Monday, June 20, 2016

Is There a Spectrum Crunch, or is There Spectrum Abundance? Yes and Yes

According to Cisco, mobile data traffic has grown 4,000-fold over the past 10 years and almost 400-million-fold over the past 15 years, while global mobile data traffic will increase nearly eight-fold between 2015 and 2020 alone.

So service providers always complain about the “spectrum crunch” that threatens to derail progress on the communications front.

On the other hand, many critics of the present spectrum allocation process say there actually is plenty of unused spectrum that could be put to use if we were smarter, and used new tools to allow use of fallow spectrum that already has been licensed.

In that view, hoarding of spectrum to prevent its use--intended or unintentional--is as big a problem as the total amount of usable communications spectrum.

The “unity of opposites” here is that both arguments are correct: there is a lack of available spectrum, as demand for communications soars, and “more” is needed.

On the other hand, there is plenty of available spectrum within the already-allocated 30 MHz to 3 GHz bands, if we could efficiently share its use, while protecting the existing license holders.

One way or the other, the simple answer for the capacity crunch is “more spectrum.”

Since most spectrum useful for communications--from 30 MHz up to about 3,000 MHz--already is allocated, spectrum sharing is the answer to gaining use of huge amounts of spectrum already licensed to existing users.

But spectrum sharing also is the key to efficient use of new spectrum in the millimeter bands (3 GHz up to 300 GHz). The notion is “use it or share it,” rather than the ability to squat on resources that nobody else can use, even if the licensee is not making any use, or only light use, of a resource.

In one important sense, spectrum sharing introduces a market mechanism for spectrum use that is efficient, and encourages licensees not to hoard valuable assets, but put them to work.

The traditional exclusive licensing of spectrum on an “exclusive right to use” basis  is inefficient. In fact, some policy advocates claim that as much as 95 percent of licensed spectrum is not used. Those claims appear to be based on a 2005 National Science Foundation study.  

Other studies of U.S. spectrum use between 30 MHz and 3 GHz did not always find that degree of fallow bandwidth, but the point is that much spectrum is not used much, most of the time.

That, plus the expected incremental demand for mobile and wireless communications, is driving innovations in network architecture (small cells), radio technology, mobile traffic offload, new spectrum (millimeter waves) and spectrum sharing.

All of that suggests the importance of two broad strategies: making better use of spectrum that is available, but lightly used, and opening up non-traditional bands of spectrum that traditionally have been very hard to use for commercial purposes.

And advancements in computing technology driven by Moore’s Law now are crucial to that effort. Simply, cheap and powerful computing makes possible lower-cost sharing of spectrum, as well as commercial use of millimeter waves (above 3 GHz and below 300 GHz) that have been too expensive and too difficult to use, in the past.

In the former case, sophisticated and cheap computing means we can allocate access in real time, in ways that literally were not possible in the past.

In the latter case, we also intentionally can design access systems, using those same techniques that allow robust sharing of resources among a number of potential users.

But just as important is the application of processing power to improve the usefulness of millimeter wave frequencies that are distance-limited (signals do not go so far) and signal propagation limited (signals cannot go through solid objects).

At the same time, the use of small cell networks helps overcome both distance and line-of-sight limitations. Better radio techniques also allow us to “bend” signals around solid objects and recover signals that have become weak or scattered.

“More spectrum” arguably is the single most important issue in communications. But spectrum sharing arguably is the most important tool for securing that needed spectrum.

As you would guess, incumbent service providers, including mobile and satellite firms, oppose sharing, while others, especially app providers, support spectrum sharing. But it seems inevitable that spectrum sharing, following the model of the 3.5 GHz Citizens Broadband Radio Service band, will be proposed.

The proposed CBRS service would use a three-tier access rights system, reserving priority access for existing licensed users, but allowing licensed secondary rights for commercial users where the primary licensee is not using the spectrum, with best effort licensing for other devices and services (on the model of Wi-Fi).

