Before the FEDERAL COMMUNICATIONS COMMISSION

Washington, D.C. 20554

________________________________

) In the Matter of )

) Unlicensed White Space Device Operations ) ET Docket No. 20-36 in the Television Bands )

) _________________________________ )

COMMENTS OF MICROSOFT CORPORATION

Paula Boyd Senior Director, Government and Regulatory Affairs Michael Daum Director, Technology Policy, CELA Privacy and Regulatory Affairs MICROSOFT CORPORATION 901 K Street NW, 11th Floor Washington, DC 20001 (202) 263-5900

Paul Margie Paul Caritj Joely Denkinger HARRIS, WILTSHIRE & GRANNIS LLP 1919 M Street NW, Suite 800 Washington, DC 20036 (202) 730-1300

May 4, 2020

i

Table of Contents

INTRODUCTION AND SUMMARY.................................................................................. 1 I. The Commission Has Broad Support to Move Forward With Its Proposed

Rules............................................................................................................................... 3 II. The Commission Should Adopt Its Proposals to Optimize the TVWS Rules............... 12

A. The Commission’s Proposals to Allow Fixed White Space Devices to Operate at Higher Power Levels and Heights Above Average Terrain Will Expand Broadband Coverage............................................................................ 12 i. The Commission should adopt its proposal to increase the radiated

power limits for fixed devices in less-congested areas........................................ 13

ii. The Commission should adopt its proposed increase in HAAT limits in any areas where it would not cause harmful inference to licensees................ 15

B. The Commission Should Adopt Its Proposal to Allow TVWS Operation on Mobile Platforms Within Geofenced Areas.......................................................... 19

C. The Commission Should Adopt Its Proposal to Clarify the Framework for Narrowband TVWS IoT Operations.......................................................................... 25

D. The Commission Should Permit TVWS Operations at Increased Power Levels on First-Adjacent Channels While Maintaining Conservative Protections for Broadcasters...................................................................................... 28

CONCLUSION...................................................................................................................... 40

1

INTRODUCTION AND SUMMARY

Microsoft Corporation strongly supports the Commission’s proposal to revise the rules

for TV White Spaces (“TVWS”) devices. The rules described in the Notice of Proposed

Rulemaking (“NPRM”) would enable expanded broadband coverage in rural and underserved

communities, improve application of TVWS technologies for smart agriculture and other

industries, and open the 600 MHz band to Internet of Things (“IoT”) devices.1

White Space devices (“WSDs”) deliver internet connectivity over longer distances than is

possible with other broadband technologies because they operate in lower frequencies than any

other unlicensed technology. This reduces the costs that Internet Service Providers (“ISPs”) must

incur to deploy networks that reach Americans in rural and remote areas that are too far from

their current coverage areas to be served using other technologies. For this reason, Microsoft and

its many ISP partners, through the Airband Initiative, rely on TVWS technology as a key tool for

closing the digital divide, and believe that it would achieve even more if it was not held back by

overly conservative technical rules.

TVWS technologies are already serving rural Americans in communities around the

country. But the Commission’s proposal recognizes that a set of practical changes to TVWS

technical rules would further advance our nation’s goal of increasing broadband access without

harming incumbents. These improvements would do so by allowing providers flexibility, but

only where doing so would not, in the real world, result in increased harmful interference.

Service providers, equipment makers, consumer advocacy and trade groups, and members of

Congress agree. The changes proposed in this NPRM are based on the real-world experience of

1 Unlicensed White Space Device Operations in the Television Bands, Notice of Proposed

Rulemaking, 35 FCC Rcd. 2101 (2020) (“TVWS NPRM”).

2

small, medium, and large companies in dozens of states working to expand broadband, and will

facilitate significant improvements in coverage areas and enable key innovative applications, all

while maintaining robust interference protections for licensees.2 Although Microsoft is not and

does not plan to become a broadband provider, it is committed to partnering with these ISPs to

use the Commission’s improved rules to expand broadband coverage.

The Commission has broad support for the rule changes it has proposed and for moving

forward with an order adopting the revised rules as soon as possible. This is because the

proposed changes will:

• Enable increased coverage and additional deployment locations in rural and remote areas by increasing the radiated power limit for fixed WSDs in less-congested areas to 16 watts EIRP, and by increasing the height above average terrain (“HAAT”) limit from 250 meters to 500 meters.

• Facilitate innovation in the smart agriculture and mining industries and provide additional connectivity opportunities for people living in rural communities by allowing TVWS operation on geofenced, mobile platforms at the power limits that apply to fixed devices.

• Accelerate and improve machine-to-machine networks by adopting a clarified

framework for narrowband operations and by permitting narrowband IoT applications.

Broadband providers across the country also support Commission action to allow fixed

WSDs to operate at higher power levels on unoccupied frequencies that are closer to an occupied

broadcast channel. The use of well-established propagation models, and the recent engineering

analysis and field tests described in these comments, demonstrate that such operation would

2 Microsoft’s proposals are also informed by the experiences of other Airband partners from

deployments around the world, including in areas of United Kingdom, Ghana, Kenya, Colombia, India, and many other countries. See id. at Statement of Commissioner Brendan Carr (noting the “power of connectivity” that can result from a TVWS network).

3

improve broadband coverage for American communities without risking harmful interference to

broadcasters.

Accordingly, Microsoft requests that the FCC adopt the changes to its rules described in

these comments. Removing the regulatory barriers that stand in the way of more robust TVWS

broadband deployments will allow the Commission to advance its goal of making broadband

internet access services available to all Americans.

