Broadband.gov
Federal Communications Commission



Is There Enough Spectrum?

October 6th, 2009 by Charles Mathias - Assistant Bureau Chief, Wireless Telecommunications Bureau

Charles Mathias BBLast week, the FCC released a Public Notice (PN) asking for comment about whether there is sufficient spectrum available for wireless services.  Specifically, we asked for focused comment on the current spectrum allocations available in spectrum bands and whether that amount of spectrum is sufficient for our broadband needs.

The questions we ask in the PN arose as a result of the information we have already received in response to the National Broadband Plan Notice of Inquiry and the discussions at the workshops we have already held.  Multiple commenters have raised the issue that the United States will not have sufficient spectrum available to meet demands for wireless broadband in the near future.  Because of these comments, we're asking for your view on the fundamental question of whether current spectrum allocations are adequate to support near- and longer-term demands of wireless broadband.  The PN requests that commenters provide detailed, fact-based responses and to the extent possible provide quantitative data and analytical justification for their arguments.

The amount of spectrum available for use for broadband devices is crucial in determining an overall national broadband plan.  With the continued rise of the use of smartphones, and the needs for spectrum associated with their use, we have to look to the future availability of spectrum and where that spectrum is located.

The questions are as follows:

  1. What is the ability of current spectrum allocations to support next-generation build-outs and the anticipated surge in demand and throughput requirements?
  2. What spectrum bands are best positioned to support mobile wireless broadband?
  3. What spectrum bands are best positioned to support fixed wireless broadband?
  4. What are the key issues in moving spectrum allocations toward their highest and best use in the public interest?
  5. What is the ability of current spectrum allocations to support both the fixed and mobile wireless backhaul market?

Some people may ask "Why is the FCC asking for information on these questions?  Don't you already have the answers?"  What we are looking for are creative, as well as practical, opinions on spectrum availability.  New technologies can drive innovation; ideas for the availability of spectrum may be found or you may have creative ideas on to use the current spectrum allocation more efficiently.

Regardless, it is important for us to hear back from all different sides in the equation.  We want your input, no matter your background or your thoughts on the subject.

The easiest way to comment is to post on this blog.  Your comments will be included in the record for the National Broadband Plan.

You can also file comments with the FCC's Electronic Comment Filing System, using ECFS Express for short comments, or our standard submission page if you need to attach a file or to file in other dockets.  Please title comments and reply comments responsive to this Notice as "Comments (or Reply Comments)-NBP Public Notice # 6." See the PN for information about filing your comments in other relevant dockets.

4 Responses to “Is There Enough Spectrum?”

  1. Drew Childress says:

    Quite simply there isn't enough spectrum. State of the art modulation techniques achieve about 3 bits per second of throughput per hertz of spectrum utilized. Even if we take all the spectrum (no more broadcast TV or police radios) below 5 gigahertz (spectrum above 5 gigahertz is not very usable for non line-of-sight applications) that will only provide 15 gigabits per second of total bandwidth. At 50 Mbps that yields about 300 households. How many areas of the country have enough households to justify a tower contrsuction and operation but less than 300 households in the total footprint? The rosy projections for 4G with thousands of users are all based on large over-subscription ratios, where the same bandwidth/spectrum is shared by dozens of households. Unfortunately the nature of Internet usage is rapidly shifting to always-on video. With IPTV your DVR is recording 24x7 whether your TV is on or not. So wireless needs to stay focused on ad-hoc mobile applications and a very, very few access applications. Stop looking for bits of spectrum and start thinking smart.

  2. Brett Glass says:

    I intend to submit a comment in the docket (time permitting; my business is extremely busy and we are short handed). But since there is ice on the rooftops this morning and we cannot safely install wireless broadband for customers for another hour or two, I would like to post an informal comment here.

