This is the latest in a series of white papers designed to spark conversation about the issues related to broadband for the public safety community. Check back on our blog for new posts each week, and feel free to add your comments below. Download a printable version of Broadband Conversations #22.
Broadband Conversations #22: The Fundamental Problem for States in Relation to FirstNet and the PSBN
I. Central Problem
The central problem for states in relation to FirstNet and the PSBN is the present inability to know with reasonable certainty whether or not FirstNet will offer states radio access networks (“RANs”) which truly meet public safety requirements at reasonable subscriber costs.
II. The Reasons for Uncertainty
The ultimate question or the question which drives the need for strategy is this: Whether FirstNet will provide states with a radio access network which truly meets public safety needs at an acceptable cost.
If that question could be answered now with certainty, the issues of strategy would be less complex and possibly even disappear, but: The inescapable fact is that no state now has enough information to answer that ultimate question with any reasonable degree of assurance. This conclusion is inescapable for reasons that fall into four categories:
• [Viability] FirstNet has itself raised questions regarding its own viability.
• [Construction Variation] FirstNet has itself indicated that it is considering constructing widely-varying number of sites (between 14,000 and 35,000), and that wide variation has direct implications for the satisfaction of public safety needs and for the resulting degree of reliance upon commercial wireless and satellite networks and the consequent incurrence of roaming charges.
• [Standards] FirstNet’s approach to RAN design diverges from certain applicable standards, and that divergence suggests that FirstNet’s RAN designs may not meet public safety requirements.
• [Consultation Process] The consultation process as envisioned by FirstNet may not be able, either in principle or in practice, to produce RAN designs satisfactory to the public safety user community.
III. Uncertainty Established
FirstNet has raised questions regarding its own viability.
• [No Business Plan] FirstNet has not disclosed its business plan and recently indicated that it has no business plan, but, rather, only a roadmap for the development of a business plan.
• [Spectrum and Financial Viability] FirstNet has acknowledged that its financial viability depends upon realization of the value of its “excess radio spectrum,” but has not yet been able, after a year-long heavily-staffed effort, to identify a plan to realize that value.
• [Question of “Excess” Spectrum] The value of that spectrum is very problematic especially in the largest metropolitan areas because, although in principle the spectrum is most valuable in those areas, in practice very little “excess” spectrum may be available in those highly-populated markets.
o [What Seybold Says] As Andy Seybold noted recently: “In major metro areas, until the NPSBN is up and running, it is difficult to convey to potential partners exactly how much spectrum will be available to them at any given time.”
o [What Seybold Does Not Say] What Andy does not say is that, if FirstNet builds a network of inadequate capacity, as seems likely in view of FirstNet’s sole focus on coverage and its ignoring capacity, the availability of “excess” spectrum will be proportionally reduced or eliminated. FirstNet does not seem to be aware that its limited construction plans undermine its spectrum leasing efforts because with inadequate capacity built, deployed spectrum will be heavily utilized leaving little, if any, left to lease. As a result, there will be less money available to FirstNet for construction which means that even the inadequate planned construction will have to be curtailed leaving even less spectrum to lease and consequently less funding. This is Joseph Heller’s “Catch 22.”
• [Roaming and Financial Viability] If FirstNet relies upon the availability of roaming to make up for deficiencies in coverage and capacity, high roaming charges could induce public safety subscribers to abandon FirstNet and to take service directly from a commercial wireless carrier with dire consequences to FirstNet’s revenue and financial viability.
B. Construction Variation
The extent of FirstNet’s network construction is uncertain and subject to wide variation.
• [Three Widely-Varying Levels] FirstNet has indicated that it is considering three levels of construction: 35,000, 24,000, or 14,000 sites.
o [No Real 35,000 Site Case] The 35,000 site case may be off the table because that was the number of sites assumed in the failed effort to produce a business plan.
o [35,000 Site Case Inadequate] Even the 35,000 site case is not nearly capable of producing a public safety broadband network that will meet applicable requirements as will be demonstrated later.