I. THE COMMISSION HAS BROAD SUPPORT TO MOVE FORWARD WITH ITS PROPOSED RULES.

A diverse array of stakeholders—including broadband providers, technology companies,

public advocacy groups, and trade associations—support the Commission’s proposed

improvements to the TVWS rules.3 The proposed changes also enjoy unanimous, bipartisan

3 See, e.g., Comments of ACT | The App Association at 2, ET Docket No. 14-165 (filed June

10, 2019) (“App Association Comments”); Comments of Adaptrum, Inc. at 2, ET Docket No. 14-165 (filed June 10, 2019) (“Adaptrum Comments”); Comments of Connect Americans Now at 2, ET Docket No. 14-165 (filed June 10, 2019) (“CAN Comments”); Comments of Dynamic Spectrum Alliance at 3, ET Docket No. 14-165 (filed June 10, 2019) (“DSA Comments”); Comments of Evolve Cellular, Inc. and Skylark Wireless LLC at 5, ET Docket No. 14-165 (filed June 10, 2019) (“Evolve/Skylark Comments”); Comments of Nominet at 1, ET Docket No. 14-165 (filed June 10, 2019) (“Nominet Comments”); Comments of Open Technology Institute at New America, Next Century Cities, Gigabit Libraries Network, Tribal Digital Village, and Public Knowledge at 1–2, ET Docket No. 14-165 (filed June 10, 2019) (“Public Interest Comments”); Comments of RADWIN LTD. at 3, ET Docket No. 14-165 (filed June 10, 2019) (“RADWIN Comments”); Comments of Rise Broadband at 1, ET Docket No. 14-165 (filed June 10, 2019) (“Rise Broadband Comments”); Comments of 6Harmonics Inc., Agile Networks, Cal.net, Declaration Networks Group, Evolve Cellular, Fairspectrum Oy, Network Business Systems Inc., Nextlink Internet, Packerland Broadband, RADWIN, RTO Wireless, Sacred Wind Communications, Inc., Skylark Wireless, Vistabeam Internet, Watch Communications, and WON Communications at 1, ET Docket No. 14-165 (filed June 10, 2019) (“Rural Partners Comments”); Comments of Sacred Wind Communications, Inc. at 1–2, ET Docket No. 14-165 (filed June 5, 2019) (“Sacred Wind Comments”); Comments of the Wireless Internet Service Providers Association at 2, ET Docket No. 14-165 (filed June 10, 2019) (“WISPA Comments”); Comments of Wi-Fi Alliance at 3, ET Docket No. 14-165 (filed June 10, 2019) (“Wi-Fi Alliance Comments”).

4

support at the FCC. As Chairman Pai has noted, the FCC’s NPRM will be a “game changer” for

broadband coverage, spurring significant growth of the White Spaces ecosystem and extending

internet connectivity to more Americans.4

As the Commission has recognized, advancements in technology to provide affordable

and accessible broadband service can help effectively address the digital divide in the U.S.5

These solutions are critically needed, as the Commission’s 2020 Broadband Deployment Report

notes that “the gap in rural and Tribal America remains notable,” explaining that 18.3 million

Americans lack access to broadband internet, including 22.3 percent of Americans living in rural

areas.6 Other estimates put the number of unconnected Americans closer to 42 million.7 The

COVID-19 pandemic has further highlighted the essential nature of internet access—and the

consequences of lacking access—as millions of Americans work, attend school, shop, and

receive medical care from home using their internet connection. Rule changes and technological

innovations that will facilitate broadband coverage for unserved Americans are more critical now

than ever before. Expanding connectivity will provide an immediate and dramatic benefit to

many consumers that need internet access.

4 See TVWS NPRM at Statement of Chairman Ajit Pai. See also id. at Statement of

Commissioner Michael O’Rielly (affirming the “wide variety of use cases” that the proposal will allow); Statement of Commissioner Geoffrey Starks (stating that broadband operations in “unused white space between TV stations present[] a valuable opportunity that could significantly change the wireless communications landscape”).

5 See id. ¶ 1, see also id. at Statement of Chairman Ajit Pai, Statement of Commissioner Brendan Carr, and Statement of Commissioner Jessica Rosenworcel.

6 Inquiry Concerning Deployment of Advanced Telecommunications Capability to All Americans in a Reasonable and Timely Fashion, 2020 Broadband Deployment Report, FCC No. 20-50, GN Docket No. 19-285, ¶¶ 36, 94 (rel. Apr. 24, 2020).

7 See BroadbandNow Research, FCC Reports Broadband Unavailable to 21.3 Million Americans, BroadbandNow Study Indicates 42 Million Do Not Have Access (Feb. 3, 2020), https://broadbandnow.com/research/fcc-underestimates-unserved-by-50-percent.

5

In partnership with private and public sector stakeholders, Microsoft is working to

harness the power of White Spaces technologies to improve the economics of delivering

connectivity to rural America. Less than three years ago, Microsoft launched the Airband

Initiative, which aims to facilitate that goal. Through the Airband Initiative, Microsoft is working

with local and regional ISPs to extend broadband access to three million people in unserved rural

areas of the United States by July 4, 2022.8 More specifically, the Airband Initiative creates an

ecosystem that makes it easier to deploy broadband networks using TVWS technologies

alongside other wireless and wired technology solutions. Microsoft, however, has no intention of

operating broadband networks. Instead, it partners with ISPs, equipment manufacturers, and

community organizations to provide them with expertise and resources that support their work to

expand their networks to reach the unconnected.

To date, Microsoft’s Airband Initiative has launched commercial partnerships in twenty-

four states and Puerto Rico, with pilot projects in three additional states. Building these

partnerships has given us insight into the real-world challenges rural ISPs encounter when

bringing services to unserved customers in difficult to reach areas. Indeed, Microsoft’s partners

have already extended broadband access to over 633,000 previously unserved people in rural

areas;9 however, they can and should be permitted to reach even further. Microsoft therefore

supports the critically needed changes to the TVWS rules proposed in the NPRM. These

proposals, as reflected in Microsoft’s 2019 Petition for Rulemaking (“Petition”), are based on the

feedback we received from partner ISPs and will support increased investment and connectivity

8 Shelley McKinley, Microsoft Airband: An annual update on connecting rural America,

Microsoft (Mar. 5, 2020), https://blogs.microsoft.com/on-the-issues/2020/03/05/update-connecting-rural-america/.