    I am a WISP -- a terrestrial wireless Internet service provider. Last week, I climbed a 40 foot ladder, on a day with winds gusting to 35 MPH, to put up a wireless access point on an apartment building . The intent was to serve a new housing development where the new homeowners were interested in receiving broadband service. Because I am a small, local businessperson, I cannot afford the huge sums to which the "spectrum barons" have bid up licensed spectrum (in part, to foreclose competition), and none of these companies will partition, sublicense, or sell any spectrum that they have already acquired. And so, even though a spectrum analyzer shows that most of the spectrum in my small town and its surrounding areas is completely vacant, I mounted -- clinging to my safety lanyard with fingers numbing due to the wind chill -- an access point which operates on the 2.4 GHz unlicensed band. The unit I chose is a high quality, noise resistant model made by a Georgia company called Deliberant.

    I then went out to install radio equipment for the first new customer.

    To my dismay, the link -- even though it was shorter than many we've constructed and had a completely clear line of sight -- did not work. The neighborhoods between the access point and the new customer contained so many Wi-Fi access points, baby monitors, DSL modems with Wi-Fi, and other consumer wireless devices operating on this unlicensed band that no matter what channel we used, spikes of noise interrupted the signal.

    I went back to the drawing board, spent another thousand dollars on equipment, and -- after another scary climb on a steep, tall roof -- mounted an access point that operates on a different unlicensed frequency band. Unfortunately, the FCC rules for that band are much more restrictive. Lower maximum signal levels and higher path loss allow the signal to propagate for a much shorter distance, limiting the number of subscribers it can cover. And "unwritten" rules, imposed upon the manufacturer not by Part 15 of the FCC rules but by the FCC engineers who approve the equipment, force the access point to pick its channel at random rather than choosing the clearest channel, greatly increasing the potential for interference from other users of that frequency band and preventing users of the spectrum from coordinating with one another to avoid interference.

    This story -- at least -- ended happily. Due to our persistence and the potential customer's patience (he very much wanted to use a local provider rather than the telephone or cable company), we finally got the service working, but it was an unnecessarily expensive and difficult ordeal. And while the access point and customer premise equipment for the link cost much more than those for 2.4 GHz, the range of the new access point is severely limited.

    The second case study did not, alas, end as well. We were called by a rancher, in a very rural area, to provide Internet access for his children, who needed it for school. We already had a wireless access point about 5 miles away, but when we arrived we discovered that a hill was blocking the path. We tried the most powerful antennas we had, but the strength of the signal which made it over the hill was simply too low to serve him. Radiating a stronger signal to reach him would have disturbed no one in that vast, lonely expanse of prairie, but would have violated the FCC's rules.

    We attempted to find a location for a relay station from which we could serve the ranch, but were unable to do so in time. With the school year fast approaching, the rancher opted for satellite service, even though this service had long latency and a download cap which was relatively restrictive for use by three children.

    The third case study involved a house in town which is rented to several graduate students at the University of Wyoming. This house is in Laramie's "Tree Area," a part of town which is filled with deciduous trees (mostly oak-like "Plains Cottonwoods"). We have access points on the 900 MHz and 2.4 GHz unlicensed bands only six blocks from the house.

    The students in the house called us to provide high speed Internet service -- preferring us over the cable company and the phone company because we do not require a "bundle" or a long term contract to give the customer a reasonable monthly rate. But when we got to the house, we discovered that we could not make an adequate connection. The six blocks of trees completely absorbed the 2.4 GHz signal (which is limited to 4 watts of effective isotropic radiated power and can typically penetrate only one or two trees before it becomes too weak for reliable wireless broadband delivery). The 900 MHz signal -- with an even lower power limit of 1 watt of EIRP -- was mostly absorbed and the remainder blotted out by noise from consumer devices. (To see how noisy the 900 MHz band is, even in our small city, view the spectrum analyzer display at http://www.brettglass.com/Laramie900.jpg. Also note that this reading was taken in 2005; the noise levels have actually grown since that time.)

    The final case is a somewhat amusing one. An existing customer of our wireless broadband service called to report that his service seemed to be affected by weather, but not in the way one might expect; it was failing in GOOD weather, not bad. On sunny days, his Web browsing was dramatically slower and sometimes stopped altogether.