• [Consequences of Wide Variances] The widely-varying possible construction levels have a number of material implications, both performance-related and cost-related
o [Coverage and Capacity Provided] Those disparate construction levels must implicate wide variations in the degree of satisfaction of public safety needs in terms of coverage and capacity.
o [Degree of Reliance upon Roaming] Those disparate construction levels must implicate wide variations in the degree of reliance upon commercial wireless networks and satellite networks, at additional costs.
o [Cost and other Consequences of Reliance upon Roaming] Reliance upon commercial wireless networks or satellite networks for coverage or capacity will mean the incurrence of roaming charges which could make all-in subscription costs too high even if FirstNet’s own subscription charges are reasonable, and, while commercial wireless roaming costs can be quite expensive, satellite services costs will shock most public safety users (e.g., $4-$15/MB);
The nature of FirstNet’s design approach appears to diverge from certain applicable standards.
• [NFPA 1221] OEC with the support of FirstNet has established four levels of coverage quality, the highest of which only provides one wall in-building portable coverage, but that coverage level does not comply with the NFPA 1221, Standard for the Installation, Maintenance, and Use of Emergency Services Communications Systems (2013). That standard provides that the radio coverage required within structures shall meet certain minimum signal strength levels to and from a portable radio worn on the hip where those minimum signal strength levels depend upon the nature of the structure, where high-rise buildings require the highest minimum signal strength, and where one-wall in-building coverage would be non-compliant. RCC has worked with hundreds of Fire departments on the development and implementation of their land-mobile radio systems. Few, if any, of RCC’s Fire department clients would accept a one-wall level of in-building coverage. Most say that they need in-building radio coverage wherever the next fire is, quite a contrast with FirstNet’s proposed level of in-building coverage, particularly if the PSBN is ever to be relied on to provide public safety voice communications.
• [Requirements Engineering] FirstNet has not required the development of formal needs assessments even though the international standards [ISO/IEC/IEEE] and public safety standards [U.S. Department of Justice, Community Oriented Policing Services] make “requirements engineering” (formal needs assessments) mandatory in the development of systems and software. In fact, few, if any, public safety radio systems have, to RCC’s knowledge, been successful where they were not based on a proper needs assessment. Trade magazines are replete with articles about failed public safety land-mobile radio systems that were not based on actual user input.
• [Interoperability] Nationwide interoperability is by statute to be the hallmark of the PSBN, but FirstNet’s plan to rely upon satellite communications for rural areas will effectively undermine the legislatively-imposed interoperability standard because first responders from densely populated areas where satellite services are not used will not be able to “drop in” to rural areas and start communicating as one FirstNet representative indicated in his recent presentation to one of the larger states (http://www.youtube.com/watch?v=pdfIoN4JhYw) because the devices carried by those first responders will not include satellite connectivity for which they have no need in their home territories.
D. Consultation Process
The consultation process as envisioned by FirstNet may be unable to produce satisfactory RAN designs.
• [Top-Down/Bottoms-UP] FirstNet and OEC are taking a top-down approach to the consulting process by establishing a “coverage baseline” and grade of coverage of their own making instead of taking a bottoms-up approach and starting with the coverage requirements established in a proper needs assessment as required by applicable standards and public safety best practices. The old saying “You never get a second chance to make a first impression” applies here. If after years of development, public safety users begin using the PSBN in their state and find the coverage or capacity to be insufficient or too costly, they will be highly critical and may not use the network at all.
• [Capacity Misunderstood] FirstNet and OEC seem to assume that adequate coverage assures adequate capacity. That assumption is in error because, unlike LMR systems which are designed so that capacity does not vary with coverage, in LTE systems capacity declines with the distance from the cell site. Accordingly, while there may be coverage at the cell edge, the capacity at the cell edge will be much lower than the capacity closer to the cell site, and the capacity at the cell edge may be utterly inadequate to support required applications properly.
• [Capacity Ignored] FirstNet and OEC are entirely or substantially ignoring capacity in their approach to the consultation process. The March 10, 2014, detailed diagram of the consultation process issued by FirstNet never once mentions capacity or throughput. In addition, the current FirstNet Website page about consultation with the states makes no mention of capacity in the envisioned discussions with states (http://firstnet.gov/consultation).