9 Id.

6

in rural services, while ensuring that incumbent licensees remain protected from harmful

interference.10

Service providers that already use White Spaces technology have explained that certain

practical changes to the TVWS rules will “allow [them] to build on that work and connect even

more of rural America.”11 For example, Evolve Cellular and Skylark Wireless agree that

updating the TVWS rules to better reflect deployment realities will “encourage innovation in

new radio technologies and vastly improve rural broadband performance while maintaining

stringent requirements of non-interference.”12 Declaration Networks Group explains that the

U.S. “can make great strides in closing the digital gap if the FCC moves forward with a

Rulemaking that helps spur increased private-sector deployment of broadband services using

White Spaces technology.”13 And as Rise Broadband has commented, “rule changes to increase

the utility and reach of TV white space spectrum,” will boost the equipment ecosystem, “making

equipment more competitive and affordable, and making the business case a reality for service

providers.”14

Companies that develop and manufacture the TVWS equipment and software likewise

support practical changes to the rules. Adaptrum, a TVWS radio manufacturer, has explained

that practical changes to the TVWS rules such as those in Microsoft’s Petition and proposed in

10 See Petition for Rulemaking of Microsoft Corporation, ET Docket No. 14-165 (filed May 3,

2019) (“Microsoft Petition”); TVWS NPRM ¶ 6. 11 Rural Partners Comments at 2. 12 Evolve/Skylark Comments at 5. 13 Letter from Bob Nichols, CEO, Declaration Networks Group, Inc., to Chairman Ajit Pai,

Commissioner Brendan Carr, Commissioner Michael O’Rielly, Commissioner Jessica Rosenworcel, and Commissioner Geoffrey Starks, FCC, ET Docket No. 14-165, WC Docket Nos. 11-10, 19-195, at 1–2 (filed Dec. 23, 2019).

14 Rise Broadband Comments at 1.

7

the NPRM “will provide enhanced opportunities for providing service with TVWS devices and

thereby have a direct impact in both improving the quality of life of rural Americans and

strengthening the rural economy.”15 As RADWIN, another TVWS technology provider, has

noted, targeted changes to the TVWS rules “will make the available white space spectrum an

even better tool for serving rural and underserved areas.”16

Trade groups in the agricultural sector also recognize that improving the operation and

coverage of WSDs will benefit the country. The American Farm Bureau Federation supports the

Commission’s proposal to improve the TVWS rules and explains that “[f]armers and ranchers

depend on broadband for the viability of their operations,” and that “the latest yield maximizing

farming techniques require wireless broadband connections for data collection and analysis

performed both on the farm and in remote data centers.”17 The National Grange has commended

the Commission for moving forward to “clear regulatory barriers to TV white space technology

as a key tool to help eliminate the digital divide,” and “encourage[s] the Commission to finalize

and enact these new rules in 2020.”18

Technology associations and standards groups agree. ACT, The App Association, further

explains that “Commission action will address the growing need for broadband connectivity in

rural areas via TV white spaces,” and that the additional connectivity enabled by TVWS will

15 Adaptrum Comments at 2. 16 RADWIN Comments at 3. 17 Letter from Paul Schlegel, Vice President of Public Affairs, American Farm Bureau

Federation, to Marlene H. Dortch, Secretary, FCC, ET Docket No. 20-36, at 1 (filed Feb. 25, 2020).

18 National Grange (@NationalGrange), Twitter (Feb. 28, 2020, 1:53 PM), https://twitter.com/NationalGrange/status/1233465186083909633?s=20.

8

allow “small businesses to grow and contribute to local economies.”19 Wi-Fi Alliance explains

that, given the many changes in the WSD landscape since the last time the Commission reviewed

the technical rules, including the success of geo-location and database technologies, the

feasibility and value of narrowband WSDs, and the continued need for innovative tools to close

the digital divide, the Commission should “reevaluate a number of the rules governing white

space devices.”20 Similarly, the Wireless Internet Service Providers Association states that

changes to the TVWS rules “offer real promise that deployment of fixed wireless networks on

TV white space spectrum can develop into a prominent means of delivering broadband services

to rural Americans.”21

Microsoft is particularly appreciative of the efforts of the National Association of

Broadcasters (“NAB”) to come to consensus on the four proposed changes in the TVWS NPRM.

That consensus came after careful collaboration to ensure that changes to the FCC’s rules that

will expand TVWS operations will do so without harming existing broadcasting operations. As

discussed in the NAB Comments on Microsoft’s Petition, it supports the FCC’s inclusion of the

four proposed rule changes in this proceeding.22 While there is not yet agreement on how to

allow rural ISPs to operate at higher power on channels closer in frequency to broadcast

channels, we hope that the record in this proceeding and continued discussions with broadcasters

will contribute to a solution.

19 App Association Comments at 1–2. 20 Wi-Fi Alliance Comments at 2. 21 WISPA Comments at 2. 22 Comments of the National Association of Broadcasters at 2–5, ET Docket No. 14-165 (filed

June 10, 2019) (“NAB Comments”).

9

Public interest groups and membership organizations that seek to facilitate increased

access to technology and connectivity likewise support TVWS rule changes that will enable

greater coverage. Connect Americans Now explains that changing the TVWS rules will “both

better enable ISPs to utilize TVWS technology to bring broadband to rural areas and open up the

technology to a variety of new user groups and use cases.”23 And a group of organizations that

includes Tribal Digital Village, the Gigabit Libraries Network, and Public Knowledge have

explained that “pragmatic and long-overdue changes to the TV White Space rules in Part 15 . . .

present the Commission with an opportunity to take important steps to bridge the rural-urban

digital divide,” and “can empower providers to extend higher-speed internet access to more

unserved areas.”24 Voices for Innovation explains that the Commission can improve broadband

connectivity by enabling connected school buses, agricultural equipment, and new IoT for

precision agriculture, and also encourages the Commission to “pursue more rigorous broadband

use of the TV band spectrum by allowing the use of channels adjacent to broadcaster channels

which otherwise will lay fallow.”25 The Schools, Health & Libraries Broadband Coalition

(“SHLB”) encourages the Commission to act quickly to adopt “new technical rules to promote

TV White Spaces use and availability,” noting that “[a]llowing greater power levels . . . could

23 CAN Comments at 2. 24 Public Interest Comments at 1–2 (internal quotation marks omitted). 25 Letter from Voices for Innovation to Chairman Ajit Pai, Commissioner Brendan Carr,

Commissioner Michael O’Rielly, Commissioner Jessica Rosenworcel, and Commissioner Geoffrey Starks, FCC, ET Docket Nos. 14-165, 20-36, et al., at 1 (filed Feb. 21, 2020).

