page image

CHAPTER IV - Federal Policy for Spectrum

The 5G race is thought about in three ways: security, economics, and a resulting 5G dependent ecosystem. The success of all of this is predicated on our national communications policy.

Federal spectrum policy needs more flexible tools to increase access to federal spectrum, particularly on a shared basis. The most important near-term change is to reform the Spectrum Relocation Fund (SRF) to expand the ways in which federal users could use SRF monies. The longer-term objective is an automated spectrum management environment, building on the work done for the CBRS band.

The Desired End State for Spectrum Access
It is a given that the United States must respond to widespread demand for increased spectrum access. The desired end state is access to spectrum that is not a constraint on economic growth, social progress, public safety, or national security. Therefore, sufficient spectrum should be available for all of these purposes, although multiple purposes might be accommodated in a single band. On the mobile and commercial wireless side, there is a new emphasis on competition from a global perspective and national security. Spectrum for commercial uses, including mobile, should be made available through a balance of licensed and unlicensed access. The desired end state should include access to cloud services and applications for all.

A full and ongoing inventory of spectrum uses and a well-planned and consistently maintained spectrum map for both federal and nonfederal spectrum are needed, particularly if the U.S. is to have an automated national spectrum management capability that will enable additional spectrum sharing. For example, if the management of spectrum for wireless services is tied in with the expansion of fixed telecom infrastructure (e.g., continued build-out of fiber), shifting some services and uses to the fixed infrastructure will free up spectrum for other purposes. In addition, the following are likely to be required: cooperative federal and non-federal relationships that enable private transactions, incentives for efficient spectrum use, and federal user adoption of commercial services and technologies.

Government should also consider how to reform the Spectrum Relocation Fund. Directing additional funds towards the spectrum management ecosystem would increase the chances of getting to the desired end state. Monies could come from interference enforcement fines and application fees, as well as auction proceeds. These funds could be targeted toward automated spectrum management systems and a national spectrum inventory. A key change would be enhancing the Spectrum Relocation Fund (SRF), which is used to reimburse federal agencies for their transition costs when a band is re-allocated for commercial use, and also may be used under certain circumstances to study spectrum pipeline plans for other bands. It would be extremely helpful to allow federal agencies to use SRF monies toward better technology and system upgrades, rather than just comparable capability. This would create incentive for win-win situations in which both federal and commercial users can benefit. One of the PCAST report recommendations is “to create a spectrum efficiency fund so that revenues associated with spectrum are not limited to spending on that same set of frequencies but may be used flexibly” for other frequencies or equipment, stated Michael Calabrese of the Open Technology Institute.

The group discussed other ways to use funds, e.g., auction proceeds or fines from enforcement of spectrum regulations, in a manner that would facilitate increased access to federal spectrum. Many of these approaches would require legislation. Funds could be used for spectrum research, development, testing and evaluation, including test beds. As sharing and congestion increase, funds will also be required for interference protection, detection and resolution. Auction funds might also be used for broadband infrastructure deployment, as well as mapping the middle mile and the last mile. Other ideas include improving receiver performance, especially for critical infrastructure related systems, and providing the NTIA with more responsibility and authority to manage federal spectrum, analogous to the manner in which the General Services Administration (GSA) manages property for the federal government.

Government should also evaluate spectrum sharing approaches. The Citizens Broadband Radio Service (CBRS) is a radio-frequency band established to facilitate shared federal and non-federal radio spectrum use. The CBRS band is a poster child for the PCAST report, but it is important to have a clear-eyed assessment of how it is working, and lessons learned from the implementation of the sharing framework in that band. Specifically, there is a need to know whether the band is working for General Authorized Access (GAA) users, which do not get individual licenses, as well as for Priority Access License (PAL) holders, once the auction is complete, and the PALs have been issued. It is also imperative to understand whether the Spectrum Access System (SAS) model is repeatable—and to what end, and what can be learned about dynamic sharing. The evaluation should also include whether it is possible to speed up the process of developing the sharing mechanism. One participant said that his organization is getting feedback “that some parts of the CBRS ecosystem may not be the way forward, [such as] the sensor” component. Perhaps there are new incumbent informing models that could be used, and the framework could be tweaked a bit. And economic results can and should be compared to outcomes using different models of spectrum allocation in the 3.5 GHz band that are being widely deployed in other countries.

