Licensed Access: The Smarter Policy for TV White Space and Broadband Internet Access By Charles L. Jackson and Dorothy Robyn Smart Radio: Smart Markets and Policies April 6, 2007 Context for Our Study Vacant regions ( white space ) will remain in TV band, even after DTV transition FCC has three options 1. Leave such spectrum space idle. 2. Allow access by unlicensed devices. 3. Auction off flexible, tradable rights (licenses). Debate has focused on the unlicensed option specifically, would it cause harmful interference to broadcasters? Our goal: better define and analyze the licensed option How much white space will there be? Comparison of UL and L options on three criteria: efficient use of spectrum, incentives for investment, and long term evolution of TV band Feasibility of licensed option 2 2 1
How Much White Space is Available? Our approach: Calculate a protected region around each station. Calculate available spectrum channel by channel in each census block group. Aggregate results to the MTA level. We calculate results under five scenarios, corresponding to different assumptions regarding likely FCC interference-protection standards Our fundamental view: FCC standards can be less protective in a licensed regime because of greater ability to control interference 3 3 Measure of White Space under Alternative Interference-Protection Scenarios Measures of White Space under Alternative Interference-Protection Scenarios Included Facilities Co-Channel Protection Adjacent- Channel Protection Average White White Space Space Total MHz-Pops in Percent of MHz- White Space Bandwidth Bandwidth Covering Bandwidth White Space Pops in White Covering 100% of Total 95%+ of Total Available (Millions) Space Population Population Nationwide 2-51 5-51 14-51 2-51 5-51 14-51 2-51 5-51 14-51 2-51 5-51 14-51 2-51 5-51 14-51 All US, Canadian, and Mexican Scenario regular and Class A stations and FCC Radius X land systems in the UHF TV None 53,678 49,232 37,829 64% 63% 60% 78 60 36 108 90 60 188 173 133 spectrum. All US, Canadian, and Mexican Scenario regular and Class A stations and FCC Radius Y land systems in the UHF TV FCC Radius 28,266 24,532 17,379 34% 31% 27% 0 0 0 24 12 6 99 86 61 spectrum. All US, Canadian, and Mexican Scenario regular and Class A stations, Z land systems in the UHF TV FCC Radius FCC Radius 27,156 23,523 16,547 32% 30% 26% 0 0 0 24 12 6 95 82 58 spectrum, and all TV translators. FCC Radius plus All US, Canadian, and Mexican 46, 30, and 17 Scenario regular and Class A stations and FCC Radius miles for low UL-1 land systems in the UHF TV plus 5 miles VHF, high VHF, spectrum. and UHF Scenario UL-2 All US, Canadian, and Mexican FCC Radius plus regular and Class A stations and 46, 30, and 17 land systems in the UHF TV miles for low spectrum. Channels 2-4 and 14-20 VHF, high VHF, excluded. and UHF FCC Radius plus 5 miles 21,028 18,093 12,752 25% 23% 20% 0 0 0 12 6 0 74 63 45 15,160 15,160 9,820 18% 19% 16% 0 0 0 0 0 0 53 53 34 4 4 2
For Scenario Z, we estimate there will be at least 12 MHz of white space available everywhere except in portions of the San Francisco MTA Minimum Bandwidth of White Space by MTA 300 Minimum MHz Available 250 200 150 100 50 0 10 8 114 90 84 84 78 78 60 60 66 66 66 66 72 72 72 72 72 54 54 54 54 48 48 48 48 48 48 42 42 42 42 42 36 36 30 30 30 24 24 24 24 18 18 18 18 12 12 0 San Francisco-Oakland-San Jose Philadelphia Los Angeles-San Diego New York Puerto Rico-U.S. Virgin Islands Tampa-St. Petersburg-Orlando Chicago Detroit Milwaukee Cleveland Charlotte-Greensboro-Greenville-Raleigh Washington-Baltimore Miami-Fort Lauderdale Salt Lake City Seattle Atlanta Buffalo-Rochester Knoxville Tulsa Boston-Providence Birmingham Denver Richmond-Norfolk San Antonio Dallas-Fort Worth Houston El Paso-Albuquerque Oklahoma City Indianapolis Nashville Alaska St. Louis Portland Pittsburgh Louisville-Lexington-Evansville Kansas City Little Rock Jacksonville Cincinnati-Dayton New Orleans-Baton Rouge Des Moines-Quad Cities Honolulu Columbus Memphis-Jackson Phoenix Minneapolis-St. Paul Spokane-Billings Omaha Wichita MTA (Reverse Rank Order by Bandwidth) 5 5 For Scenario Z, we estimate that 24 MHz would be available to more than 95% of the US population Percent of Population With Given Amount of White Space 0, 100% 100% 18, 9 9 % 24, 97% 30, 91% 90% 42, 82% 80% 60, 70% Percent of Population 70% 60% 50% 40% 30% 72, 6 3% 90, 50% 20% 10% 0% 100% of Pops 99.2% of Pops 97.1% of Pops 90.8% of Pops 81.9% of Pops 70.3% of Pops 62.9% of Pops 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150 156 162 168 174 180 186 192 198 204 210 216 222 228 234 240 246 252 258 264 270 276 282 288 294 50.2% of Pops MHz of White Space 6 6 3