10

encourage more schools, libraries and other anchor institutions to deploy antennas extending

wireless broadband service to their communities.”26

Support from members of Congress demonstrates the increasing importance of closing

the digital divide in their home communities, and the ability of TVWS technology to do so.

Senator Roger Wicker, Chairman of the Commerce, Science, and Transportation Committee

supports the FCC’s efforts to increase access to TVWS spectrum and highlighted the

Committee’s statement explaining that “[e]xpanding access to broadband is essential for rural

communities,” and that the FCC’s action to “support greater unlicensed spectrum uses for

broadband deployments is a positive step that can increase the pace of progress to bridge the

digital divide.”27 Senators Kevin Cramer, from North Dakota, and Angus King, from Maine,

wrote to the FCC to support its action on TVWS, noting that they were “especially pleased to see

that the NPRM addresses the use of TVWS for narrowband applications, such as precision

agriculture, as well as moving platforms, such as school buses,” and encouraged the FCC to

adopt final rules by the end of 2020.28 Likewise, Senator Mark Warner recently emphasized the

importance of TVWS technology in helping to close the digital divide in his state, Virginia,

explaining that the proposed rules “will support greater utilization of TVWS technology and help

26 Letter from John Windhausen, Jr., Executive Director, Schools, Health & Libraries

Broadband (SHLB) Coalition to Chairman Ajit Pai, Commissioner Brendan Carr, Commissioner Michael O’Rielly, Commissioner Jessica Rosenworcel, and Commissioner Geoffrey Starks, FCC, ET Docket No. 14-165, et al., at 7 (filed Mar. 17, 2020).

27 Senate Commerce (@SenateCommerce), Twitter (Mar. 4, 2020, 3:29 PM), https://twitter.com/SenateCommerce/status/1235301356015099906?s=20.

28 Letter from Senators Kevin Cramer (R-ND) and Angus King (I-ME) to Chairman Ajit Pai, FCC, at 1 (Mar. 10, 2020).

11

to bring affordable, reliable broadband to millions of Americans stuck behind the digital divide,”

and that “[p]rogress on this front has never been more important.”29

Finally, because the proposed rule changes will increase the coverage and cost-

effectiveness of TVWS networks and enable providers to connect more Americans, there is

already strong unanimous, bipartisan support at the FCC for practical changes to the TVWS

rules. Chairman Pai’s statement to the U.S. House of Representatives in a recent hearing

highlighted that the FCC’s work to advance TVWS technology was “aimed at closing the digital

divide,” and that the NPRM’s proposed rules “would provide for more robust service and

efficient use of White Space devices particularly in rural areas, without increasing the risk of

harmful interference to protected services in the TV bands.”30 Commissioner O’Rielly has also

noted of TVWS that “[i]t is common sense to make these frequencies available for additional

non-interfering wireless services, including those that bring broadband access to unserved

households.”31 And Commissioner Rosenworcel has explained that engineering and policy

decisions at the FCC starting in 2008 have allowed White Spaces technology to achieve progress

and promise as a means to close the digital divide, and that the changes proposed in the NPRM

“can open up new possibilities for using these airwaves to power the Internet of Things and

extend the reach of broadband networks.”32

29 Letter from Senator Mark Warner (D-VA) to Chairman Ajit Pai, FCC, and Commissioners

Rosenworcel, O’Rielly, Carr, and Starks, FCC, at 2 (Apr. 14, 2020). 30 Statement of Chairman Ajit Pai, FCC, Hearing on the FCC’s Fiscal Year 2021 Budget

Request, Before the Subcommittee on Financial Services and General Government Committee on Appropriations, U.S. House of Representatives, at 2 (Mar. 11, 2020), https://docs.fcc.gov/public/attachments/DOC-363663A1.pdf.

31 TVWS NPRM at Statement of Commissioner Michael O’Rielly. 32 Id. at Statement of Commissioner Jessica Rosenworcel.

12

II. THE COMMISSION SHOULD ADOPT ITS PROPOSALS TO OPTIMIZE THE TVWS RULES.

The NPRM proposes a set of practical and focused changes to the TVWS rules that will

enable broadband expansion without increasing the risk of harmful interference to licensed

services. The Commission should, as it proposes, (1) increase power limits and HAAT limits for

fixed devices subject to appropriate separation distances and coordination requirements; (2)

allow WSDs to operate on mobile platforms within pre-cleared geofenced areas at the higher

power levels permitted for fixed devices to enable connectivity in additional use cases; and (3)

adopt its proposed rules for narrowband WSDs to enable important new use cases in rural and

industrial settings while maintaining the protections for other users of the TV bands.

Additionally, the Commission should (4) adopt rule changes that allow fixed WSDs to operate at

higher power levels on first-adjacent channels to broadcasting operations governed by technical

rules that maintain conservative interference protections, and should implement a modern,

terrain-based propagation model in order to accurately reflect the actual operating environment.

A. The Commission’s Proposals to Allow Fixed White Space Devices to Operate at Higher Power Levels and Heights Above Average Terrain Will Expand Broadband Coverage.

