After the 

Spectrum Inventory ...

The legislation for a spectrum inventory is gaining momentum and hearings will be held shortly.  This week I sent a paper to the NTIA Commerce Spectrum Management Advisory Committee on what is likely to happen after the spectrum inventory is completed.  My prediction: gridlock.

Let us take the example of the technical issues in the AWS-3/M2Z controversy.  Everyone agrees that the band has no primary occupants in it anywhere in the US.  The key technical controversy that has dragged on for 3+ years is how the band can be used without impacting the incumbents in the lower adjacent band - mainly T-Mobile.  This, in turn revolves on what constitutes "harmful interference" and to a lesser degree on what receiver immunity is reasonable before an adjacent channel users  can complain of interference.  So if this is the mess we get with a completely vacant band, imagine the mess we will get with bands that have use in some areas but not others and have intermittent users.

I propose that we decrease this controversy by having NTIA and FCC work in parallel with the inventory to improve the definition of harmful interference, develop improved transparent procedures for making harmful interference determinations in a timely way, and clarify receiver expectations.

Otherwise the inventory will likely be, in the words of the Bard of Avon:

... a tale
Told by an idiot, full of sound and fury,
Signifying nothing.

Commissioner Meredith Attwell Baker

Speaks on 

Spectrum Policy

On December 3, Comm. Baker gave a speech entitled "A Spectrum Management Framework" to  The Phoenix Center for Advanced Legal &  Economic Public Policy Studies 2009 Annual U.S. Telecoms Symposium.  As far as I am aware, this was the first speech this year by a commissioner to basic spectrum policy issues.  Comm. Baker is particularly well qualified in this area as a past head of NTIA.

Her key points were :

"1.    An up-to-date, strategic spectrum management framework that includes achievable and clear short and longer term goals. A transparent plan will provide a predictable flow of spectrum resources to broadband providers to allow for planning by both existing providers and new entrants, as well as more flexible use of existing allocations.

2.    Economic and regulatory policies to facilitate investment in mobile broadband networks—including deploying 4G networks and enhancing 3G infrastructures. These policies need to support the continued success of the competitive wireless market, but not to the exclusion of entrepreneurs and new entrants. We also need to align the incentives to deploy wireless services to unserved and underserved areas.

3.    A policy plan to actively promote innovation, including cutting edge research and development in areas that will help increase the efficiency of spectrum use. Mobile broadband would not exist today without basic research conducted years ago and relentless applied research and development. We cannot expect to benefit from the types of advances that took us from brick phones to smartphones without a comprehensive commitment to world-class research and development in the area of mobile technologies—here in the United States. This could include enhanced collaboration with
technical advisors and other spectrum experts and the adoption of policies that foster further innovation."

Let me focus on the 3rd point, "A policy plan to actively promote innovation, including cutting edge research and development in areas that will help increase the efficiency of spectrum use." The free enterprise system has done amazingly well in this country in bringing us the benefits of the information society.  Wireless innovation lags innovation in other areas because of government regulation of spectrum which is needed in some degree because of the limited nature of the spectrum resource and the potential for interference.  (My Mac doesn't interfere with your PC so there is no need for government to get involved in computer technology regulation.  Nor is there a finite amount of semiconductors available for computers.)

The key policy challenge for FCC and NTIA is how to craft spectrum policy to encourage private capital formation for R&D.  (Since many new innovations raise questions of interference to federal users and NTIA has effective veto power over FCC actions in such areas, NTIA is also a key player here.) Such policy should deal fairly with both existing manufacturers and operators and entrepreneurial ones. (Remember not that long ago both Microsoft and Qualcomm were startups!)  The current spectrum policy deliberations are so drawn out and complex that capital formation is certainly inhibited.  Whether you agree with M2Z's business  model or not, should it really take 3+ years to resolve whether TDD can be used in AWS-3?