    We went to the customer's house with a spectrum analyzer, and finally were able to root out the problem. On warm sunny days, the customer's teenage daughter, who liked to sunbathe, would lie on the deck in the back yard... and would take her 2.4 GHz cordless phone to chat with friends, as teenagers are wont to do. Because the antenna for wireless broadband service was mounted directly above the deck, this placed the cordless phone right in front of the antenna and shut down the customer's wireless broadband link. The phone wasn't labeled with the frequency it used, so the consumer had no easy way of knowing that this was the cause, but our spectrum analyzer -- an expensive piece of gear to which consumers don't normally have access -- revealed the problem.

    We arranged for the customer to get a DECT phone, which operates on a different frequency band, and this solved the problem.

    (Note: Unfortunately, interference problems such as this one are quite common, and are exacerbated by the fact that current FCC rules do not require consumer wireless devices to be prominently labeled with the frequencies they use -- both on the packaging and on the device. As a result, consumer devices do not just interfere with wireless broadband service; there is also frequent interference among them within homes and offices. A simple labeling requirement would help consumers to avoid self-interference when they purchased these devices.)

    These are but four of the many instances in which consumers' broadband choices and/or peformance -- were limited not by a lack of spectrum (again, most of the spectrum in our small town sits idle) but by our inability to obtain permission to use unique frequencies or to transmit at adequate power to surmount obstacles in an outdoor environment.

    There are several obvious solutions to these problems. One -- endorsed publicly by FCC Commissioner Adelstein in 2005 (see http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-257737A1.pdf) -- is to allow bona fide wireless broadband providers who are offering services to the public to radiate stronger signals, overcoming both physical obstructions and the "buzz" of noise from consumer devices. An 8 dB increase in allowed power over the existing Part 15 limits on the 900 MHz, 2.4 GHz, 5.3-5.8 GHz, 24 GHz, and 60 GHz bands -- which could be limited to counties with populations of under 250,000 if there were concerns about interference in urban areas -- would greatly enhance the coverage of wireless broadband providers operating on unlicensed spectrum. Such an increase could be easily implemented in existing systems by allowing the use of signal boosters with automatic power control -- some of which are already available, though not yet approved for use by anyone other than amateur radio operators. Such boosters could be required to include bandpass filters to ensure that the louder signals remained within the proper frequency band.

    A second solution would be to approve the use of more unlicensed, or nonexclusively licensed, spectrum specifically for use in the delivery of wireless broadband but not for consumer devices. (No one has presented any credible argument that the large amount of spectrum available for such devices is inadequate.) Available bands that come to mind include the AWS-3 band (which has already been considered as a vehicle for delivery of wireless broadband), the 700 MHz "D" block (in which providers' equipment could be programmed to step aside -- via a DFS-like mechanism -- for emergency traffic), and the LMDS bands (in which licenses are now expiring without any substantial buildout). The advantages of nonexclusive licensing, rather than making the new spectrum unlicensed, would be the ability to limit use of the band to bona fide ISPs and the ability to revoke the licenses if they did not conform to the rules or coordinate with other users of the band.

    A third solution would be to adopt a strict "use it or lose it" policy for spectrum. Most of the spectrum in our small city is, in fact, licensed but is completely unused. Requiring exclusive licensees to give up their licenses if they did not use the spectrum would likely free up quite a bit of spectrum, making productive use possible. A principle of "adverse possession" for spectrum -- in which it would be possible to obtain the right to use spectrum by putting it to productive use when the licensee did not -- would likewise ensure that innovators and entrepreneurs who came even one day after the end of a spectrum auction were not forever "locked out" of the use of spectrum.

    For more on these issues, see my slides from the recent broadband workshop at

    http://www.brettglass.com/Wireless

    Brett Glass Owner and Founder, LARIAT The world's first Wireless ISP (WISP)

    To contact me regarding this or other issues, see http://www.brettglass.com/mailbrett.html

  3. Wayne Longman says:

    The Commission and the public should be aware that the majority if not all economists who are proponents for the application of economic theory to the radio spectrum have no real-world hands-on experience in the planning or implementation of real radio systems. They display no awareness of the enormous complexities in the planning, implementing, and resolving conflicts of frequency selection and coordination, for the design, installation and user experience. They have never committed their money or energy to the hardware and the infrastructure needed and the people involved in building it. To those who know and care about the radio spectrum, the application of economic theory to the spectrum seems akin to economists dictating how doctors will deliver health care. Radio is in the realm of physics and engineering, not economics.