• [Capacity Limitations Recognized] Until some pilot systems are up and operating, it is not completely clear how much data capacity will be needed for first responders. Some commentators have, however, offered attempts at quantification. Andy Seybold, in his Corner Stone Project in 2011, estimated that 6.3 Mbps uplink would be needed for one type of scenario (http://www.andyseybold.com). Tait Communications, in a presentation last year, estimated that an average scenario might require 4.8 Mbps uplink (http://www.radioclubofamerica.org/ 2013 Tech Symposium). Tait also found that while a 10 MHz x 10 MHz (the same as the FirstNet Band 14 system) LTE cell site could support up to 12 uplink video feeds of 1 Mbps each at 1.3 miles from the cell tower, that same site would only support ONE uplink video feed of 1 Mbps at 6.5 miles from the cell tower. In other words, at a little over 6 miles from the cell tower the LTE site could not nearly support an average incident’s requirements of 4-6 Mbps on the uplink. Thus preliminary indications, which need to be validated with more testing, suggest that a network designed for coverage only will not support even the basic data needs of routine incidents.
• [Population-based Design vs. Incident-based Design] FirstNet and OEC approach RAN design substantially on the basis of population density and assume that the required quality of coverage is a function of population density. That approach is closely associated with the design of commercial wireless networks which few public safety agencies rely on for their primary communications. By contrast, public safety networks are designed with a view to providing adequate coverage and capacity where incidents requiring a first response may occur which are not always areas of high population density (e.g., at a nuclear generating station in a sparsely populated area). Moreover, serious incidents can and often do result in a concentrated response of first responders whose capacity requirements are focused in a small geographical area. Assuming that those capacity requirements are a function of population density at the site of the incident is plain error.
• [Residual Resources] It is not apparent how much funding will be available to enable state-initiated additions to the baseline coverage plan. Available funding will be further limited if FirstNet, in search of users and revenues, diverts resources from providing service where needed by civil first responders to providing service to military bases.
IV. Response to Uncertainty
States have responded in diverse ways to the uncertainty relating to the ultimate question: whether FirstNet will offer states radio access networks which truly meet public safety requirements at reasonable subscriber costs.
• [Doing nothing]: Some states are simply paralyzed.
• [Following FirstNet’s lead without truly knowing where they will be led]: Some states follow FirstNet’s program without much by way of evidence to support the view that the program will produce a satisfactory RAN.
• [Answering the ultimate question positively without adequate foundation]: Some adopt a favorable disposition toward FirstNet and state they will “opt in” even though they have no information on the RAN they propose to opt into.
• [Answering the ultimate question negatively without adequate foundation]: Some states adopt an unfavorable disposition toward FirstNet and suggest they will “opt out” even though they have no information on the RAN they propose to opt out of.
• [A Few States] A few states understand that the question whether FirstNet will provide a state with a RAN that meets public safety requirements at a reasonable subscription cost is not now and will not, for some time, be certain and that, therefore, it is necessary for a state to hedge its bets and to prepare for the range of possibilities that the presently unavoidable uncertainty makes it necessary to consider.
Consider the following propositions as the conclusions to be drawn from the previous discussion:
• [Risk] The risk that FirstNet RANs will be unable to meet public safety needs is real. The risk that the limitations upon FirstNet RANs will drive public safety users to commercial wireless and satellite networks is real. The risk of the imposition of material roaming charges is real. The risk that public safety users in some areas will not use the PSBN is real.
• [States; Risk Management Responsibility] States have a responsibility to manage and to provide protection to itself and its constituent public safety agencies against those risks because only states are in a position to do so, and, in so doing, they must make their constituent public safety agencies an integral part of that risk management process.
With over 30 years of corporate experience in assisting hundreds of cities, counties, and states in the development, procurement and successful implementation of high quality public safety wireless communications systems, RCC fully supports the successful planning and deployment of the PSBN and the improvement of public safety communications for all first responders. RCC has developed a strategy for risk management in the context of the uncertainties relating to FirstNet’s current plans and capabilities and has also developed a unique set of processes and tools to implement that strategy. RCC’s processes, tools, and system development support are available to the public safety community and also to FirstNet to help dramatically improve the likelihood of success for everyone involved in development of the PSBN.
If you would like to learn more about how RCC’s strategy, processes, and tools, please contact RCC at http://www.rcc.com/contactus.shtml.
If you would like to learn more about why designing for capacity is important, download RCC’s 2014 IWCE presentation "Modeling Incident Scenarios on the Public Safety Broadband Network” at http://www.rcc.com/resources/presentations.shtml, where other resources which may be helpful are also available.