She fully supports the spectrum inventory concept

"One critical tool government and industry needs is a spectrum inventory to better understand how spectrum is being used today across all bands. Such an inventory should be dynamic and focused on data that will inform and facilitate additional spectrum use. The output should be a user-friendly resource for all interested parties and should be able to be incorporated into more sophisticated spectrum management tools. Such an inventory will be critical to government efforts to manage spectrum more effectively as well as spectrum users trying to find fallow spectrum that can be transformed into greater connectivity and new services."

I support it also.  But I recognize that without some progress on clarifying "harmful interference" and speeding adjudications of "harmful interference" the spectrum inventory will be a waste of time and resources.  Why?  Going back to the AWS-3 case, everyone agrees that 2155-2175 is empty of other primary users.  But there is huge disagreement on what use of this band is acceptable without causing "harmful interference" to the lower adjacent incumbents.  This is a pattern repeated many times for innovative technologies.  We need a system that deals with these issues in a timely and transparent way.  The Docket 09-157 NOI asked some key questions in this area (para. 34-37).  It states

Spectrum allocations and access often hinge on controlling interference between new services and incumbent services, as do licensing and service rules to some extent. The resolution of disputes about potential or actual interference in rulemakings can pose a major impediment to the introduction of new services, devices and technologies, either as a result of long delays in the establishment of service rules or the imposition of onerous and perhaps unachievable technical standards.

These "long delays" and "onerous/unachievable standards" are also key factors in inhibiting innovation.  Unfortunately, few of the commenting parties so far have offered helpful improvements.  In general, the "haves" are happy with the status quo and they don't realize that it will prevent the inventory from having much impact.

IEEE Publishes Spectrum Inventory Article

My article on practical aspects of implementing change after the pending legislation on a spectrum inventory has been published in IEEE Wireless Communications. Comments are welcome here in the blog.

An excerpt:

Repeated surveys of spectrum occupancy in the U.S have shown large amounts of idle spectrum even in urban areas. This is the inevitable result of spectrum allocations based on peak demands of individual users in areas with uneven terrain and uneven population density. There is reason to believe that the same phenomenon occurs in all other countries although data is not as readily available. Progress in spectrum inventories is necessary to identify how much spectrum is used and how much might be available for new uses. However, an inventory by itself will lead to many further questions and time consuming deliberations unless it is accompanied by progress in defining harmful interference, determining receiver standards or at least expectations, and making the resolution of such issues more transparent so that new entrants can participate.

Spectrum Inventory Bill Moving Through Congress

BroadbandCensus.com reports that The Senate Commerce Committee on Wednesday approved S. 649, the Radio Spectrum Inventory Act, which would give the NTIA and FCC 180 days to present Congress with a complete inventory of the radio frequencies that they manage from 300 Megahertz to 3.5 Gigahertz.

The key provision of the bill is a new Section 342 of the Communications Act that would read:

(a) Radio Spectrum Inventory- Not later than 180 days after the date of enactment of this section, and biennially thereafter, the National Telecommunications and Information Administration and the Federal Communications Commission, in consultation with the Office of Science and Technology, shall carry out each of the following activities:

(1) Conduct a report that includes an inventory of each radio spectrum band, from 300 Megahertz to 3.5 Gigahertz, managed by each such agency. Such report shall include--

(A) the licenses or government user assigned in the band;

(B) the total spectrum allocation, by band, of each licensee or government user (in percentage terms and in sum);

(C) the number of intentional radiators and end-user intentional radiators that have been deployed in the band with each license or government user; and

(D) if such information is available--

(i) the type of intentional radiators operating in the band;

(ii) the type of unlicensed intentional radiators authorized to operate in the band;

(iii) contour maps that illustrate signal coverage and strength; and

(iv) the approximate geo-location of base stations or fixed transmitters.

CTIA immediately applauded this event. Not surprising since the Center for Responsive Politics reports that the following organizations are formally lobbying for this bill:

Cellular Telecom & Internet Assn
National Amusements Inc
Verizon Communications
Wireless Broadband Coalition
National Cable & Telecommunications Assn
Technology Assn of America
Alcatel-Lucent
AT&T Inc
Broadwing Inc
Consumer Electronics Assn
Entergy Corp
Lockheed Martin
MetroPCS Communications

I believe that more information on spectrum use is a long needed idea. My former FCC colleague, Paul Kolodzy, used to say during the Spectrum Policy Task Force/SPTF deliberations that the US spends more counting agricultural production like corn and cattle than spectrum even though spectrum continues more to the GDP.

