Bandwidth is a hot—and limited—commodity. DoD will have to make a good case for why it needs access.
A recent article in Electronic Design sums up the exploding area of personal communications: "Nothing epitomizes the global nature of everyday life quite like satellite communications.... While many attribute the shrinking world to the Internet, satellites have been steadily shrinking our world for decades."
Alone, this is just an interesting observation. When joined with the next comment, however, the root of an ever-deepening crisis is made clear: "Regardless of the application, almost all satellite technologies have one major requirement—more bandwidth."
"Global-mobile" communications is the greatest "need" of today's world community, and the key to meeting that need is the electromagnetic radio frequency (RF) spectrum. Whether we are considering satellite, ground-based relay, or sensor communications, there is a common denominator—more bandwidth is needed. This burgeoning demand for bandwidth is driving the great collision between today's two spectrum-using worlds: commercial/consumer and governmental/ military.
Allocating the Resource
RF spectrum is a finite resource. There essentially is no unused spectrum, at least not in the frequency regions having practical application, and especially those of use to mobile communicators—meaning below about 3 gigahertz (GHz). The bottom line is that new uses of the practical regions of the RF spectrum must either replace or somehow be sandwiched into, on top of, or next to those already in place.
As a result, there is intense competition for access to the spectrum by wouldbe new users and equally intense resistance by current users. Entities must demonstrate the importance of the functions for which they employ the spectrum, and they must use the spectrum more efficiently. The best way for a user to get more operational effectiveness out of this limited resource is, of course, to use it more efficiently, and once a new user is allowed into a given band, the only way to avoid having to evict the incumbent user is to find a way for them both to use the spectrum without unacceptable mutual interference.
The debates and struggles extend to frequencies over 3 GHz as well, but the region below 3 GHz is the most significant. These lower bands have the highest performance capabilities and therefore are in the greatest demand by mobile users, both commercial and governmental. About 90% of all military frequency assignments are in the bands below 3.1 GHz. The reasons for this are tied up in antenna theory, the need for non-directional antennas on mobile platforms, and the relationship between frequency and antenna size. Frequencies above 3.1 GHz just do not work as well for many terrestrial applications.
Because demand for this spectrum band is acute and growing on both the commercial/consumer and government/military fronts, the problem quickly focuses on how to optimize access. We must either decide which use is more critical and relocate the other or find a way to share.
Military Spectrum Reliance
To address the first option—relocating users—it is important to understand DoD use of the RF spectrum. With few exceptions, modern military equipment depends in some way on access to the RF spectrum. Without adequate access, it cannot function properly. In addition, planning for future military operations hinges on current and emerging information technologies. The decision superiority called for in "Joint Vision 2020," for example, requires an uninterrupted flow of relevant information at the appropriate time—such as the provision of intelligence to operational commanders on demand. These strategic concepts enable an increase in our nation's military capabilities despite the significant reduction in DoD force structure, but only through a greater dependence on spectrum access. RF spectrum is the only medium that supports the mobility and flexibility the military requires to accomplish its missions.
So DoD needs access to RF spectrum to carry out its directed missions and, in most instances, those parts of the spectrum in which we currently operate are the most optimum. This brings us back to our two options: either we convince the spectrum authorities that DoD has a critical need for sole access to a given portion of spectrum, or we share it.
Given the pressing demand for worldwide global-mobile personal communications and the tremendous amount of capital involved in meeting it, convincing spectrum authorities that DoD has a critical need for sole access to certain regions of the spectrum would be a formidable task, especially in international forums. A number of newly reallocated bands recently were auctioned off for billions of dollars. That leaves only one real option: DoD must focus its energies toward sharing. This focus also should incorporate a long-range plan for increasing the amount of spectrum DoD has access to for executing its missions.
Sharing As a Way of Life
Spectrum sharing is not simply subdividing a given band into separate pieces, with each piece being used exclusively by a different user. Such sub-allocation actually is very inefficient. True spectrum sharing means two or more users use the same frequencies, relying on the available discriminants to permit compatible, interference-free operation by all. True spectrum sharing also implies sharing frequencies between military and commercial users regardless of allocation (contrary to some recent articles that suggest only the sharing of DoD spectrum with commercial interests, without mentioning the converse).
In principle, sharing seems rather simple—just do not use the same frequencies at the same time and in the same place. In practice, however, sharing is a complex undertaking. The time issue is particularly difficult. Most commercial enterprises will not accept scheduling constraints. Imagine, for example, a cellular telephone company telling its customers they can use their phones only during certain hours or on certain days. Also, instead of the more abbreviated voice communications that previously were the norm, today's mobile connection may involve significant and extended data transmission to/from the user's Internet service provider. The location issue is not much better. Mobile implies anywhere, especially now that it is global-mobile.
Fortunately, with modern signal processing, the use of power-level control, and the careful design of more powerful, interference-resistant receivers, users who are separated by very little distance often can use the same frequencies at the same time effectively and compatibly. And if sufficient separation cannot be achieved, "smart" radios can search for a band and find a channel that is not being used at that moment (a practice employed aggressively by personal communication systems). However, none of this happens by accident.
