Congress recently mandated that the Federal Communications Commission (FCC) make additional spectrum available through a novel incentive auction designed to transition television broadcast spectrum to mobile use. The FCC’s task is to adequately compensate television broadcasters for relinquishing their spectrum while ensuring such spectrum is rapidly transitioned to mobile uses that benefit consumers nationwide.
This will be the most challenging and complex auction design the FCC has ever attempted. The FCC cannot avoid the complexity inherent in this unique auction design, but it can emphasize simplicity and exercise restraint when considering the other service rules that will govern this new spectrum. To maximize its opportunity for success in this daunting circumstance, the FCC should leverage proven policies wherever possible.
Successful spectrum policies are critical to sustaining innovation, economic growth, and global competitiveness in the mobile era. Today, consumer demand for tablets and smartphones is straining the limits of mobile Internet capacity, which is threatening our nation’s lead in mobile innovation. The quickest and least costly way to increase mobile network capacity is to add spectrum, and the incentive auction is how the FCC intends to bolster our spectrum resources. The continued success of the mobile Internet thus depends on the success of the incentive auction, and the auction’s success depends on policy decisions that must be made by the FCC.
The world’s first incentive auction is not the time to reinvent spectrum policy from scratch. The FCC has already pioneered proven spectrum policies, including technological neutrality – a policy based on hard-won experience in the early days of cellular. Back in 1981, when the FCC established rules governing the first cellular service, it required that all cellular providers use a particular analog technology (the “AMPS” standard). The analog regulatory requirement remained in place for 27 years – long after its commercial viability had passed. By the time cellular providers were allowed to shut down their antiquated analog networks in 2008, many mobile broadband providers were already deploying 3rdgeneration mobile broadband technologies.
The FCC’s experience with the technological stagnation caused by its analog cellular requirement prompted it to adopt “flexible use” mobile policies beginning in 1993 with the Broadband PCS band and continuing through 2007 with the 700 MHz band. These flexible use policies allow spectrum licensees to make determinations regarding the services they will provide and the technologies they will use (including FDD- and TDD-based technologies) so long as they comply with the FCC’s technical rules. (For an explanation of FDD and TDD, see the “Technical Appendix” at the bottom of this post.)
Mobile service providers in the United States have generally supported the FCC’s technology neutral spectrum policies. For example, in 2004, when the FCC was considering a new band plan in the Broadband Radio Service, the “overwhelming majority” of commenters, including Sprint, argued that the band plan should be technology neutral. The FCC adopted Sprint’s view that the public interest would be best served by “not restricting the band to a particular technology,” which would allow “licensees and systems operators to deploy either FDD or TDD technology, and freely switch between the two as the technology develops and the marketplace demands evolve.”
Sprint recently reversed course. In its comments in the incentive auction proceeding, Sprint is asking the FCC to abandon its technology neutral approach to spectrum and mandate a TDD-only band plan. The result would be a regulatory prohibition on the deployment of FDD-based technologies in the new spectrum band. In addition to being inconsistent with proven policy, a TDD-only approach would prohibit proven practices. FDD-based deployments have predominated in the United States with excellent results. Though the rules governing the 700 MHz band are technology neutrality, the LTE deployments that made the U.S. the world leader in mobile broadband use FDD, and that trend is expected to continue.
The return to restrictive spectrum policies Sprint seeks would be unwise. Sprint’s attempts to emphasize the potential advantages of TDD technologies (e.g., dynamic downlink/uplink ratios) while minimizing their potential disadvantages (e.g., synchronization and coverage limitations) are largely irrelevant. Even if Sprint were “right” today (i.e., that the potential advantages of TDD technologies currently outweigh their potential disadvantages for all potential use scenarios), it could very well be “wrong” tomorrow. Spectrum technologies and uses can change more rapidly than regulations, and spectrum users should have the flexibility to adapt to those changes without seeking permission from the regulator – which is precisely the point of technology neutrality.
It seems counterintuitive for mobile operators to advocate for regulations that limit their future options. Unsurprisingly, the overwhelming majority of mobile providers commenting in the incentive auction proceeding last week supported the FCC’s proposal to adopt a band plan capable of supporting FDD-based technologies. Sprint was the only mobile provider that asked the FCC to limit the flexibility of mobile providers to deploy FDD networks.
I expect Softbank, the Japanese company that is acquiring Sprint, may have influenced this advocacy. Softbank was one of the founding members of the Global TD-LTE Initiative launched in 2011 and the first mobile provider to deploy a commercial TDD-based LTE network in Asia. To Softbank, advocating for a TDD-only band wouldn’t seem unusual: many Asian countries mandate the deployment of TDD-based technologies. But it is unusual here, where technology neutrality has helped make the United States the global leader in mobile broadband deployment.
There may be many valuable things we can learn from other countries around the world, but restrictive technology policies aren’t among them. We have several decades of our own experience with the damage that such policies can do to innovation, and nothing indicates that importing them today would produce different results.
What’s the difference between FDD and TDD technologies? A mobile radio can’t transmit and receive on the same frequency simultaneously. Two-way transmissions must be divided in some way. In mobile systems, they are either divided by frequency (FDD) or time (TDD).
FDD
Frequency Division Duplex (FDD) allows simultaneous communications in both directions using two (or more) frequency bands that are divided by a “guard band” (also known as a duplex band or frequency offset). One or more frequency bands are used for communications from “base stations” to subscribers (the downlink), and another frequency band is used for communications from subscribers to base stations (the uplink).
An analogy for FDD is a two-lane road that allows cars traveling in opposite directions to pass one another safely.
TDD
Time Division Duplexing (TDD) uses the same frequency band for downlink and uplink transmissions but divides them into different timeslots. Instead of using a “guard band” to divide the downlink and uplink frequencies, TDD uses a “guard interval” to divide the transmissions by time. TDD systemsemulate simultaneous communications by using brief timeslots that are not perceived by users.
A TDD system is analogous to a one-lane road with traffic controllers at each end.
FDD TDD Comparison
Both FDD and TDD have their advantages and disadvantages.
| Attribute | FDD | TDD |
| Spectrum | Requires two divided channels | Requires only one channel |
| Traffic | Uplink and downlink capacity is fixed by channel sizes (though channel sizes may be asymmetric) | Can adjust the ratio of uplink and downlink capacity |
| Distance | Not affected by distance | Guard interval increases with distance as signal propagation time increases, which decreases channel efficiency |
| Latency | No additional latency (Tx and Rx channels are always available) | Latency may increase with multiplexing due to switching times (between Tx & Rx) |
| Equipment | Additional filters and a duplexer are required | No significant additional costs (though a switch is required) |
| Synchronization | Adjacent channel interference is much lower | Systems must be synchronized to avoid adjacent channel interference |
A detailed discussion of the tradeoffs between FDD and TDD systems is beyond the scope of this blog post. In general, however, FDD may be preferable for mobile applications designed to cover longer distances when traffic is relatively balanced and synchronization issues are likely; TDD may be preferable for mobile applications designed to cover shorter distances when traffic is relatively unbalanced and synchronization issues are limited.
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology.