Upgrade and Repack; Repeat Forever

I released a new ITIF report on spectrum policy last week at a star-studded event on Capitol Hill. You can get the report and see a video of the release event here and my slide deck here. While the report is 50+ pages long and the summary took nearly an hour, I can summarize the report in the five words that headline this post.

For quite some time, all the RF spectrum of interest to commercial service providers (most of it below 3 GHz) has been allocated. The strategy that has enabled new spectrum-based services to come about begins with incumbent services such as broadcast TV and radio, communication by satellite, specialized land-based two way radio services, and various government applications in both the civil and military sectors.

We free up spectrum for new services by improving the old ones. Almost every American adult is aware of the digital TV transition at some level, although most weren’t affected by it. Before 2009, 500 MHz of RF spectrum was assigned to television broadcasting (not counting satellite TV):

  • 54-72 MHz channels 2-4
  • 76-88 MHz channels 5-6
  • 174-216 MHz channels 7-13
  • 470-896 MHz: channels 14-83

After the transition from analog to digital TV, 200 MHz formerly occupied by UHF channels 52-83 (698-896 MHz) was reassigned to a variety of uses, mainly cellular and public safety networks, and White Spaces networking and wireless microphones were permitted between allocated TV channels in local markets according to a database.

On its face, the transition freed up 50% of the former allocation for analog TV (including the White Spaces allocation.) If we dig a little deeper, however, we find some more interesting implications.

For one, the digital transition started on June 12, 2009, but it’s not over yet. Cable TV companies have been required to transmit many channels in the old analog format to their subscribers, despite the fact that there hasn’t been any analog programming over the air for three years. They’re finally able to convert all their channels to digital, which will make management of the spectrum in their cables much easier. In case you didn’t know, the cable system is an RF broadcast domain that essentially mirrors the OTA regime. A number of low power TV stations still broadcast in analog, although they’re supposed to vacate or upgrade this year. The last vestiges of analog TV won’t be off the air before 2015, but we’re well underway.

Even more interesting is the fact that the allocation for each local TV broadcaster actually increased while the overall assignment for TV decreased. This is the effect of digital technology. While the broadcasters retain six MHz each, these allocations allow them to transmit anywhere from three to eight separate program streams, where they could only transmit one program apiece before. The flexibility comes about from the format and compression options the the broadcasters have. The six MHz assignment translates into a bit stream of 20 Mbps, which allows eight programs to be transmitted at 2.5 Mbps for SDTV, two streams at 10 Mbps for HDTV, or any combination that doesn’t exceed 20 Mbps.

So before the transition we had the capacity for a total of 82 possible TV programs, but afterwards we have the ability to broadcast as many as 392, roughly a five times increase. These figures aren’t per market because DTV channels can be more closely packed; on a per-market basis, the increase is two to eight times, depending on the encoding choices that broadcasters make. It’s fair to say the increased program capacity is at least triple, so we can now have three times as many TV programs moving over the air in half the spectrum that was formerly used.

An upgrade of this kind is easy for policy makers to mandate. The broadcasters can’t say no to three times as many TV programs, and the larger interest outside the TV world was happy to get access to 300 MHz of “new” spectrum. Regulators got a special bonus out of the deal because they got to decide how to divide the “digital dividend” of new spectrum. They handled the dividend in the traditional manner: a little here for cellular, a little there for public safety, and a little more for unlicensed use for super wireless LANs and rural broadband. And we’re not done yet: The broadcasters have the opportunity to team up with each other to share channels and auction off excess spectrum in the “incentive auction” recommended by the National Broadband Plan and authorized by Congress. Instead of your local NBC station sending one good program and two filler programs, it can share spectrum with the local CBS station, dispense with the filler, and bank a large check for spectrum rights given them for free. That’s good work if you can get it.

This system is called “upgrade and repack,” and it works for every system that uses spectrum. The results are not always going to be as dramatic as they were in the case of DTV, but they can be significant. The DTV upgrade improved spectral efficiency by six times, more or less. Upgrades to Wi-Fi between the original 802.11 standard in 1997 and the 802.11g standard in 2003  increased spectral efficiency by more than 20 times, from 2 Mbps to 54, without using additional spectrum. 802.11a and 802.11y allow Wi-Fi to use alternate bands, at the same level of efficiency as 802.11g. We don’t get into fatter channels for Wi-Fi until 802.11n, effectively in 2008, and 802.11ac, effectively this year. By now, the DTV standard used in the U. S. (ATSC) is already obsolete so we really should be looking at a second DTV upgrade that follows the technology curve.

Technology upgrades are routine for players who pay for spectrum rights, such as the cellular providers. The first cell phones were analog devices, and they were replaced by 2G digital devices, which in turn were rendered obsolete by 3G and 4G devices. Old devices aren’t made non-functional right away, but it’s sensible to regard cellular as driven by a perpetual upgrade cycle, as Wi-Fi is. Last week, AT&T announced it’s going to sunset its 2G service by 2016 in order to re-use 2G spectrum for 4G LTE. Sensible decision. Satellite TV networks are also in a routine upgrade cycle, replacing MPEG-2 encoding with MPEG-4 to pack higher resolution programming into their limited spectrum allocations as well.

So when we look at the spectrum landscape for additional bands to use for mobile broadband service, we’re looking for two opportunities:

  1. Applications that are no longer needed, such as satellite phones; and
  2. Applications that can be made more efficient by the application of technology upgrades.

This is abundantly clear, or should be clear, to everyone who’s involved in spectrum policy. So when the President’s Council of Advisors on Science and Technology (PCAST) declare, as they recently did, that the traditional method of making new spectrum available is “unsustainable,” they’re mischaracterizing  the system:

This report…concludes that the traditional practice of clearing government-held spectrum of Federal users and auctioning it for commercial use is not sustainable.

In fact, the traditional practice is “upgrade and repack,” and it’s extremely sustainable: its critical input is the freeing of spectrum by the advance of technology. Wi-Fi got twenty times better between 1997 and 2003, so isn’t it reasonable to require federal systems to improve over time as well?

All we need to make that happen is the right incentives in the right places. The technology is already there.



  • Rob Colburn

    Great article Richard. I’ve been curious, does wireless spectrum have issues with fragmentation? Whenever I see one of those spectrum map graphics, it looks like a file partition in need of a defrag. Do the various wireless standards 2G, 3G, … function better when they have a solid chunk of spectrum to use rather than tiny slices all over the map? I remember learning a long while back that higher frequencies transmit bigger payloads of data easier, but have more issues with interference. So, it would seem difficult to build a network with such a wide range of frequencies to use.

    Btw: I’m a big fan of the blog, have been since George Ou came aboard.

  • Richard Bennett

    LTE has some features that allow it to aggregate spectrum in small chunks, but it’s more efficient to have larger chunks to work with. For mobile networks, the spectrum between 500 MHz and 3 GHz is most prized because of environmental issues and antenna size.

  • […] Upgrade and Repack; Repeat Forever « High Tech Forum. Share this:TwitterFacebookLinkedInEmailPrintLike this:LikeBe the first to like this. This entry […]

Comments are closed.