Spectrum Inventory: “Same Bed, Different Dreams.”

I find myself in complete agreement with the wireless industry on supporting The Radio Spectrum Inventory Act. This Bill, sponsored by John Kerry (D-MA), Chair of The Subcommittee on Communications of the Senate commerce Committee, and Olympia Snowe (R-ME), Bill Nelson (D-FL) and Roger Wicker (R-MS), requires NTIA and the FCC to account for every MHz of spectrum between 300 MHz and 3.5 GHz within 180 days of the bill’s passage. You can see Kerry’s statement here, and Snowe’s statement in here.

“Same bed, different dreams” was the title of a book by David Lampton on Sino-US relations, and comes from a Chinese expression describing people whose lives are fundamentally intertwined but who do not fundamentally communicate with each other. That pretty much sums up supporters of more licensed spectrum and supporters of more unlicensed spectrum, both of whom very much back this bill. Kerry caught this duality nicely by saluting both the 700 MHz auction and the opening of the broadcast white spaces as demonstrating the value of spectrum access and the need for the inventory.

A bit more on this below . . .

Mind you, the fact that we need to have an act of Congress ordering this inventory is something of a disgrace. The bill would require a very detailed inventory of the range from 300 MHz to 3.5 GHz (the range of greatest interest to the cellular industry), “including information on the licenses or government user operating in each band, the total spectrum allocation of each licensee or government user, the number and types of radiators that have been deployed in each band, and contour maps illustrating signal coverage and strength.” You would think we would already have this information, right? After all, the NTIA exists to coordinate government use of spectrum, and the FCC requires a detailed license for every use of wireless not covered by the “unlicensed” rules of Part 15.

Sadly, given the vital importance of managing spectrum, we don’t have this information in anything like a useable form. The FCC has been (to be polite about it) extremely variable and inconsistent about how it maintains license information, how it puts this information online, and what information it requires from each licensee in any given service. Nip over to the FCC’s e-filing page and you will note a “Universal Licensing System” (ULS) alongside a few other databases for various services. Nor can anyone say with a straight face that the FCC has done any serious work to ensure that the information in the database is accurate. and yes, I get that inventorying the licenses and auditing them costs money. But ensuring the accuracy of spectrum licensing is supposed to be the whole point of the FCC’s licensing regime — lest we discover we don’t need licensing to avoid interference have horrible interference from poorly managed spectrum.

In the FCC’s defense, the federal government makes the FCC’s licensing databases look hyper-efficient and user-friendly. I’d like to blame NTIA, but NTIA finds itself in the less than enviable position of being responsible for overall federal spectrum use and coordination but without any actual authority over any other agency. In 2004 Bush ordered NTIA to inventory federal use of spectrum and develop a spectrum plan. It took NTIA until 2008 to compile a report amounting to generalities, wishful thinking, and almost 200 pages of vague handwaving by other federal agencies laying claim to various swaths of spectrum without much explanation as to what they use it for.

This frustration leads to two different sets of dreams in the same spectrum bed. The wireless industry thinks it can find some decent blocks of contiguous spectrum to auction off for more exclusive licensing. Meanwhile, supporters of unlicensed spectrum use (or at least various flavors of sharing and non-exclusive use) see opportunities to open up more spectrum in a variety of ways that don’t involve exclusive licenses. For example, the federal government could open the “federal white spaces” by allowing use in bands allocated for federal use but where there is no federal activity. Or the feds could borrow a lesson from Google’s proposal for realtime auctions and spectrum sharing from the 700 MHz auction.

Personally, I think the idea that we can find and clear a large enough band of spectrum to auction, let alone a set of paired bands, is not very realistic. Heck, even clearing the AWS band auctioned in 2006 has proven to be a royal pain in the neck for the auction winners. We’ve already cleared out all the easy to clear spectrum and even all the modestly difficult to clear spectrum in ranges useful for exclusive commercial licensing. We have hit peak spectrum, and we need to start developing more sustainable spectrum access models that reflect this reality.

But the believers in exclusive licensing won’t give up easily on the dream of some vast underutilized spectrum held by federal users who can be relocated and fitted together like Tetris blocks. They know its out there, so we should damn well try to find it. Even if I am right and they are wrong on the existence of blocks of spectrum easily cleared for auction, we still need an inventory. We can’t make a real spectrum development plan (which, frankly, ought to be part of the national broadband plan) until we have a clear idea of what’s really out there and how it’s being used. So no matter what flavor of wireless access one supports, a thorough spectrum inventory is the obvious first move for formulating spectrum policy.

For now, the bed is big enough for folks with different dreams of how to better manage spectrum access. The Senators sponsoring the bill have been careful in leaving open all possibilities once we actually have the data. With luck, we have a bill around which all sides of the “licensed v. unlicensed” debate can rally and push through quickly.

Stay tuned . . . .