Spectrum (communications capacity) is a complicated matter. On one hand, there never seems to be enough spectrum to handle ever-growing numbers of users, the growing number of connected devices and higher-bandwidth applications such as full-motion video.


According to Cisco, mobile data traffic has grown 4,000-fold over the past 10 years and almost 400-million-fold over the past 15 years, while global mobile data traffic will increase nearly eight-fold between 2015 and 2020 alone.

So service providers always complain about the “spectrum crunch” that threatens to derail progress on the communications front.

On the other hand, many critics of the present spectrum allocation process say there actually is plenty of unused spectrum that could be put to use if we were smarter, and used new tools to allow use of fallow spectrum that already has been licensed.
---------------------------------------------------------------------------------------------------------------------
About Spectrum Futures
--------------------------------------------------------------------------------------------------

In that view, hoarding of spectrum to prevent its use--intended or unintentional--is as big a problem as the total amount of usable communications spectrum.

The “unity of opposites” here is that both arguments are correct: there is a lack of available spectrum, as demand for communications soars, and “more” is needed.

On the other hand, there is plenty of available spectrum within the already-allocated 30 MHz to 3 GHz bands, if we could efficiently share its use, while protecting the existing license holders.

One way or the other, the simple answer for the capacity crunch is “more spectrum.”

Since most spectrum useful for communications--from 30 MHz up to about 3,000 MHz--already is allocated, spectrum sharing is the answer to gaining use of huge amounts of spectrum already licensed to existing users.

But spectrum sharing also is the key to efficient use of new spectrum in the millimeter bands (3 GHz up to 300 GHz). The notion is “use it or share it,” rather than the ability to squat on resources that nobody else can use, even if the licensee is not making any use, or only light use, of a resource.

In one important sense, spectrum sharing introduces a market mechanism for spectrum use that is efficient, and encourages licensees not to hoard valuable assets, but put them to work.

The traditional exclusive licensing of spectrum on an “exclusive right to use” basis  is inefficient. In fact, some policy advocates claim that as much as 95 percent of licensed spectrum is not used. Those claims appear to be based on a 2005 National Science Foundation study.  

Other studies of U.S. spectrum use between 30 MHz and 3 GHz did not always find that degree of fallow bandwidth, but the point is that much spectrum is not used much, most of the time.

That, plus the expected incremental demand for mobile and wireless communications, is driving innovations in network architecture (small cells), radio technology, mobile traffic offload, new spectrum (millimeter waves) and spectrum sharing.

All of that suggests the importance of two broad strategies: making better use of spectrum that is available, but lightly used, and opening up non-traditional bands of spectrum that traditionally have been very hard to use for commercial purposes.

And advancements in computing technology driven by Moore’s Law now are crucial to that effort. Simply, cheap and powerful computing makes possible lower-cost sharing of spectrum, as well as commercial use of millimeter waves (above 3 GHz and below 300 GHz) that have been too expensive and too difficult to use, in the past.

In the former case, sophisticated and cheap computing means we can allocate access in real time, in ways that literally were not possible in the past.

In the latter case, we also intentionally can design access systems, using those same techniques that allow robust sharing of resources among a number of potential users.

But just as important is the application of processing power to improve the usefulness of millimeter wave frequencies that are distance-limited (signals do not go so far) and signal propagation limited (signals cannot go through solid objects).

At the same time, the use of small cell networks helps overcome both distance and line-of-sight limitations. Better radio techniques also allow us to “bend” signals around solid objects and recover signals that have become weak or scattered.

“More spectrum” arguably is the single most important issue in communications. But spectrum sharing arguably is the most important tool for securing that needed spectrum.


No comments:

"Tokens" are the New "FLOPS," "MIPS" or "Gbps"

Modern computing has some virtually-universal reference metrics. For Gemini 1.5 and other large language models, tokens are a basic measure...