    Recent wireless social and economic achievements are the result of technology advances, and business acumen - not the result of economic theory. The best that can be said is that an auction process was substituted for a broken-down comparative process. Rather than face a refurbished administrative process, the parties opted for a financial shoot-out, where the richest won. Economists are wont to say that the market is the best way to distribute a scarce resource. Practical experience says that auctions also take place when there is an abundance of a resource, such as foreclosed homes or traded-in automobiles. In a true shortage, items are dealt with on a case-by-case basis. The traditional spectrum allocation and assignment process does just that. It also creates competition for spectrum between non-competitive users, something the market can't. A used car dealer will not bid for a home to sell.

    Admittedly, the U.S. administrative process is cracked if not broken, but only in certain aspects. There are thousands of radio systems successfully operating today that will continue to do so under the existing processes. New approaches are selectively needed, but there is no reason to believe there are only market based solutions.

    Economic theory did not cause the success of public mobile communication. Success was the result of a combination of technology and the demand for the technology. The recitation of benefits derived from public mobile radio, to support economic approaches is clearly erroneous. If the spectrum had been allocated by another process, the same result, more or less, would have happened, without the hidden cost and indirect taxation of auctions. In pursuit of economic benevolence, the auction costs are mysteriously identified as sunk costs, rather than input costs that must be recovered, and with a profit. The transfer of auction revenue to the government is a tax that is paid for by most of the public. This same selective analytic attitude is similar to that which led to the dismissal of risk that led to the economic collapse, and in terms of radio can have similar deleterious consequences.

    Abandoning the conventional frequency allocation and assignment processes leaves no mechanism for the certain prevention or cure of radio interference. There is absolutely no evidence, proof or experience that a "commons" approach will work to provide reliable communication, and, to the contrary, there is overwhelming experience it will not where reliable data communications is an absolute necessity. The experience to date with unlicensed devices has been mixed. Their design is based on an assumption that interference will be rare because of their limited range of use. As they become more popular, resulting in an increased density of use, this assumption fails, and so will the devices. A similar assumption and its failure apply to spread spectrum and opportunistic devices.

    Secondary spectrum markets will lead to spectrum hoarding, higher costs, fragmented spectrum and services, and loss of public involvement and future flexibility. Enough said.

    Mobile communication is not expected to be fully reliable because of the physics of the environment, so there is some real world congruence with the economic theory of laissez-faire spectrum management. The application of economic theory in lieu of engineering theory to radio spectrum choices will make it even less reliable. Additional spectrum should be allocated to mobile for only those services that can't be delivered by non-spectrum means. The spectrum usable by mobile is finite so it must be reserved for needs with no other choice. This also means that mobile or fixed radio systems should not be allocated spectrum to provide high bandwidth fixed broadband services that require high reliability.

    In conclusion, economic theory has not been shown to provide a general solution to the use of the radio spectrum, nor has it been shown that it has been applied beneficially in any case that might have been achieved through other means. Markets have amply demonstrated that they cannot look out for their own good, let alone that of the public.

    There are other solutions that will work.

  4. Guest says:

    test

Leave a Reply



Capture The Phone Numbers Using Your Camera Phone

If you have a camera and a 2D matrix code reader on your mobile phone, you can capture the FCC Phone numbers right to your phone by following these three easy steps:
Step 1: Take a photograph of one of the codes below using the camera on your mobile phone.
Step 2: Use your phone's Datamatrix or QR Code reader to decode the information on the photograph. Please note, these code readers are device specific and are available to download on the internet.
Step 3: Store the decoded address information to your phone's address book and use it with your Maps or GPS application.

Datamatrix and QR FCC Phones