Wi-Fi in the TV White Spaces - 802.11af task group underway

The commercial success of unlicensed devices in TV whites spaces remains an open question, but Wi-Fi support could tip the balance, so it's good the IEEE standards association has officially chartered an 802.11 task group to:

... create an amendment whose implementation in solutions is likely to receive FCC approval for operation in the TV White Spaces under the 47 CFR Part 15 subpart H rules.

The group is 802.11 "TGaf." Their first meeting will be in Los Angeles the week of January 18th and their initial schedule calls for final votes on a new spec by the summer of 2011. My guess is they'll hit that 2011 date since many of the more difficult white space issues were figured out for 802.11y. In particular, 802.11y specifies "dependent station enablement (DSE)" where low cost devices (clients or access points) can operate under the supervision of a more intelligent (e.g. more expensive) device that actually consults FCC-mandated databases and provides sophisticated sensing.

TV White Spaces may yet be a commercial failure

At least as defined so far, TV white space rules are so restrictive that a market may never emerge. Indeed TV white spaces could follow the same path as Ultra Wide Band (UWB) - much hoopla, some significant investments, but no significant commercial success.  As specified so far, the only places where meaningful amounts of white space spectrum are available are rural areas where the population is small. This means sales volumes will be small and prices will be high - not exactly a formula for commercial success.

The good news is Wi-Fi operation will inherit much of the high volume silicon advantages created for other bands; only the actual RF amplifiers and antennas are specific to TV frequencies. If anything can succeed, Wi-Fi should be it.

Applications that could benefit from white spaces operation

Wi-Fi at TV frequencies could be useful for:

1. Rural broadband access for subscribers in densely forested areas.  If there is a line of sight, Wi-Fi at 5 GHz is more useful (and has more capacity and low cost) but when subscribers are hidden by trees, TV frequencies are scattered less.
   .
2. Wireless LANs inside heavy masonry buildings or those with plaster on metal wire lath.

In most other applications, 5 GHz is preferable to TV frequencies.  That may sound nuts to wireless engineers used to free space path loss calculations, but those calculations assume the antenna gets smaller as the frequency goes up.  With comparable antenna apertures, path loss in the atmosphere is roughly flat from 50 MHz to nearly 10 GHz.  Thus in open air, 5 GHz photons go just as far as 500 MHz (US channel 19) photons or indeed photons for Channel 2 or Channel 50. 

There's also a lot more spectrum available at 5 GHz.  What's more it's easier to form highly directional radio beams at 5 GHz than it is at TV frequencies.  Finally, at 5 GHz you need less open area around a line-of-sight transmission path, i.e. the Fresnel zone is smaller.

The primary place where 500 MHz does better than 5 GHz is going through heavy masonry.  The US National Institute of Standards and Technology (NIST) has done detailed measurements of how radio waves are attenuated as they pass through various building materials.  Their report is here.  Masonry significantly obstructs radio signals but it's much worse at 5 GHz than it is at 500 MHz.

Dense forests also obstruct radio waves.  The situation is a lot more complex as the randomly distributed leaves, twigs, branches and tree trunks cause attenuation, scattering, diffraction and absorption. There's been quite a bit of study of radio in forests, both for radio communication and for satellite observation of natural resources.  A good summary is here.  In short, wet forests are more of a problem than dry forests and lower frequencies do better than higher frequencies.

Wi-Fi:  the best bet for commercial success with TV white space

Wi-Fi has several years lead over WiMAX or LTE in deployment of so-called 4G technology.  In addition, the Wi-Fi market is large so prices are low (versus WiMAX or LTE where handsets may eventually be low cost, but infrastructure is expensive).  If there there is commercial success in the TV white spaces, it's most likely to be with Wi-Fi, in the 2012-2014 time frame.

Comments

[Peter Stanforth]

The final rules for use of TV White Spaces are being debated, so anything at this stage is somewhat speculative. However I would suggest that the real breakthrough here is not the amount of TV White Space that is available on day one but how easy it will be to add to it in the future. Once a class of frequency agile radios exist that use a database to find out where they can operate it is relatively easy to add to the spectrum pool. The FCC no longer has to find nationwide chunks of spectrum to create new unlicensed bands. Think of a scenario where there is "white Spaces" congestion in a major urban area. The FCC could identify some additional spectrum locally, maybe a single TV channel in that urban area, and add it to the pool. The effect on the community would be reduced congestion but with no change to any of the network infrastructure or operation.

Your analysis of the characteristics and assumption that using this spectrum as additional WiFi bandwidth (with increased propagation) is reasonable but I would not discount other new applications. The FCC is not defining the technology or the service so it is possible that, like the ISM bands, different types of service (like bluetooth, cordless phones and WiFi) will co-exist. Several areas of interest, beyond WiFi, include smart grid and smart city applications as well as tele-medicine.

[brough]

Thanks Peter. I agree the long term breakthough will be access to the white space that is spread throughout the spectrum. As I've pointed out on several occasions (http://su.pr/5wYiJv, http://su.pr/1kSY2q), while virtually all spectrum is licensed to somebody, most wireless spectrum is unused at most locations most of the time. The first step for shared access on a white space like basis was the FCC's action on the 3650-3700 MHz band and then the 802.11y protocols for using that spectrum. Out biggest efforts going forward should be to obtain secondary use of additional spectrum on a similar basis.

I agree that other users like Bluetooth or cordless phones or some other as yet unknown service could also benefit from TV white spaces, but I'll bet the first uses are Wi-Fi and WiMAX as rural broadband is a clear market and at least the Wi-Fi community already has most of the protocol work done (via DSE in 802.11y).