For Scenario UL-2, we estimate there will be at least 6 MHz of white space available everywhere except in portions of seven MTAs Minimum Bandwidth of White Space by MTA 300 Minimum MHz Available 250 200 150 100 50 0 0 0 0 0 0 0 0 San Francisco-Oakland-San Jose Philadelphia Miami-Fort Lauderdale Los Angeles-San Diego New York Puerto Rico-U.S. Virgin Islands Houston 90 72 66 60 60 54 48 48 48 42 36 36 30 30 24 24 24 24 18 18 18 18 18 18 18 12 12 12 12 12 12 12 12 12 12 12 6 6 6 6 6 6 Tampa-St. Petersburg-Orlando Chicago Detroit Milwaukee Washington-Baltimore Birmingham Buffalo-Rochester Cleveland Charlotte-Greensboro-Greenville-Raleigh Seattle St. Louis Kansas City Tulsa Boston-Providence Nashville Louisville-Lexington-Evansville Cincinnati-Dayton Atlanta Little Rock Jacksonville Knoxville Dallas-Fort Worth Oklahoma City New Orleans-Baton Rouge Pittsburgh Denver Indianapolis Des Moines-Quad Cities San Antonio Memphis-Jackson Richmond-Norfolk Columbus Salt Lake City Phoenix Alaska Honolulu Minneapolis-St. Paul Portland El Paso-Albuquerque Omaha Spokane-Billings Wichita MTA (Reverse Rank Order by Bandwidth) 7 7 For scenario UL-2, we estimate that more than 5% of the population would have no spectrum. Percent of Population With Given Amount of White Space 0, 100% 100% 6, 92% 90% 12, 8 7% 80% 24, 73% Percent of Population 70% 60% 50% 40% 30% 30, 67% 42, 55% 20% 10% 0% 0 100% of Pops 92.1% of Pops 87% of Pops 72.9% of Pops 66.7% of Pops 55.4% of Pops 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108 114 120 126 132 138 144 150 156 162 168 174 180 186 192 198 204 210 216 222 228 234 240 246 252 258 264 270 276 282 288 294 MHz of White Space 8 8 4
How Much White Space Will There Be? Key Results Under Scenario Z, 97% of the population will live in locations that will have access to at least 24 MHz of white space Only one MTA (SF/Oak/SJ) has less than 6 MHz of white space in the worst-case census tract block group. Every other MTA has at least 12 MHz, and most MTAs have 18 MHz or more in the worst-case location. Under Scenario UL-2, only about half as much white space will be available Only 73% of pop will have access to at least 24 MHz Seven MTAs have less than 6 MHz in the worst-case location, and another six MTAs have only 6 MHz in the worst-case. The difference between Scenarios Z and UL-2 can be thought of as the loss in white space associated with using an unlicensed, as opposed to a licensed, approach. 9 9 A Licensed Regime Offers Three Key Advantages over an Unlicensed Regime More efficient use of the white space Better incentives for investment in wireless broadband Internet access Facilitates, rather than impedes, the long-term evolution of the TV band to higher value applications 10 10 5
Advantage #1: More Efficient Use of the White Space Licensed use would permit white-space licensees and broadcasters to negotiate expanded service. Such bargaining would permit more efficient operation. Such negotiations happen today not theory FCC standards could be less restrictive in a licensed regime. These two effects (less restrictive standards and negotiations ) should result in substantially more white space being used than in an unlicensed regime. 11 11 Advantage #2: Better Incentives for Investment Investment in long-range applications would be impeded because of the threat of interference Short-range wireless LAN operations could crowd out long-range operations. Such crowding out is already happening in the 2.4 GHz and 5 GHz bands The FCC s ongoing 3.65 GHz proceeding supports the view that lack of exclusive rights discourages investment in long-range infrastructure FCC rules call for non-exclusive licenses Service providers and manufacturers of equipment for long-range services (Intel, Motorola) oppose the rules because of the potential tragedy of the commons. By controlling interference, licensed use provides better incentives for the provision of long-range services, such as wireless Internet access, for which this spectrum is particularly well-suited Thus, a licensed approach is far more likely to produce the large investments in long-range infrastructure that rural broadband requires 12 12 6
