The Federal Communications Commission on Wednesday disclosed plans to scrap its flawed Mobility Fund Phase II proceeding after an investigation determined carriers, including T-Mobile, Verizon and U.S. Cellular, submitted unreliable and overstated 4G LTE coverage maps.
Instead, FCC Chairman Ajit Pai has proposed subsidizing up to $9 billion (including the roughly $4.5 billion originally earmarked for 4G LTE under MF-II) for rural mobile 5G deployments over the next 10 years under a new program. Of that, $1 billion will be set aside specifically for precision agriculture.
The FCC’s Rural Broadband Auctions Task Force report (PDF) recommended the agency terminate its MF-II process, concluding that Verizon, T-Mobile and U.S. Cellular’s MF-II coverage maps didn’t accurately reflect real-world experience.
The year-long investigation included nearly 25,000 speed tests in 12 states. According to the report, FCC staff achieved a minimum download speed of 5 Mbps where carriers claimed to have coverage in 45% of tests for U.S. Cellular, 63.2% of for T-Mobile and 64.3% for Verizon. The investigation couldn’t find any 4G LTE signal for 21.3% of drive tests on T-Mobile’s network, 16.2% on Verizon’s network, and 38% on U.S. Cellular’s, “despite each provider reporting coverage in the relevant area.”
Inaccuracies in operators’ coverage maps is not necessarily a surprising issue, but reliable data was key to the design of MF-II since federal support for smaller carriers’ 4G LTE deployments was predicated on the FCC’s ability to use the maps to determine which areas of the country were unserved and therefore eligible.
T-Mobile for its part still feels its maps are accurate and in a statement said: “We stand behind our network coverage and all of our maps, but agree with the FCC that there is an opportunity to improve their procedures for collection of broadband coverage data for the Mobility Fund maps. We look forward to working with them and Congress to revamp the process.”
None of the carriers will be penalized for their inaccurate maps though, as the FCC didn't determine that carriers violated a specific commission rule that would warrant enforcement action, nor did it find enough evidence to say they deliberately submitted overstated coverage data.
Instead the report recommended the FCC release a broader Enforcement Advisory on broadband deployment submissions and penalties, as well as adopt policies that include clearer and standardized specifications for mobile coverage reporting.
A flawed process
The FCC undertook its own investigation into the Verizon, T-Mobile and U.S. Cellular maps after early MF-II challenge results and concerns from smaller operators and industry groups called in question the accuracy of coverage maps. That includes the Rural Wireless Association, which previously claimed Verizon “grossly overstated” its nationwide 4G LTE reach, particularly in rural areas. During the challenge process, where stakeholders could dispute carriers’ coverage claims, more than 20 million speed tests were submitted to the FCC.
The Competitive Carriers Association (CAA), an industry group representing smaller and rural operators, said it supported the move to end the MF-II challenge process, and pointed to inadequate reporting methods.
“I thank the FCC for its findings that the MF II initial eligible areas maps are fatally flawed, and its focus on improving the maps. The maps overstated coverage because the parameters relied on by the FCC model overstate coverage,” said CCA president and CEO Steven Berry in a statement. “I welcome the recommendation to update data specifications, including signal strength, cell loading factors, cell edge probability, clutter factors, and fading statistics.”
The CCA also urged swift consideration of pending Broadband DATA Act version in both the House and Senate to address updating reporting parameters.
The FCC report also noted large discrepancies between the challenger-submitted data and agency results. For example, when comparing a subset of challenger submitted speed tests (87,958) and FCC tests that were conducted within 100 meters of each other, challengers recorded zero Mbps in 60.8% of the tests, and at least 5 Mbps only 22.2% of the time, while the FCC results recorded zero Mbps in only 4.4% of tests that connected just to 4G LTE, and 23.2% when any network connection was present. The commission hit the 5 Mbps threshold in 71.4% of tests on 4G LTE and 54.8% overall.
The FCC did not identify a specific cause for these differences, but said it was not confident that individual challenger speed tests results accurately reflect the typical consumer experience.
“The causes of the large differences in measured download speed between the staff and challenger speed tests taken within the same geographic areas, as well as of the high percentage of tests with a download speed of zero in the challenger data, are difficult to determine,” according to the report. “Discrepancies may be attributable to differences in testing methodologies, network factors at the time of test, difference in how speed test apps or drive test software process data, or other factors.”
In dispersing additional Universal Service Fund subsidies for 5G, the commission hopes to avoid the problem of needing industry-provided coverage maps to determine eligibility because unlike 4G, 5G is relatively greenfield and not deployed in most parts of the country.
Instead of using maps to determine where service is lacking, the FCC would use methods like drive testing, auditing and independent oversight after the fact, to ensure carriers that receive support from the 5G fund actually build out those services in rural America.
Pai is expected to circulate a Notice of Proposed Rulemaking for the 5G fund next year. Exact eligibility parameters have not been decided.
With carriers currently deploying varying flavors of 5G using millimeter, mid- and low-band spectrum, what constitutes 5G under the new fund will be based on two criteria: meeting the 5G New Radio (NR) technical standard under the 3GPP, and a yet determined speed metric that would be verified by a drive test.
The FCC will seek comment through the NPRM as to what that speed metric should be.