What do WSC-3 radios, Spy-1 radar and telemetry pods have in common? They operate using radio frequencies allocated for Defense Department systems by the National Telecommunications and Information Administration. NTIA can only grant permission to operate in specific frequency bands, and bands are occasionally targeted for reallocation by outside interests, both national and international. Spectrum allocations are globally harmonized through the United Nations’ International Telecommunications Union (ITU).
Every four years, the ITU holds the
World Radiocommunications Conference (WRC) to review and revise the international treaty governing the use of the radio frequency (RF) spectrum.
The next WRC will be held in November 2015, where reallocation of bands may lead to the Defense Department having to move equipment out of specific bands. Figure 1 shows the range of spectrum to be considered at WRC-15 and the number of Department of the Navy licenses it encompasses. The office of the DON Chief Information Officer has been strategizing how best to prioritize bands to protect department equities.
Problem Statement
What is the most accurate and efficient way to prioritize RF bands that are targeted for reallocation? The traditional approach is to send out a data-call to all spectrum owners asking for cost and impact estimates, wait for their responses, rack and stack the results, and then argue and adjust the answer.
The problem with this approach is that the process is manpower intensive, time-consuming, and results in a consensus decision not always backed by consistent measures. The problem is further compounded by the need for the DON to repeat this drill every few years because RF bands are continually targeted for adjustment.
In essence, the problem is two-fold.
- First, if asked to vacate some amount of spectrum, how would one determine which spectrum loss would have the least impact on operations and would cost the least to enact?
- Second, given how often this type of assessment must be made, is there a more efficient way to do this?
Can we look at data that exists for spectrum licenses and come up with a process that ranks frequency licenses in terms of their advantage to military operations and their cost to vacate? Is there an easy button? Can a decision-maker enter a list of targeted frequencies into an application and get an output ranking those frequencies in terms of suitability to vacate?
Last year, the Deputy Assistant Secretary of the Navy for Command, Control, Communications, Computers and Intelligence (DASN C4I) in the office of the Assistant Secretary of the Navy for Research, Development and Acquisition (RDA) and the DON CIO asked these questions of the Center for Naval Analyses (CNA): Can a methodology be devised that would rank the suitability of targeted RF bands by keying on basic data representative of military and cost impact, and furthermore could this method be performed using standard existing databases. The CNA is a federally funded research and development center for DON studies and analyses.
Findings
The CNA undertook this project and, using the Government Master File (GMF) as the data source, looked at Navy and Marine Corps license data. The GMF is the federal license data repository, where frequency assignment, equipment, location, operating specs, and other information is kept for all federal agencies.
The Center for Naval Analyses examined the data fields within the GMF and developed several proxy measures of effectiveness for both cost and mission impact. Some of the proxies used single data fields (e.g., counts of the number of licenses within a targeted RF band); other proxies used combinations of data (e.g., platform, function and equipment type). Combining proxies, the CNA was able to produce a relative ranking of the targeted RF bands. Because of the limitation of using a single database with cases of incomplete or ambiguous data, subject matter experts were still needed to review the rankings.
Overall the CNA was able to show that the concept of ranking bands using database information alone has merit and with some additional data inputs, more complete data, and with refined measures of effectiveness, a semi-automated process is feasible. Although results will still need some review, and effort must be spent to populate or update databases, there is potential to save time and money by implementing this automated process.
The Center for Naval Analyses will continue to refine the method this year and will be looking at data availability throughout DoD. Key issues for improving the ranking process include:
- Improving the quantitative measure of mission impact (e.g., are systems unique or are there contingency systems that operate on different frequencies).
- Improving the assessment of the regional reallocations (i.e., can systems retain licenses in specified geographic regions).
- Measuring the cost of reallocation as a function of whether the band is vacated or shared.
DoN CIO is not the only group pursuing these issues. The Defense Information Systems Agency (DISA) Defense Spectrum organization (DSO) is populating the Joint Spectrum Data Repository (JSDR), a multi-source compendium of spectrum-related information, and is planning future enhancements to allow user-specified queries of the data. This work will help provide a better understanding of military spectrum usage and needs, and allow the DoD to better protect the spectrum needs of the warfighter.
Stephen Munchak and Ariel Michelman-Ribeiro are research scientists at the Center for Naval Analyses.