Advantage #2: Better Incentives for Investment, cont d For short-range data transfer, unlicensed TV spectrum would be inferior to the existing unlicensed bands at 2.4 GHz and 5 GHz for the market needs served by today s wireless local area networks (LANs) Limited data rates reduce attractiveness. Network externalities (802.11) also favor the existing bands. Manufacturers will also face the cost of the capabilities needed to avoid interference with incumbents. Similar factors fatally impaired the unlicensed PCS (UPCS) band. Although unlicensed white space (unlike UPCS) would attract users, the likely short-range applications would not add significant value beyond those found in existing unlicensed bands today. 13 13 Scope of WLAN Services in White Space (Illustrative) 14 14 7
Scope of WLAN Services in White Space (Illustrative) Users Active Here Destroy Usefulness Here 15 15 WISPA Comments on White Space WISPA Comments on TV Whitespaces Further Notice We would encourage the Commission to create rules that promote fair and efficient sharing of the spectrum and efficiency of spectrum utilization. No more spectrum hogs! WISPA is opposed to any use of the Whitespaces for personal portable devices at this time... we do NOT wish to see a spectrum issue similar to the current 2.4 GHz WiFi band. In the 2.4 GHz band channel 6 has become all but useless for large-scale, wide area deployments... We do not wish to see this situation played out in any new bands. Also, we believe that it is NOT in the consumers best interests to have personal portable devices with propagation properties that would naturally allow them to pass through interior walls but also exterior ones... We believe that personal portable devices, especially in urban and suburban, markets would be best left to the higher frequency bands. 16 16 8
Advantage #3: Long-Term Evolution of the TV Band Licensed access would put in place a mechanism that would facilitate transition of TV bands to more productive uses ( Negroponte Switch ). By contrast, UL access would impede transition of TV band to higher value uses by creating a constituency of tens of millions of spectrum squatters and their suppliers; together with an embedded base of equipment that would interfere with any new use. UL advocates claim that TV band is a junk band and thus can only be used for UL operations is ironic the band is junk only because of perverse incentives created by the FCC s legacy command-and-control regulatory regime. An UL regime, through its necessary reliance on command and control regulation, would perpetuate the existing culture, and the attendant rent seeking and make sure it stays a junk band. One spectrum policy goal should be to put in place a regime that would refine the junk to gold. 17 17 Is it Feasible to License the White Space? Licensing of the white space is feasible FCC has conducted at least 14 auctions of white space in other bands. Licensed land mobile systems have operated in TV band for decades. Unlicensed advocates characterize licensing as infeasible. High transaction from Swiss cheese and uncertainties about users and use But, transaction costs would be a sign of efficient spectrum use. They would occur only if licensed operations were more extensive or efficient than unlicensed activity. Moreover, coordination activities need not be prohibitively complex or costly. The FCC typically structures overlay licenses with an eye to limiting coordination costs the same could be done with the white space. Licenses could be structured so that most broadcasters would deal with only a single licensee. Uncertainties about use (who and how) should not be an impediment to licensing. Well-designed rules would reduce uncertainties. The market will factor any remaining uncertainty into the price of the licenses. 18 18 9
A Licensed Regime Offers Three Key Advantages over an Unlicensed Regime More efficient use of the white space Better incentives for investment in wireless broadband Internet access Facilitates, rather than impedes, the long-term evolution of the TV band to higher value applications And it s feasible! 19 19 A Closing Thought from Another Commenter I am of the belief that if this TV whitespace is allowed to be utilized as 802.11x it will soon fall in the same category and be considered the trash of the spectrum. It is not uncommon to see noise floor levels as high as -70 in Monroe Louisiana as every one has a Linksys wireless router, a microwave oven, cordless telephones...etc. To turn all of these types of devices loose on such great spectrum would be an absolute shame. 20 20 10