The proposed increases to fixed device power in less-congested areas and HAAT will, as

the Commission notes, “benefit American consumers in rural and underserved areas.”33 The

Commission should adopt both of these proposals. The separation distance changes proposed in

the NPRM are conservative and should assure incumbent licensees that no increased risk of

harmful interference will result from these rule changes. To ensure that new technical rules allow

service providers to fully realize the benefits of increased power limits for fixed devices in less

congested areas, the Commission should implement this change in a way that allows effective

33 See TVWS NPRM ¶ 8.

13

use of multiple-input-multiple-output (“MIMO”) devices. Additionally, in implementing this

proposed HAAT improvement, the Commission should apply the HAAT limit change to all

areas—not just less-congested areas. This would avoid unnecessarily foreclosing deployments

where an increased HAAT can fill coverage gaps but would not interfere with existing services.

i. The Commission should adopt its proposal to increase the radiated power limits for fixed devices in less-congested areas.

The Commission’s proposal to increase the maximum permissible radiated power from

10 to 16 watts EIRP for fixed WSDs in less-congested areas will, as the Commission recognizes,

allow service providers to expand their service areas to expand and improve broadband coverage

for hard-to-reach areas.34 Implementing this change by allowing an increase in antenna gain

rather than in conducted power will, as the Commission explains, “improve spectrum efficiency

by ensuring that less white space device energy is directed outside the main antenna beam than

would be the case if [the rules] permitted higher transmitter power using lower gain, less

directional antennas.”35 As Microsoft noted in its Petition, this change will allow significant

improvement in the economies of rural coverage.36

Service providers and TVWS equipment makers have confirmed this benefit. Adaptrum,

for example, has explained that, “a modest increase in effective radiated power . . . would

improve the ability of TVWS devices to provide service to users in rural areas, thereby

improving the economics of deploying TVWS in these currently underserved locations.”37

Overall, “[i]ncreasing radiated power by allowing greater directional gain will directly improve

34 See id. ¶¶ 9, 12, App. A (Proposed Rules) at § 15.709(a)(2). 35 Id. ¶ 13. 36 See Microsoft Petition at 5. 37 Adaptrum Comments at 2.

14

the cost-to-coverage ratio for providers and allow them to serve more Americans by enabling

more homes to be served from a single tower.”38

The Commission can adopt this rule change while continuing robust protection of

licensed services using the band. The proposal maintains protection for broadcasting services

because the increased power levels are accompanied by increased separation distances from

broadcasters.39 As the Commission notes, NAB does not oppose this proposal, subject to

appropriate separation distances established to protect broadcasters.40 The proposed change

would protect other incumbent services as well, including wireless medical telemetry service

(“WMTS”) operations because the proposed changes do not alter the protections for the WMTS

operations in Channel 37, and the new EIRP limits would only apply below Channel 37.41

Further, the NPRM does not propose changes to the existing radiated power limits applicable to

Channel 37, which specify that fixed WSDs may only operate up to 4 W (36 dBm) EIRP

between 602 and 620 MHz.42 The proposed rules would ensure that the TVWS Database

continues to protect wireless microphone users as it does today without change.43

Because the proposed rules will effectively protect against harmful interference, and the

economic benefits of connecting additional unserved households to broadband access are great,

Microsoft strongly encourages the Commission to adopt the proposed increased power limit. The

Commission additionally asks whether a different maximum EIRP from that proposed would be

38 Rural Partners Comments at 3. 39 See TVWS NPRM at App. A, tbl. 3 to § 15.712(a)(2)(v) & tbl. to § 15.712(a)(2)(v). 40 Id. ¶ 11; NAB Comments at 1–2. 41 See TVWS NPRM ¶ 13. 42 See 47 C.F.R. § 15.709(a)(2)(i). 43 See 47 C.F.R. § 15.713.

15

more appropriate to enable service to rural areas and whether it should allow even higher power

levels under certain circumstances.44 Microsoft would welcome the flexibility for ISPs to operate

at radiated power levels greater than 42 dBm, and we encourage the Commission to explore this

possibility, but support adopting the 16 Watt EIRP power level now rather than waiting for an

inquiry into higher power levels.

Finally, and importantly, the Commission should adopt rules that treat each antenna

element separately for purposes of power level measurement for WSDs with multiple streams

using MIMO. Doing so will enable broadband providers to more effectively deliver broadband

connectivity using MIMO technologies and would not change the interference analysis.

ii. The Commission should adopt its proposed increase in HAAT limits in any areas where it would not cause harmful inference to licensees.

Microsoft supports the Commission’s proposal to increase the maximum permissible

HAAT for fixed WSDs to 500 meters because it will increase coverage areas and transmitter site

availability, particularly in remote or mountainous areas.45 This tool will be valuable for network

planners seeking to serve populations in challenging geographies, particularly in mountain

foothills and valleys, where there are few if any tower sites that comply with the existing HAAT

limit. Often the only available location is a natural feature such as a ridge, but, because these

natural features are often well above the average elevation of the surrounding terrain, it can be

impossible to deploy fixed WSDs on them below the current 250-meter HAAT limit.

44 TVWS NPRM ¶ 14. 45 See id. ¶¶ 17, 18 (asking about the benefits of a higher HAAT limit in terms of improved

rural coverage and increased transmitter site availability in high elevation areas).

16

Rural technology and service providers have noted this concern and the utility of an

increased HAAT limit.46 Specifically, Adaptrum has noted that the current HAAT limit prevents

operators from providing service in important mountainous rural communities: “For instance, a

community in Southern Virginia was exploring use of TVWS systems, but many of the sites

targeted were not viable due to the HAAT limit.”47 And Sacred Wind has explained that it

already has two existing communications towers it could use for White Spaces operations, but

for the current HAAT limit, noting that it also “has surveyed other areas for new tower

installations that are above 250 meters HAAT, that could otherwise accommodate TVWS

antennae in order to serve many homes beyond reach of other spectrum.”48 The Commission’s

proposed rule changes would directly address these missed opportunities to connect additional

homes, schools, and communities, as well as provide flexibility in other areas where operators

face HAAT limitations.

