This article was authored by John C. Tanner, and was originally posted on telecomasia.net.
ITEM: The US FCC has approved a plan for allowing spectrum sharing in the 3.5 GHz band for small cells. The band in question (3550-3650 MHz) is currently used for US Navy radar operations and covers 60% of the US population.
In a Notice Of Proposed Rulemaking (NPRM), the FCC proposed “a three-tiered licensing and interference protection framework” to manage access to and use of the 3.5 GHz Band (via a geolocation-enabled dynamic spectrum access database modeled on the existing TV white spaces database), providing different levels of protection for different levels of access.
Technology Review explains how the process would work:
Under the proposed rule, wireless carriers, corporate offices, or researchers could reserve pieces of that spectrum in different regions and at different times—a system managed by a central database. The approach guarantees that the spectrum will be available and not subject to interference in certain areas by a crush of new users, as might happen if the new chunk of spectrum were made available with no regulation at all.
Although the checked-out spectrum might be free of charge at first, eventually, depending on the level of demand, a pricing system could be implemented. In theory, this could allow a wireless carrier to pay for priority access in rare cases of extremely high demand. This would give it the right to temporarily evict another party—such as an academic lab, which might be willing to tolerate short-term interruptions in exchange for low- or no-cost access to the spectrum.
Technology Review goes so far as to say that the spectrum-sharing rule “spells the beginning of the end of a system in which spectrum is either exclusively owned by a private company, walled off for government and military use, or unlicensed and crowded.”
That might be putting things a little strongly, especially in the US, where major cellcos very much like the private-spectrum system and have been enthusiastically critical of the stated benefits of spectrum sharing.
On the other hand, it’s worth noting that recent spectrum auctions – particularly in the Asia-Pacific region – have been damp squibs (at least compared to the salad days of past auctions), indicating that at least some cellcos are losing their taste for buying additional spectrum – at least at the prices being asked for it.
For example, India’s recent 2G spectrum re-auction fizzled, with the government selling less than half of the 1800-MHz spectrum up for bid and earning less than a quarter of the revenue it had expected. Thailand’s recent 3G spectrum auctionfetched a piddly 2.8% premium. And Australia’s upcoming 4G spectrum is so overpriced that Vodafone Australia has bowed out and Optus says it will do the same if the government doesn’t lower the reserve price.
On the other hand, Monday's LTE spectrum auctions in The Netherlands yielded around eight times more than the government expected.
Meanwhile, the FCC isn't the only regulator pushing for shared spectrum. TheEuropean Commission wants to put together a shared spectrum plan by the middle next year.
Here in APAC, India is already doing shared spectrum for cellular, albeit only for 2G, and not without the usual controversy that comes with anything spectrum-related in India. Meanwhile, the Center For Internet & Society is also advocating the use of shared spectrum for Super Wi-Fi services using white-space spectrum.
Whether other regulatory regimes adopt spectrum sharing will probably depend on the usual factors – namely, whether existing operators see it as an opportunity or a threat, and whether regulators and ministers think they can still milk the spectrum cash cow for a decent return.
Either way, one thing to bear in mind is that from a technical viewpoint, spectrum sharing isn’t exactly a plug-and-play endeavor.
Over on Fierce Broadband Wireless, Peter Rysavy of Rysavy Research recently pointed out the challenges involved in spectrum sharing for LTE. And over on GigaOM, Rysavy also pointed out the security issues raised by dynamic spectrum sharing that will also have to be addressed.