In a bold move, the President’s Council of Advisors on Science and Technology (PCAST) recommended for federal spectrum to be shared between federal and private sector. This radical departure from “business as usual” when it comes to spectrum management holds much promise for accelerating innovation in the wireless industry.
But what is the motivation for spectrum sharing? First, it’s cost. It takes too long to clear a band and move incumbents to another one; the disruption is too high (e.g. NTIA – National Telecommunication and Information Agency – concluded it would take 10 years and cost $18 billion to clear a 95 MHz band). Second, the net revenue realized by auctioning spectrum is relatively small (e.g. estimated net revenue to federal government for 2006 AWS auction was $5.35 billion amortized over 10 years.) Third, the need for spectrum by government is increasing (e.g. the number of unmanned aerial systems operated by the Department of Defense has increased from 167 to nearly 7500 from 2002 to 2010, with greater volumes of intelligence, surveillance, and reconnaissance data.)
The claim of spectrum scarcity is rejected as an artifact of the present auction practice. Spectrum sharing, rather than exclusivity, is therefore not only a mean to achieve better utilization of spectrum, but can also yield higher revenue as it will offer greater opportunity for relicensing. The recommendation was to release 1,000 MHz of spectrum to increase the effective capacity by 1000x. Identifying such large amount of spectrum may not be as daunting as by NTIA estimate, 60% of frequencies between 225 and 3700 MHz are allocated to Federal use (629 MHz or 18.1% is exclusive to Federal users, 1058 MHz or 30.4% is exclusive to non-Federal users, remainder is assigned primarily to Federal users and secondarily to the private sector.)
Spectrum use by Federal Agencies
Spectrum sharing is used today in certain applications. For example, WiFi systems in the 5 GHz band implement DFS (Dynamic Frequency Selection) to avoid radars. 700 MHz TV band White Space spectrum mandates geo-location capability, although the availability of such devices and systems making use of this spectrum is very limited at the current moment. Medical devices can share spectrum with radar in 413–457 MHz and with aeronautical telemetry systems in the new Medical Body Area Network 2.36-2.39 GHz band.
The faith in such efficiency is bolstered by new technological and architectural advancements. First is small cells architecture which reduces the potential of interference. In fact, UK regulator Ofcom inquired last year about a shared spectrum band for small cells as part of the preparation for the upcoming 800 and 2600 MHz spectrum auction. Second is enhanced tolerance to interference. Taking advantage of these innovations requires large blocks of spectrum spanning several hundred megahertz rather than fragmented exclusive blocks. The 1000x capacity increase over current usage is achieved over large bands by dynamic sharing techniques that optimize the use of frequency, geography, time and other physical properties of the new radio systems. Cognitive radio capabilities (e.g. sensing and dynamic spectrum access) as well as geo-location would feature prominently in future wireless systems, as would time division duplex (TDD) instead of frequency division duplex (FDD) which is almost universally used today in wide area access networks.
That’s why this recommendation is bold and implementing it will surely define the future of wireless communications. The central essence of spectrum sharing is not just a way to manage spectrum, but will result in developing new technologies that are more efficient in their use of wireless spectrum. It follows also that a new business model can evolve for operating companies that would take advantage of this arrangement. It is no surprise then that large mobile operators such as AT&T have been vocal in their criticism of such plans. Spectrum sharing has the potential to fundamentally change the mobile service value chain. I believe that spectrum sharing would unleash a new era of development specifically to address the myriad of technical and other issues that need to be solved.
That’s why I think we can be at the cusp of a new revolution that can take wireless to new heights.