The Commission need not limit a HAAT change to less congested areas.49 Such a

limitation is unnecessary to protect incumbent users because WSD use under the revised rules

would still have to comply with channel and location availability as indicated by the White

Spaces Database (“WSDB”), which “will continue to ensure that incumbent operations will be

protected from harmful interference.”50 There is therefore no benefit in precluding higher HAAT

46 See Comments of 6Harmonics at 4, ET Docket No. 14-165 (filed June 10, 2019)

(“6Harmonics Comments”); Adaptrum Comments at 3; CAN Comments at 3; DSA Comments at 6; NAB Comments at 3; Public Interest Comments at 7–8; Rural Partners Comments at 5–6; Wi-Fi Alliance Comments at 4; WISPA Comments at 3–4.

47 Adaptrum Comments at 3. 48 Sacred Wind Comments at 6–7. 49 See TVWS NPRM ¶ 18 (asking whether to limit HAAT over 250 meters to less-congested

areas). 50 Wi-Fi Alliance Comments at 4.

17

deployments in any area where it would otherwise be permitted by the separation distance

requirements and the database availability parameters. In fact, from an engineering perspective,

the TVWS rules do not need a HAAT limit at all. This is because by adjusting the minimum

separation distances the Commission could fully protect incumbents from TVWS deployments at

any HAAT value.51 However, in Microsoft’s experience, increasing the HAAT limit to 500

meters will cover the deployment cases most commonly precluded by current HAAT limits,

which are typically near or in the foothills close to mountains.

The Commission also asks whether the increased fixed WSD transmission range

associated with higher HAATs will limit opportunities for spectrum sharing between different

unlicensed WSDs.52 In our experience, such issues are unlikely to be significant in practice.

Generally, there is only one ISP offering broadband service in the areas where a company could

use the increased HAAT limits because only rural areas with low population densities will allow

for the large separation distances needed for higher HAAT values under the Commission’s

proposed rules. Because there is likely to be only one ISP with fixed White Space operations in

such an area, any TVWS-to-TVWS interference issues would be identified and avoided by the

ISPs network design and channel planning process. Further, to the extent that there are multiple

TVWS operators even in such rural area, because effective coverage area is a function of both

EIRP and HAAT and the correct solution is area specific, the net effect of increased HAAT and

the increased directional gain the Commission has proposed to authorize may actually improve

net frequency reuse by encouraging the deployment of narrower coverage cones.53 Additionally,

51 See TVWS NPRM ¶ 18 (asking whether a HAAT limit greater than 500 meters would be more

appropriate). 52 Id. ¶ 18. 53 See id. ¶ 13.

18

the low duty cycles typically used by WSDs, combined with the “burst” nature of the

transmissions—short periods of transmission activity followed by longer periods of inactivity—

mean that WSDs operating on the same spectrum will be able to effectively share and avoid

causing harmful interference to one another.

Microsoft also supports the streamlined coordination procedure for higher HAAT

facilities as proposed in the NPRM, including the Commission’s proposal not to require a 30-day

trial period. To implement this coordination, Microsoft agrees that potentially affected protected

entities should be defined as stations with a “broadcast contour . . . within the separation distance

corresponding to an assumed HAAT 50 meters higher than the actual deployment,” and should

encompass any licensees with spectrum assignments operating on the WSD’s proposed channel

of operation.54 The Commission should permit White Space operators to contact potentially

affected licensees via email, telephone, or any other form of electronic communication used by

the licensee. Finally, we agree with the Commission that there is no need to require new

coordination or notification for small HAAT increases within a 50-meter step.55

Relatedly, the Commission is correct to ask questions regarding its current antenna above

ground height (“AGL”) limit.56 Because the current AGL limit could unnecessarily limit

operators’ ability to design and deploy White Spaces networks, Microsoft supports exploring

different approaches. Because, as the Commission notes, separation distances are based on

HAAT rather than the antenna height above ground, increasing or eliminating the antenna AGL

should not have any effect on the potential for interference to a protected service. Notably, the

54 Id. ¶¶ 20, 21 (asking about coordination details). 55 See id. ¶ 22. 56 See id. ¶¶ 24–26.

19

Commission’s decision to increase the limit from 30 to 100 meters last year has significantly

improved the ability of operators to build and deploy networks without increasing the risk of

harmful interference.57 If the Commission does revise the antenna above ground limit, any

revision should be applied in all areas rather than only in less-congested areas, which would

unnecessarily increase the complexity of the WSDB.

Microsoft notes, however, that if the Commission chooses to promote greater flexibility

by relying exclusively on HAAT limits, the Commission should account for the fact that some

locations will have negative HAAT values due to the terrain-averaging function of the

calculation. A fixed WSD operating towards the center of a valley, for example, could have a

negative HAAT.

B. The Commission Should Adopt its Proposal to Allow TVWS Operation on Mobile Platforms Within Geofenced Areas.

Allowing WSDs to operate at higher powers on mobile platforms within a pre-cleared

geofenced area will provide significant benefits for rural communities. Rural industries such as

agriculture will benefit from more information-intensive farming practices and remote livestock

monitoring.58 Additionally, WSDs can be placed on the roof of a school bus operating on a rural

route, allowing children to do homework during often long travel times. As Microsoft explained

in its Petition, this rule change “would allow a wide variety of operat[ions] . . . in well-defined

57 See Amendment of Part 15 of the Commission’s Rules for Unlicensed White Space Devices;

Amendment of Part 15 of the Commission’s Rules for Unlicensed Operations in the Television Bands, Repurposed 600 MHz Band, 600 MHz Guard Bands and Duplex Gap, and Channel 37; Expanding the Economic and Innovation Opportunities of Spectrum Through Incentive Auctions, Report and Order and Order on Reconsideration, 34 FCC Rcd. 1827, ¶ 64 (2019) (“White Spaces Order on Reconsideration”) (increasing the allowable fixed white space device antenna height above ground level from 30 meters to 100 meters in less congested areas).

