Connecting Across the Big Sky
Connecting Across the Big Sky
The State of Montana implemented a state-wide broadband service availability map. The project was funded by a grant from the National Telecommunications and Information Administration (NTIA) and is a component of the American Recovery and Reinvestment Act (ARRA) program. NTIA provides guidance for the project and defines the specific data to be collected, level of mapping detail, and final data formats.
The results of the Montana mapping effort became part of the NTIA National Broadband Map, to be used for planning and development of broadband in Montana.
For Broadband Mapping, the National Telecommunications and Information Administration requires that all wired (DSL, Cable Modem, and Fiber to the End User) broadband service be reported at the census block level (the smallest geographic unit used by the United States Census Bureau) and wireless broadband service reported as a map area or coverage. In very large census blocks (greater than 2 square miles) wired coverage is report by street segments.
The NTIA defines broadband availability as follows: "'broadband service' is 'available' at a location if the provider does, or could, within a typical service interval (7 to 10 business days) without an extraordinary commitment of resources, provision two-way data transmission with advertised speeds of at least 768 kilobits per second (kbps) downstream and greater than 200 kbps upstream to end-users at that location."
To meet the objectives of this project, Montana developed a flexible and robust set of mapping methodologies. The methods implemented depend on the level of cooperation of broadband service providers, type of broadband technology being mapped, available public data, and other factors.
Working with Providers
Provider participation in this project was critical to accurately conveying the status of broadband service and availability in Montana.
The program typically asked a provider for the following types of data or information:
When we were not able to get detailed information from providers of their service areas that could be mapped at the census block level, we started by modeling coverage based on known infrastructure points. The mapping process used infrastructure points (central offices, remote terminals, wireless towers and antenna locations, middle mile and backhaul), cable franchise areas, and anchor institution addresses. Coverage models were derived dynamically from this infrastructure based on geoprocessing techniques specific to each broadband technology. Examples of the techniques include developing propagation models using the Longley-Rice model for wireless coverage and using infrastructure points in conjunction with the road network to predict the area served for DSL coverage.
Putting the Information to Use
Using this information, a variety of manual mapping techniques were used to compile the information into standardized mapping databases for use in the project. These techniques included hand compellation, tracing and digitizing, and extract, transform, load (ETL).
Montana developed an extensive system for quality assurance and for quantifying "validated" data for the purpose of determining what is suitable for delivery to NTIA. Using reliability and validity codes, together with completeness checks, we tracked which data elements were complete or still in process of refinement. Infrastructure was compared to public data, independent measurements, and telecommunications provider submittals at varying levels of geography. As more data was obtained from providers and systematically checked against infrastructure points, the reliability and validity progressed from 1 (not validated or reliable) to 10 (validated and reliable) and was recorded. Completeness was primarily dependent on provider input, and could be supplemented in many instances with independent measurements. The process was iterative. Our validation methods provided the ability to use general information and iteratively cross check and improve the mapping information as more accurate data was obtained.
Montana had developed a very robust operational data model to support our broadband mapping efforts. This data model allowed us to store data at a high degree of granularity and to derive a variety of output data products from the mapping database. Our operational model could support any reasonable modifications to NTIA requirements over the five-year life of this project.
Montana had also implemented two very innovative and aggressive independent verification methods that we used to support mapping and quality assurance. The first method was a broadband speed test application and web site. The speed test web site resulted in over 12,000 tests for Montana. The test integrated a standard web speed test into a full application that validated users address, segregated general public and anchor institution tests, and performed a reverse "whois" lookup to validate provider information.
The second independent verification method used in Montana was direct measurement field tests of wireless coverage. This was accomplished using custom smart phone programs that logged geographic location, signal strength, data throughput, cell tower information, and provider. Using this tool we drove over 6000 miles in Montana and bordering states to record wireless availability. The information gathered was used to support signal propagation modeling and for verification of overall mapping accuracy.