The group discussed how to define success in the CBRS band, and in what timeframe this will be possible. As GAA use begins, that will provide some information, but many people thought that a full assessment would not be possible until the PALs have been issued and deployment has occurred. It is important for the SAS to be working as a technical matter and enabling sharing, said Steve Sharkey, Vice President of Government Affairs, Technology and Engineering Policy at T-Mobile, but the critical question will be whether the SAS is working to enable services to be available in the way that is needed. Preston Marshall posited that the title of the PCAST report is not “lots of radios,” it is “economic growth.” In evaluating the success of the CBRS experiment the question should be whether there is growth in new startups, new employment, and new activity.

The process now for CBRS use is in fact very regulatory, and to change it you have to go back to the regulator, stated Charla Rath. This is unlike the process of dynamic sharing that takes place in exclusively licensed spectrum. Preston Marshall added that “the FCC gave [the wireless] community incredible latitude,” for example, “to decide coexistence and channel assignments.” This is because the FCC assumed that the multi-stakeholder process would function, without companies complaining to the FCC or the Hill.

The problem with reform continues to be fragmentary rights, stated Tom Hazlett. He continued:

GAA is not exclusive whereas PALs have some exclusivity associated with them. The SAS has its algorithms for assigning channels. And this fragmentation of ownership of the use is an expensive transaction, and that is why you will need to go back to the regulators. Suppose it is just a mess and you have got to clear it up? Who do you make a deal with? …Contrast this to the L Band, where LightSquared made a deal with Inmarsat to make contiguous spectrum and interleaving in the band became an issue. Hundreds of millions of dollars changed hands between the satellite carriers to get that done. The part they could not fix because of fragmentary rights, and there is no owner in GPS, are those border conditions that are still ongoing 15 years later. This is a transaction cost that is blocking something of value and eliminating the ability to make a deal.

The CBRS discussion revealed some disagreement over Hazlett’s conclusions: others at the conference thought that the path the FCC took, as imperfect as it was, was significantly better than any practical alternative, given the widespread federal incumbents in the band. Jonathan Chaplin said if CBRS is a disaster, for example because of the fragmentation of ownership described by Hazlett, it only demonstrates that it is extremely difficult to fix certain types of problems. But if it is not a disaster, “it could unleash significant value across bands of spectrum that are difficult to get to today,” said Chaplin.

Hazlett offered conclusory thoughts: When evaluating regulatory options, such as the CBRS sharing framework, it is important to consider the opportunity cost; and, in asking whether a regime works, the question should be whether it works better than the alternatives. We should “constantly be looking at new models and considering the costs and benefits,” and perhaps we could call that dynamic regulatory thinking.

Bidirectional Sharing is access sharing among federal and commercial users: federal government has access to commercial spectrum as well as commercial users having access to federal government spectrum. In the big picture, the federal footprint of spectrum use is constantly shrinking over time, typically “because [of] repurposing for commercial use,” said one participant. Generally, the assumption is that federal spectrum use is not going to expand in the band that is repurposed. But as federal uses change over time, it is worth seeing whether there are opportunities for expanded bi-directional sharing.

For certain federal systems, such as the military air-to-air/air-toground system, testing requires a huge peak spectrum need, one participant noted. This kind of occasional use might be a candidate for bi-directional sharing. There are also geographically isolated areas where the military would like to train using specific spectrum they will be using in other venues. “Every wireless carrier does those kinds of agreements on a regular basis,” stated Rath. Steve Sharkey added that those spatially isolated areas are part of the mix, but that federal users also ask for spectrum in more populated areas. Many think that the greatest amount of unused capacity remains in federal bands. And the federal users are not moving any time soon. “There are lots of things that the NTIA and the federal government could be doing to promote further sharing and that would be in line with the PCAST idea of abundance,” stated Calabrese. “It could be clearing, but more likely in the near term it is going to be sharing and that may require a federal SAS.”

Title Goes Here
Close [X]