Systems must be designed expressly for their intended RF environment to permit a reasonable amount of sharing. The problem here is that designing for difficult sharing environments tends to increase the cost not only of the design effort but also of the resultant consumer equipment. Such cost increases are the enemy of the commercial developer, as price is an obvious factor in the marketplace. And when U.S. manufacturers do expend the effort to develop a more sophisticated product, they frequently face competition from foreign manufacturers whose devices have not been designed with RF spectrum sharing in mind and therefore are less expensive. If we want to maintain a level playing field, we need regulations that mandate smart design and that govern the sale and use of inferior products as well. If we are to make spectrum sharing a way of life, regulatory agencies must occur to bring it about properly.
Sharing, of course, is not feasible in every situation. Attempts to share frequencies/bands between systems intended for widely different purposes generally are unsuccessful because of significant technical differences. In some cases, geography and demographics combine to make sharing extremely challenging. The U.S. population is concentrated in certain major urban and coastal areas. When these areas are collocated with major DoD test and exercise locations, the geographic separation required between users to permit interference-free operation may be difficult to achieve. Nonetheless, it is important to study the possibilities in every case, as advances in technology often can effect solutions that were not possible in the past.
Spectrum Efficiency
Many speak glibly about bits and Hertz and how digital modulation will solve all our spectral efficiency problems, but the situation is more complex than it appears. Even agreeing on a definition of spectrum efficiency is difficult. Bits/Hertz seems to be an appropriate measure, but is a baby monitoring system really more spectrum efficient than a fire/rescue communications system, which must be designed for a much more demanding environment, just because the former can transfer more bits per second in a given bandwidth? How then does one factor in the missions of the systems being compared? One definition that has emerged is that the system under consideration be compared (on a bits/Hz basis) to the best, conceptual state-of-the-art system that could serve the same mission requirements.
In addition, the expected usage for a given system should be a factor. For example, if a system is used continuously, then is it not a more efficient use of the spectrum of concern than one that is used only rarely? One would think so, but what about the system that carries an alarm signal that is almost never exercised but critical when it is needed? These few examples illustrate the difficulty in defining the spectrum efficiency issue.
Digital modulation is another topic that should be addressed. In DoD, we consider all forms and types of modulation technology when new systems are being designed. When digital operation is the best solution, we use it. In fact, more than 60% of our systems are now digital. But although digitization of the baseband information does allow source encoding and information compression by eliminating much of the normal redundancy in common information formats, it does not always allow such compression. In addition, it is more complex, it costs more, and in some cases it suffers from interference susceptibility, requiring a higher signal-to-noise ratio than other, simpler forms of modulation. This is not to say that there is not a better way to operate than our traditional approach, and DoD is constantly researching new means to do its job even better.
Even in the best allocation and sharing situations, DoD may lose spectrum access in some cases, so we must find the best ways to use those parts of the spectrum that remain available to us. Indeed, even if we were never to lose more spectrum, we still would need to develop more efficient means to use what we have, just to account for the military's ever-increasing spectrum-dependent operations.
Spectrum Allocation Harmonization
Complicating the spectrum allocation issue, especially for DoD, is the fact that each country retains sovereignty to regulate the use of spectrum within its borders. Countries in the same International Telecommunication Union (ITU) region tend to allocate spectrum use in a similar fashion, but they all retain some national reservations or exceptions, as the United States does. Because military units deploy worldwide, DoD must know how each country regulates it spectrum resource to avoid causing interference to host-nation commercial systems and having their systems interfere with our military systems. This requirement causes difficulties domestically, as most other government agencies are not concerned with allocations in other regions and thus are not very sympathetic to this DoD-unique problem.
The other side of this coin can be just as difficult. Now that the ability to use the same mobile phone in any location around the world is beginning to be accepted as a reasonable goal, there is a strong push in the ITU and other international forums to allocate such bands identically in every country in a given region, and even in every region. This is called harmonization of allocations, and while it would ease the problems associated with disparities among national allocations, it would hinder our ability to use unique, U.S.-only allocation arrangements to operate on a worldwide basis.
We Must Get Out in Front
The collision between military and commercial spectrum access requirements is inevitable. It is, in fact, under way. DoD already has lost spectrum availability, but we have worked with the national authorities to identify those portions of spectrum (for reallocation) that were the least critical to DoD operations. By virtue of this work and the fact that much of the reallocated spectrum has not yet been put into use by the new "residents," the impact on operations thus far has been minimal. Unfortunately, very little of the remaining spectrum used by DoD could be called noncritical. The actions we believe are necessary—sharing, regulatory activities, development of higher spectrum efficiency systems, etc.—quickly are becoming critical to our future military operations. We must get on with these activities, and we must begin immediately. We must be out in front proposing strategies that protect national spectrum priorities, promote sharing, and transform us from the current contentious environment of frequency spectrum competition to one of mutual benefit for both military need and commercial opportunities.
General Raduege is director of the Defense Information System Agency and manager of the National Communications System.