferred, and the system gets it to you as fast as it can, using empty "slots" in the circuit switched data channels.   

       I think this is not only a good idea but also practicable, and more deserving of an RFC than a mere croissant (which is what I'm giving it).
8th of 7, Nov 26 2002
  

       "They charge nothing for the service."
WIBNI
"The same thing that stops you from bidding on half of eBay..."
Oh, come on now. Someone will try to corner the market in any commodity - can you honestly compete against the Bill Gates' of the world? Ol' Bill would buy it all up just to keep Oracle, AOL, Sun, etc from being able to. Then he'd turn around and sell it back to you as Microsoft AirTime '03 for twice what it's worth. That's why the airwaves belong to the public in the first place and are held in trust by the government.
phoenix, Nov 26 2002
  

       "They charge nothing for the service. - WIBNI"   

       The government is not providing a service in the first place, so how can they charge a fee for it? (This would be a tax, if charged.)
dalek, Nov 26 2002
  

       [phoenix], the government doesn't charge for all sorts of services. Normally tax money pays for them. But this is actually a money-maker for the government; the price you pay for using the airwaves is actually paid to the Feds, who should easily make enough to do whatever limited management and enforcement is necessary.   

       You bring up the issue of monopoly control, but that's hardly specific to RF spectrum -- you could just as easily argue that milk should belong to the public and be held in trust by the government because otherwise Bill Gates will buy up all the milk and then sell it at an expensive price. Monopoly regulation is necessary but totally orthogonal.
egnor, Nov 26 2002
  

       I think it would be beneficial to experiment with something like this for certain applications, in certain spectral ranges. Provisions should be made public interest, and for situations and applications for which bidding is not an appropriate distribution strategy. The technology should be designed to be minimally cumbersome, and the market should be designed to avoid catastrophic failures (like California's energy de-regulation scheme).   

       For example, cell phones seem like a good application for this kind of experiment. Cell phone networks should be able to handle normal traffic; in fact, a minimum standard of service could be set by regulators - say, at least 95% of the time. When too many people want to use the phone at the same time, say, when a disaster strikes, or there's a major news event, or a particular circuit is overloaded because it's Mother's Day, or whatever, then the bidding system kicks in. To keep things simple, I'm imagining a one-step bidding process. After you finish dialing, you get a message that there's too much traffic. You have the option of either a.) hanging up b.) pressing 1 to have your phone ring when the traffic has cleared up, or c.) pressing 3 to pay $x.xx for 3 minutes of Priority airtime. (You could have periodic announcements - "You have one minute remaining. If you stay on the line, you will be billed an additional $y.yy for the next 3 minutes." - but that's starting to get complicated.) The system might start $x.xx at say, $0.25, but raise it further if lots of callers were making Priority calls. 911 calls would always be granted Top Priority, without charge. The government could also give special phones with Top Priority capability to key officials and safety crews, for use in case primary communications were disabled. (I believe that's already being done in some places.)   

       As it stands right now, I wouldn't want landline ISPs to use bidding; I like paying a flat rate for unlimited usage. It's OK for performance to gracefully degrade slightly when things are unexpectedly busy, but if it happens a lot, I expect my ISP to upgrade its equipment with the money from my monthly dues. (Life-critical applications should have alternatives to the Internet.) Wireless Internet access is a different matter, since one can't upgrade the laws of physics to expand available spectrum. I'm not sure what the user interface on that should be, though.   

       TV and radio broadcast spectrum probably shouldn't be subject to this free-market experiment - certainly not right away. (Larry Lessig once joked about PBS breaking into an emergency pledge break to keep Sesame Street on the air in conditions of heavy traffic.) I'm not sure how broadcast licenses are currently allocated, but they are probably best negotiated on a long-term basis (perhaps a year or more). TV and radio spectrum is actually supposed to be used more efficiently after what I believe is Congressionally-mandated conversion to a digital system. But license allocation is more of a public policy question than a technical one, and the information distribution marketplace is undergoing too many changes for me to predict what we will need broadcast spectrum for 20 years down the road.
beland, May 26 2003
  
      
[annotate]
  


 

back: main index

business  computer  culture  fashion  food  halfbakery  home  other  product  public  science  sport  vehicle