58 See TVWS NPRM ¶ 39.

20

areas, such as those in the agricultural and extractive industries, the necessary flexibility to use

long-range White Space links,” and provide an additional opportunity for internet connectivity in

rural areas.59 A wide range of rural service providers, equipment makers, and trade groups

support this category of TVWS operation because of its potential for enabling useful applications

in otherwise unused spectrum.60

As a group of rural ISP partners correctly explains, “[p]ermitting fixed TVWS device

operations on moveable platforms using geofencing technology will allow residents, students,

and workers in rural areas to access the internet in communities where they otherwise might not

always have reliable access.”61 Rules allowing for these uses could enable more deployments

similar to the Hillman, Michigan school bus connectivity project conducted by Microsoft and its

local partner, Allband Communications, under an experimental license.62 Operating WSDs on

the roof of the bus at the higher EIRP limits associated with fixed WSDs, within a well-defined,

pre-cleared area along the school bus route, allowed the bus to connect to TVWS base stations

over longer distances and establish a backhaul link for a Wi-Fi network on the bus. Facilitating

high-speed internet access for students to complete homework and access online resources

provides connectivity during what can be long journeys to and from school in rural areas, and

represents a valuable opportunity for students without a home internet connection to obtain

connectivity. These same TVWS and Wi-Fi equipped school buses can be used to provide

59 See Microsoft Petition at 23, 25. 60 See 6Harmonics Comments at 6; CAN Comments at 3; DSA Comments at 8–9; NAB

Comments at 3–4; Nominet Comments at 6; Public Interest Comments at 8–9; RADWIN Comments at 2; Wi-Fi Alliance Comments at 6.

61 Rural Partners Comments at 6. 62 See Microsoft Petition at 22–23; Application of Microsoft Corporation, Form 442 Exhibit 1:

Experiment Description, ELS File No. 0049-EX-CM-2018, Call Sign WJ2XCD (filed Mar. 7, 2018).

21

connectivity after hours and on weekends to schoolchildren living in communities with

households lacking broadband connections.

Microsoft’s Petition proposed allowing fixed WSDs to operate on moveable platforms

within geofenced areas.63 The Commission proposed, instead, to allow mobile Mode II

personal/portable WSDs to operate at higher power levels commensurate with that allowed for

fixed devices within pre-cleared geofenced areas. The NPRM’s proposal will, if governed by the

correct technical rules, unlock the same benefits for users of WSDs as our earlier proposal.64

Although either approach could be successful, the use of the Mode II device framework

to enable higher power geofenced operations may require more complex and extensive rule

changes than if the Commission use the existing fixed-device rules that Microsoft suggested.65

As the Commission acknowledges, its choice of approach will affect the equipment approval

process.66 Under the approach of allowing fixed devices to operate on mobile platforms within

pre-defined areas, equipment makers could certify devices under the existing fixed device

certification process, updated with an option to test the re-check and shutoff requirements

associated with geofenced operation. Under the approach of allowing Mode II portable devices

to operate at higher powers, however, fixed WSDs also intended for geofenced mobile

63 See Microsoft Petition at 22, 24. 64 See TVWS NPRM ¶ 39 (seeking comment on “the benefits or costs of this proposal with

respect to white space device users or other authorized users of the TV band spectrum”). 65 See id. ¶ 41 (asking whether the FCC should “instead permit devices operating under the

fixed device rules to operate on mobile platforms as suggested by Microsoft and others”); see also NAB Comments at 4 (explaining that, subject to the protections described in Microsoft’s Petition, “NAB believes the Commission should move forward with an FNPRM including [Microsoft’s] proposal.”)

66 See id.

22

operations would require certification under two standards: one for the fixed operations, and then

another for Mode II portable device operations at increased powers where subject to geofencing.

The Commission could, as it suggests, approve Mode II devices at a higher power level

for general use because the database would limit the amount of power permitted for operations in

any specific area.67 However, this approach would increase the complexity of the WSDB

compared to allowing fixed devices to operate on mobile platforms subject to the re-check and

shutoff requirements. Additionally, the antenna gain rules for fixed devices already

accommodate the type of operation necessary to create a link over longer distances.68

Personal/portable devices, however, must operate with an omni-directional antenna with 0 dBi

gain, meaning that if the maximum conducted power is 1 W, then the maximum power of the

TVWS radio within the geofenced area will be 1 W EIRP. 69

If the Commission adopts the approach proposed in the NPRM of allowing Mode II

personal/portable devices to operate at “higher power levels commensurate with that allowed for

fixed devices within ‘less congested’ areas,” in pre-cleared geofenced areas, the Commission

should adopt two important changes.70 First, it should ensure that the rules clearly address the

permitted power levels and antenna gain. The proposed rules, as written, do not include language

specifically addressing the power levels that would be allowed for geofenced higher power Mode

II operations, nor do they address the antenna characteristics that would be required to achieve

67 See id. 68 See 47 C.F.R. § 15.709(c)(1) and (2) (addressing antenna gain for fixed devices). 69 The Commission has clarified that the 40 mW (16 dBm) fixed WSD will need to operate

with an antenna with a directional gain of 6 dBi given the conducted power limit of 10 mW. White Spaces Order on Reconsideration, ¶ 44 n.114 (citing 47 C.F.R. § 15.709(b)(1)(ii)). Personal/portable devices, however, may operate at up to 40 mW EIRP with a non-directional antenna (0 dBi gain). Id. ¶ 45 (citing 47 C.F.R. § 15.709(b)(2)).

70 See TVWS NPRM ¶ 39.

23

these power levels. Second, the Commission should revise the current power level rules for

Mode II in Section 15.709(a)(2)(ii) of the Commission’s rules to specify that devices operating

using geofencing would be allowed to transmit at the same power levels as fixed devices in less

congested areas.71 These changes will ensure that geofenced devices are able to operate at the

power levels necessary to close the links required for the applications the Commission intends to

enable.

