The FCC organizes radio services by frequency. Each service AM, FM, satellite, cellular, radar, everything else is assigned to certain bands. And, in most services, each user is assigned to a specific channel in the band. Every walkie-talkie license identifies one particular operating frequency, out of hundreds. The entire usable spectrum is divided and subdivided into a vast number of individual channels. Most users are confined to one of them.
This form of regulation began as a matter of necessity. In the early days of radio, the only way to keep nearby users from interfering with each other was to place them on different frequencies. This suited the transmitters of the time, which could only operate on one frequency anyway. Each used a pre-tuned crystal or other components to set a fixed frequency, with the modulation and bandwidth hard-wired into the circuitry as well. Changing any property of the radio meant changing the hardware.
This approach worked well for decades. Even as demand proliferated, engineers opened up ever-higher reaches of the spectrum, and found ways to carry more voice and data in narrower bandwidths. But the transmitters remained band-specific. Today, the specifications of bands and modulations, among other properties, may be programmed into a digital signal processor, but the FCC still insists the transmitter be limited to specific bands and modulations, just as though it were still tuned by hardware.
Now the available bands are filling up again. Engineers are still pushing higher in the spectrum, but the equipment for operating at those frequencies is expensive, and its useful range is short. Further bits-per-bandwidth increases are still feasible, but are also expensive, and can only go so far.
Are there other options?
Despite all the talk about spectrum shortage, in fact much of the spectrum is vacant for most of the time. A land mobile channel, for example, is typically licensed to a single user. A busy channel might actually be in use for one-third of the working day, on a seconds-per-hour basis, but is silent the rest of the time. The spectrum shortage is actually a shortage of unlicensed channels, not of unused spectrum.
Current FCC rules are partly responsible for this waste of capacity. They generally prohibit anyone but the licensee from operating on a licensed frequency, even though most licensees do not fill their bandwidth. Moreover, the procedure for transferring a license from one entity to another typically takes several weeks. A few classes of licensees can enter into long-term management agreements, joint marketing agreements, and resale agreements. Leasing and other sharing arrangements are permitted in certain services, and the FCC recently allowed "band managers" to acquire certain specified bands for the purpose of leasing to other parties. But most spectrum is tied to the one user whose name is on the license.
The FCC is considering a rule change to facilitate spectrum leasing and other similar arrangements, so as to establish a "secondary market" in spectrum. Having acquired the right to use spectrum, the idea goes, a licensee could then lease some or all of that right to other users. Licensees could subdivide and lease out portions of their bandwidth, their geographic area of license, or their term of license, in any combinations they wish. Many such transactions could proceed without prior FCC approval.
The FCC likes this idea because it permits more efficient use of the spectrum. Licensees like it because they can make money from unused spectrum. Service providers like it because they can serve more customers. And the public will like it because more calls will get through.
A lot of details must be worked out. The licensee will presumably be responsible for making sure the lessee complies with the FCC's technical rules, especially those that protect other users against interference. But the FCC wants the right to proceed directly against the lessee as well for violations. Still an open question is whether lessees should be held to existing service rules, such as those governing qualification, eligibility, and use restrictions. Can frequencies now designated for private use be leased to handle commercial traffic? Should a licensee that received a small-business bidding credit be required to reimburse the FCC if it leases to a non-qualifying entity? Should a licensee's status as a common carrier or private carrier automatically carry over to its lessees? And, if the FCC decides service rules should not apply to lessees, then does applying those rules to licensees still make sense? A secondary market may break down the long-standing assignment of frequencies to specific services, so that frequency bands finally become interchangeable.
Software Defined Radios
Secondary markets may have another side effect as well: they may promote the development of radios capable of changing frequency or modulation on the fly.
It has long been feasible, at least in principle, to make a radio whose operating characteristics are under software control. But the FCC will not certify a software defined radio today. Its rules, geared to old-fashioned hardware transmitters, require the radio to be capable of operating only at a specific frequency range, bandwidth, output power, modulation, etc. With just a few exceptions, such as dual-band wireless phones, a unit capable of changing these specifications is ineligible for FCC approval. And even a dual-band phone is limited to just two modes of operation.
The FCC recently proposed to break with its policy and allow certification of radios whose operation can be modified by software. Within limits imposed by the hardware, these could be made expandable to new frequency bands and modulation methods simply through software changes. New software could even be delivered over the air for automatic upgrades, much as Internet web browsers automatically download newly available enhancements. Thus, for example, a wireless-phone service provider that acquires new spectrum can remotely reprogram its customers' handsets to operate on the new frequencies. If the provider wants to use a different modulation on the new frequencies, it can download software for that, too.
The proposed rules would require a manufacturer to test each hardware-software combination for compliance, much as it now tests separate radios. A new category of "permissive change" will allow continued marketing of the radio following a software modification that results in a change of frequency, bandwidth, output power, or modulation changes that today require a new certification. Manufacturers will be responsible for using encryption or other means to prevent users from performing unauthorized changes to the operating software. No one in authority wants to see a cell phone that converts easily to transmit on FAA ground-to-air frequencies. Yet, at the same time, the FCC wants to allow third-party vendors to develop specialized software applications. Details on how to achieve security and flexibility at the same time have yet to be resolved.
Some users will benefit from software defined radios even under the present frequency allocations. Ambulance attendants, for example, typically need half a dozen different radios to communicate with 911 dispatchers, police, fire, hospitals, and so on. Doubtless they would welcome one software defined radio that replaces all the equipment under the dashboard. Travelers could use a wireless phone that automatically adapts to service characteristics anywhere in the world. As wireless Internet services proliferate, demand will grow for a laptop radio that sniffs out available service and connects as needed.
Using Spectrum Efficiently
Along with developing higher-frequency portions of the spectrum, we can best relieve the spectrum "shortage" by finding ways to use bandwidth that otherwise lies idle. This is relatively easy for spectrum that will be vacant for years while an auction winner builds out its network, for example. Leasing is a natural solution in those cases, and can use conventional equipment. But the day is coming when spectrum-hungry providers will want to lease idle bandwidth by the second, perhaps even less. Software defined radios will make that possible. And the combination of software defined radios and secondary markets will enable providers to use every last hertz of spectrum for every last millisecond. The FCC is getting out of the way.
Mitchell Lazarus is a lawyer with the firm of Fletcher, Heald & Hildreth in Arlington, Virginia, www.fhh-telecomlaw.com. He has 17 years experience representing clients at the FCC, particularly on technical issues. He can be reached at 703-812-0440, or email Mlazarus@alum.mit.edu