Where exactly can you find wireless coverage in the United States? And how exactly can you measure that? Those are the questions that are facing Washington regulators as they attempt to allocate $4.5 billion in taxpayer money to help deploy wireless services in rural areas.
At issue is the FCC's Mobility Fund Phase II, a program designed to provide government money to private wireless carriers so that they will deploy wireless services in rural areas. This is obviously a key part of crossing the digital divide and delivering telecommunications services to rural areas that currently may not be connected to the internet at all.
But, before any money can be allocated, the FCC must first figure out what areas in the United States need wireless coverage. After all, the agency doesn't want to give money to a carrier to build out a network where there are already full bars from some other provider.
And that brings us to the FCC's ongoing “challenge process.” This is the process by which the agency is asking all the nation’s wireless carriers to map out their exact coverage areas. This, in theory, ought to allow the agency to figure out exactly where wireless service is available and exactly where it is not available. But, as with anything in Washington, this is no easy matter.
Challenges in the challenge process
First, the process to submit coverage information is complicated at best. Carriers and others that are participating in the process must record a variety of data points in specific locations where they are testing for coverage and then must have an engineer verify those results. Then they must upload that information through the FCC's complex reporting tool, dubbed Form 477.
And even when that is finished, not everyone agrees on the results.
For example, a group of rural wireless operators are challenging the maps submitted by Verizon as being “grossly overstated.” Not surprisingly, Verizon has rejected this argument and has pointed out that the rule carriers have a vested interest in shrinking its coverage map so that they can receive more government funding. Verizon, on the other hand, has a strong incentive to show that its coverage map is better than those of its competitors.
This has led a variety of players beyond wireless carriers to participate in the “challenge process.” For example, the Mississippi Farm Bureau Federation “has assigned an employee to review the technical requirements of the challenge process and this individual has demonstrated an understanding of the technical requirements, asking Commission staff a series of questions about the requirements,” the FCC noted in a ruling allowing the MFBF to participate in the challenge process.
In allowing the MFBF to participate in the process, the FCC also said that agency “is reminded that speed tests are only valid if conducted in accordance with our technical requirements and that in order to certify a challenge, an authorized representative of MFBF will need to certify, under penalty of perjury, that (1) a qualified engineer has examined all data submitted; and (2) the qualified engineer has certified that all data and statements contained in the submission were generated in accordance with these specifications and are true, accurate, and complete to the best of his or her knowledge, information, and belief."
In the meantime, some U.S. senators are raking FCC officials over the coals for not moving quickly enough to cross the digital divide and aid the deployment of rural wireless services. Sen. Jon Tester, for example, said the current FCC coverage maps "stink" because they show coverage where there is none.
"We've got to kick somebody's ass," Tester told FCC Chairman Ajit Pai in a hearing last week.
Partly in response, FCC Commissioner Jessica Rosenworcel said at the hearing that the FCC should do a number of things to improve the quality of its coverage maps, including conducting its own tests in its regional offices, partnering with the Universal Service Administration Company to do testing, and also soliciting testing results from regular Americans through the FCC’s "measuring Broadband America” program.
For his part, Pai pointed out that the agency has extended its comment period on the issue by 90 days, which he said will allow companies more time to file challenges to the FCC's coverage maps in order to improve the data.
Finding the best data
Another wrinkle in this whole issue is the fact that a number of third-party companies already provide coverage and speed data across the United States. Indeed, the FCC combined data submitted by wireless carriers with data from Ookla in order to create its competition report released earlier this year. That report was looking at whether the wireless industry was effectively competitive (the agency concluded it was).
And, in starting the process of allocating the Mobility Fund Phase II money, the FCC said that it did consider using data from Mosaik (recently purchased by Ookla) "but determined its limitations made it an inferior choice. For example, Mosaik data are not collected using a consistent methodology across geographic areas and service providers. In addition, Mosaik data are commercially provided subject to intellectual property protections, somewhat limiting their utility in the public policy sphere. Furthermore, the Commission has long expressed concern that Mosaik data likely overstate the extent of mobile broadband coverage. For instance, in 2010, the Commission’s National Broadband Plan relied on the data with respect to mobile broadband availability, but pointed out that the coverage was likely overstated. For these reasons, following the initial submissions of the new Form 477 data, we have relied on that data, rather than data from Mosaik, in assessing mobile broadband coverage outside the universal service context.”
What's even more interesting is, overshadowing this argument, an ongoing debate among wireless carriers about how fast and robust their networks are. Indeed, all of the nation’s wireless carriers proclaim that their networks are either the best or the fastest, depending on which data they are using. T-Mobile CTO Neville Ray even wrote an essay on the carrier’s website about how such data is collected—drive testing, crowdsourcing, and surveys—and argued that T-Mobile's data collection method was the best.
Finally, it's worth pointing out that this truly is a Sisyphean situation. All of the nation’s wireless carriers continue to spend money every day in order to improve their networks, either by deploying new wireless technologies, new wireless towers, or small cells for densification. Just as in the myth of Sisyphus, in which Sisyphus is forever pushing a boulder up a mountain only to see it roll down again, so too are regulators and others attempting to map coverage areas that continue to change and expand.
Of course, it is possible to roughly map out a wireless coverage area. Simply find where the cell towers are, estimate their general coverage area based on the technology and spectrum used, and call it good. In reality, though, that doesn't really work at a neighborhood level because geography, weather and other elements can affect coverage areas. “The impact of deciduous and conifer trees (under gusty wind conditions) suggest that the leaf density from the conifer more frequently produces heavy link losses and these, more so at higher carrier frequencies,” a recent report on millimeter wave transmissions recently noted.
This leaves a situation where wireless carriers say one thing, their rivals say another, and third-party data collection companies can only do so much. And if the FCC really wanted to conduct a full and comprehensive test of wireless coverage across the country, there wouldn't be much of the Mobility Fund’s $4.5 billions left to allocate to rural wireless carriers. A Sisyphean task indeed. – Mike | @mikeddano