Slashdot Mirror
Spectrum as Property
the economist troll writes "An article in this week's Economist argues that overcautious control of electromagnetic spectrum, on the part of regulatory agencies, has resulted in the sheer waste of up to 95% of available spectrum. The article suggests remedies for this sorry state of affairs, including (but not limited to) various methods of privatization. Peppered with history and interesting facts--for instance, did you know only 2% of America's spectrum allocation is determined by auction?--this is one article you won't want to miss."
How is it done in Europe, Africa, Asia, S. America, and in Austrailia? How is it working out for them? I hate to jump to privatization without a prescident.
Since when has this country used intellectual elite as a pejorative term?
As a ham many areas of spectrum are underutilized because the technology does not exist to successfully exploit them. For example the repeater which takes a radio signal coming on one frequency and retransmits it on another is the basis for the entire cell phone industry.
At the time the commercial interests wanted that spectrum for expansion of paging.
What financially driven interests forget frequently is that basic non-directed research is a good thing which yields benefits down the road and often entire new industries.
Like the RFID crowd wants to put high power RFID tags on the 70cm band. This interferes with both Hams, Wind profiling radar and satellite communications. The difference is someone can make a quick buck.
Also these RFID tags can be read at a distance of several miles with the right equipment. So much for RFID being a 'short range' technology
If i am lucky First Post
Spectrum auctions would be a way for the government to make money, without having to raise taxes directly. Something like this:
* There is a valuable (limited) resource that we own in common
* The government parcels out the resource to whoever is willing to pay the most for it
* That money goes "to the people". In reality it goes to the government, who uses it to buy an army, interstate highways, mink farm subsidies, whatever your representatives have put into the budget.
* The buyer makes the money back by selling you something you want (TV, cell phones, garage door openers, etc.)
The fraud problem is also a government problem. It's most easily fixed by demanding the money up front, though that tends to lock small bidders out of it. There are other ways that involve instituting various regulations. Just because the government has been stupid doesn't mean it has to be. (Or maybe it _does_ have to be, in which case the problem becomes insoluble and we're all screwed, and we'll just take guesses because that's the best we can do.)
Now, the point of the article is that spectrum isn't really a limited resource at all. Obviously that's not entirely true, otherwise we'd use just one frequency and we'd all be happy. Certainly the lower frequencies (to a point) are more valuable than the multi-GHz ones, because it travels better. But they claim that technology allows spectrum to do far, far more than we're doing with it. In that case we may not have to auction it at all, not because it's subject to speculation and fraud, but because it's not worth very much.
I think there are proposals to allow WiMAX and the like to tune into unused bands in the TV spectrum. Using ATSC's 8VSB modulation, each channel nets 20mbps, so UHF could have 120mbps or more capacity alone.
Why is formalizing the status quo the cure for the status quo?
Well, if I understand the question correctly, it indeed raises a valid point -- that the political process is a poor way to run anything, even the process of de-politicizing something.
At the time the FCC was founded, a handful of courts were settling disputes between broadcasters by applying a "homesteading" analogy rooted in common law and the concept of first/continuing use of a given freq at a certain power level in a given geographical area.
Revisionist historians have pointed out that, in hindsight, it looks a lot like the creation of the FCC (and the attendant nationalization of the airwaves) might have been pushed into place precisely to forestall that legal trend -- thereby locking the ordinary person and small business owner out of the new field and consolidfating ruling class power.
Now, if privatization is carried out as merely a sham for finalizing this expropriation from early broadcasters (who are, admittedly, almost all surely dead now), then you would be correct to be cynical. It would, however, ALSO remain true that (in general) any private ownership is a better way to manage any resource -- but that overlooks the matter of justice in the face of expropriation.
That brings us to this -- the whole debate ought to be not about whether or not to privatize, but what is the correct way to go about it and why.
-- I could tell right away that she was impressed with my HUGE Slashdot Karma.
then why do I hear two different radio stations on the same frequency so often?
Why does one over-the-air broadcast station have ghosting caused by another??
What, the free market is supposed to fix those problems magically, without government oversight, when they're still pretty bad with the FCC throwing down tons of rules *and* charging licensing fees?
I smell typical Economist free-market hype. Just let the highest bidders control your spectrum, and everything will be fine, kiddies...
I'm not saying there isn't a need for change in the way RF is used. But I am calling into question a highest-bidder-takes-all approach, and the motives of those who back such an approach.
Most people don't understand how television channels are allocated in the USA. Due to interference concerns, stations on the same and adjacent channels must be geographically separated by large distances. For VHF, I've been told that the FCC's rules can be approximated at 160 miles separation between stations on the same channel and 70 miles separation between stations on adjacent channels. The rules for the UHF band are stricter due to the increased susceptibility to interference of television receivers in the UHF band. The end result is that you can't just arbitrarily pack analog stations into a smaller UHF band. Digital stations (ATSC) are more resistant to interference and this allows the rules to be relaxed without resulting in unacceptable levels of interference.