But let us look at the current bill and what it proposes to be accomplished in 6 months. For the FCC, (a)(1)(A) and (a)(1)(B) are already done and have been for decades. (a)(1)(C) involves information that is not presently available at FCC on end user equipment. How many mobiles units does police department x or taxi company y have? Note that to make the job of the CTIA membership easier, they no longer even report cell site information to FCC. So at FCC (a)(1)(C) will require a massive new data collection, often involving small businesses which hold radio licenses. Then we have radios that are "licensed by rule" such as boat and aircraft radios on noncommercial units. FCC doesn't even know who these "licensees" are, so counting them will be a real challenge. (Should we mention the wireless mic issue at UHF where 90% of the users are unlicensed and illegal under present rules?)

Now in the case of NTIA much of the information is available in classified databases. The bill tries to deal with this in (b):

(b) National Security Exemption- A licensee or government user of spectrum may petition the Commission or the National Telecommunications and Information Administration for a partial or total exemption from inclusion on the website and in the report required by subsection (a). Such an exemption may be granted only to the extent that each such agency determines that disclosure of such information would be harmful to the national security of the United States. The licensee or government user seeking an exemption under this subsection bears the burden of justifying the exemption and shall provide clear and convincing evidence to support such an exemption. Any such exemption shall apply only for 2 years and shall expire upon the end of such 2-year period unless the licensee or government user seeks and obtains an extension in accordance with this subsection. Any information that is excluded from public disclosure pursuant to this subsection shall still be compiled and reported to the Committees of Congress described in subsection (a)(4) on a confidential basis.

Those of us use to how things work can images a huge paper pushing exercise at NTIA on how much could be declassified. Considering this problem has built up over decades, it is unlikely it will be resolved in 6 months.

But what happens when the inventory is finished? Presumably CTIA and friends want to hunt for new spectrum to use for their industry. Fine enough. When can the spectrum be used? Presumably when it doesn't cause "harmful interference" to other users? What is "harmful interference"?

CTIA and mainstream spectrum players such as broadcasters have consistently fought against FCC trying to elaborate on the current ambiguous definition of harmful interference. As "haves" and insiders the current ambiguities give them an upper hand where each issue is resolved in intensive 8th Floor lobbying. They do not want transparency.

Another topic opposed by mainstream spectrum users is receiver standards. (Note that in this context, "standards" need not mean regulation but could mean minimum performance before one is entitled to protection.) Although it is not obvious to newcomers, it is well known in the spectrum field that "most spectrum is used by receivers not transmitters". So if and when the "inventory" is finished, we will all get into a new set of massive arguments about whether a proposed new spectrum use will cause "harmful interference" to some population of receivers that has some unknown immunity to nearby signals, adjacent channel signals, signals capable of causing receiver-generated intermodulation, etc.

Finally, in the case of NTIA-controlled (Federal) Government spectrum, how do you decide if some could be used for new private uses? NTIA control is more nominal than real since in reality the NTIA management can not order more power agencies to spend their own money. This is the natural consequence of the moves by Presidents Nixon and Carter to move the Section 305 power of the President out of the White House to the Commerce Department and create NTIA. The present NTIA structure is just not up to the job.

Congressionally mandated reallocations, e.g. 47 USC 923, could be tried again, but they have long term costs in creating a more hostile atmosphere between FCC and the IRAC members who really control most of the spectrum management decisions in the federal government. I believe that these previous actions poisoned the atmosphere and made NTIA/IRAC drag heir feet on other issues and probably stimulated FAA's unsuccessful endrun on both FCC and NTIA.

So my bottom line is that the spectrum inventory will have little net success, other than employing K Street lawyers, unless it is accompanied by at least some progress in:

1. clarifying "harmful interference",
   
2. receiver standards, and
   
3. NTIA/IRAC reform.