Mea navis aericumbens anguillis abundat
The amount of information that can be transmitted over an RF link with a given frequency band and noise floor is finite. The Shannon limit describes the absolute bottom signal to noise limit, below which no useful information can be sent. With wideband spead spectrum technology and robust error detection/correction algorithms, we can finally approach it. That is the bad news. The good news, this is 100x better than most of the (mostly uncompressed analog) open air transmission methods currently being used.
Consider a regular, low noise telephone line limited to 3 KHz bandwidth, no DSL, ISDN, or other high bandwidth enhancements. The first generation telephone modems ran at 110 or 300 baud. Eventually, QAM modulated modems came out that worked at 1200 baud. Later, 2400 baud modem appeared. This proved to be the limit of pure analog op-amp filter technology. 9600 baud modems requred a DSP, to process and recover data from the incoming signal. Later, 19.2k, 28.8k, 33k, and eventually (almost) 56k modems appeared, as the DSPs got faster, and more sophisticated filtering, error detection and recovery algorithms were used. But this was the limit. Pushing more data through a bandwidth limited, voice quality phone line requires a lower noise floor, or more bandwidth. Sending symbols faster requires greater bandwidth. Using a more complex symbol constellation requires a lower noise floor, or eventually the bits smear into each other to an extent that the error recovery mechanism cannot cope.
Open RF is much the same - you have a finite slice of bandwidth to use. You can reduce the signal to noise ratio by increasing the transmitter power, but then you become a greater noise source for everybody else who is transmitting over the same spectrum. CDMA phones are constantly adjusting their transmit power up and down, depending on how well the base station is receiving them. If the BER (Bit Error Rate) is too high, the phone is told to raise its transmit power. If the BER is low, the phone is told to reduce power, in order to reduce the noise. In a CDMA system, you can always add "just one more" transmitter, but eventually the noise floor is raised to the point where calls are dropped.
Also in open RF there are other problems to contend with, that dictate the optimal method of transmission - fading, (transmitter moves behind or out from behind a building) multipath, (Signal takes multiple paths to receiver, resulting in overlap because signals arrive at different times - think of trying to talk across an echoing canyon) and dopplar shifts. (Transmitter is moving, resulting in shifted carrier frequency) In practice, open RF is a pretty crappy transmission medium as compared to any sort of physical link.
In order to preserve optimal use of the spectrum for others, you don't want to transmit omnidirectional. If the receiver is in front of you, the signal you transmit to the sides and back are just wasted transmitter power, and an unwanted noise source for everybody else. Ideally, you only want the signal to go in a laser like path between transmitter and receiver. Very tricky if you don't know where the receiver is, or if it is moving.
My rights don't need management.
First, divide the spectrum into a million different slices. Specify some of them as high power, and some as low power, some as very long distance, some as fairly long distance, some as short distance, and a few as few very short distance, aka, bluetooth. (You need less as the distance gets shorter, because, duh, everyone can use the same ones.) OSI Layer 1.
Next, come up with some protocol. It needs a sender address, an optional destination address, and data, and that's it. Include a 'logical link control' to let people move around at will. Automatically negotiate a 'backchannel'. OSI Layer 2.
Then let anyone broadcast on any power/freqency, anything. But, and this is the key, require them to always negotiate downward to the 'worst' power/freqency that works, unless they have a good reason not to. (Aka, they're a TV station, and they don't want TVs to constantly have to call them back and say 'Hey, up the power a bit more.' Possibly you'd have to license this, but that's not important because most people would use devices that are bidirectional and thus don't need to worry about it, just the broadcast people.)
But, remember, each frequency has a set power, so if they want to broadcast stronger or lower, they'd need to change power/freqency. And, yes, attennas are designed for certain frequences, so we'd need to have evenly distributed strong and weak ones. (Aka, for every antenna size, we need to make sure we have a 'right next to the tower' power/frequency we can broadcast, and a 'A strong as possible' power/frequency too, that we can both hit with that antenna.)
Then build an assload of repeater stations. For any power/freqency, with the brains to let us link through them instead of directly. (In fact, if you have these, you can ignore the TV broadcast problem. Just have the TV station aim for these guys. If you can get it direct, good for you, if you can't, get it from them.)
The problem is that we're trying to solve a technological problem with regulations. There's plenty of bandwidth for everyone. It's just that we build devices that can't move around to get more. If we stop that, if we build some sort of 'airnet' that lets me use a specifically designated low power/freqency when I'm right under a cell phone tower, but flip to what is currently UHF at near-TV station power when I need drive behind a mountain (Thus sucking all my batteries, but that's not important.), we'd never have to worry again.
And, depending on how smart we design the network on top of this infrastructure, we can magically have cell phones and TV that can tune in radio stations, and direct-connect cell phones. Because there won't be any difference, except hopefully a layer of encryption on the cell phones.
Actually, every device will need some encryption, or at least authentication. So every idiot can't wander around pretending to be CBS. But, hey, we now have a world-wide wired infrastructure to grab public keys off of. And, really, no reason we couldn't hand out keys over the airwaves...we'd just have to what happens with web browsers now. You can get a signed cert, or you can make your own. In fact, make this part of layer 2, also, just because.
Of course, no way in hell this will ever happen...it requires throwing out all TVs and radios and cell phones and everything.
If corporations are people, aren't stockholders guilty of slavery?