Rather than allowing geofenced mobile operations under the Mode II personal/portable

rules as the Commission proposes, or under the fixed device rules as Microsoft proposed, the

Commission could, as it suggests, create a new class of WSDs—"mobile white space devices.”

This approach would distinguish such devices from personal/portable WSDs, permit detachable

antennas, and simplify the inclusion of geofenced operations in the regulations.72 Using the

Mode II rules as the Commission proposes would require adjusting or making exceptions to the

antenna gain and power levels, as explained above. Adding a new “mobile white space device”

class to the rules, by contrast, would be simple and would not require as many modifications to

the existing device class rules. The new rule could be straightforward and reference the fixed

device rules for power levels, antennas, and RF exposure, and include the new geolocation rule

the Commission has proposed in Section 15.711(c)(3) of the rules.73

71 Personal/Portable devices: Up to 100 mW (20 dBm) EIRP and, when operating in less

congested areas as defined in § 15.703, subject to the geofencing procedures in § 15.711(c)(3), and consistent with the separation distances in Table 3 to § 15.712(a)(2)(v), at the power limits for fixed devices specified in § 15.709(a)(2)(i). (Recommended additions in italics.)

72 See TVWS NPRM ¶¶ 39, 41 (asking whether to create a new class of devices). 73 See id. at App. A, § 15.711(c)(3).

24

The White Spaces Database is well equipped to accommodate this new category of

geofenced operations, by whatever means the Commission decides to implement it. There is no

reason why the Commission should dictate the specifics of how information describing pre-

cleared geofenced areas should be provided to the database. But, if it does, the Commission

should allow for both flexibility and specificity in the geofenced area and not foreclose the use of

any type of area shape.

More generally, the Commission also asks for comment on any other operational

specifics, including antenna requirements and database inputs, that would make the geofenced-

operation rule revisions a success. To ensure that the rules are as useful as possible for the kind

of applications the Commission envisions, we recommend two other improvements. First, the

Commission should state that the same antenna requirements that apply to fixed WSDs will

apply to higher power mobile devices, including the provisions regarding detachable antennas

with higher gain.74 Second, as the Commission notes, technologies including electronically

steerable beams could enable mobile devices to operate with higher gain and therefore more

highly directional antennas. The rules should therefore allow mobile geofenced devices to

incorporate such technologies but should not mandate their use, as such a requirement could

unnecessarily increase the cost of producing and procuring devices for smaller manufacturers

and rural operators.

Finally, it is important to note that the NPRM’s proposed interference protection

safeguards for geofenced operations will provide conservative protection for all licensed TV

band users, including broadcasters. Commenters responding to the proposal in Microsoft’s

74 See id. ¶ 41 (asking whether the rules should allow higher power mobile devices to use

detachable, higher gain antennas, as permitted for fixed devices).

25

Petition for movable geofenced operations explained that, “[w]ith a 60 second channel

availability check any additional risk of interference is unlikely,”75 and that with the correct

computation of available channels and power limits, “this approach has no effect on the risk of

interference compared to a fixed device.”76 The proposed 60-second location re-check interval

and 1.6-kilometer shutoff distance would effectively prevent operation outside the geofence at

interstate highway speeds—a vehicle traveling 60 miles per hour travels roughly 1.6 kilometers

in 60 seconds.77 Further, the requirement that the fixed WSD continue to contact the database to

confirm that its pre-determined channel of operation remains available will account for the

possibility of changed channel availability due to wireless microphone registrations.78 Similarly,

because of the re-check and shutoff requirements, there is no need to place any limitations on the

size of the area within which higher power mobile devices can operate.

C. The Commission Should Adopt Its Proposal to Clarify the Framework for Narrowband TVWS IoT Operations.

Microsoft also supports the Commission’s proposed rules for creating a new category of

WSDs for narrowband operations. As a variety of stakeholders explained in response to the

narrowband proposal in Microsoft’s Petition, this class of WSDs will support a range of rural

industries that need to make periodic data transmissions over long distances to support IoT

operations. Narrowband WSDs will take advantage of the propagation characteristics of low-

band spectrum to “enable new, innovative uses of TVWS spectrum in the agriculture, mining,

75 6Harmonics Comments at 6. 76 Nominet Comments at 6. 77 TVWS NPRM ¶ 40. 78 Id.

26

and environmental monitoring sectors.”79 Narrowband White Spaces technology will be

especially useful for rural industries such as agriculture and mining, where IoT applications

geared for denser urban settings with far different telecommunications and electric power

environments are less effective.

White Spaces devices in the VHF and UHF bands will substantially improve IoT

applications because they will facilitate the periodic transmission of small amounts of data over

longer distances and more reliably overcome obstacles such as foliage and small buildings than

any devices in any other unlicensed band. By addressing the regulatory barriers limiting

commercial use—the power spectral density and conducted adjacent channel emission limits in

the current rules—the proposed rule changes would allow narrowband WSDs to operate at the

power levels necessary to transmit data over longer distances in very small channels. The

conducted power limit of 12.6 dBm/100 kHz proposed by the Commission for narrowband

devices is appropriate. It is equal to the conducted power spectral density limit of a WSD

operating in a full 6-MHz channel—12.6 dBm within any 100 kHz. Furthermore, excluding the

top and bottom 250 kHz of the 6-MHz channel means that if each of the fifty-five 100-kHz

channels had a narrowband IoT device operating at the maximum proposed power of 12.6 dBm,

the total conducted power in the 6 MHz channel would be 30 dBm—the same conducted power

limit for a broadband WSD operating in a 6 MHz channel. Similarly, the radiated power limit of

narrowband WSDs would be 18.6 dBm, 36 dBm total in a 6-MHz channel—the same as a fixed

WSD operating in an area that is not considered less-congested. The Commission’s proposal to

protect incumbents treats a channel containing a single narrowband WSD operating at 18.2 dBm

EIRP as a fixed WSD operating at 36 dBm EIRP.

79 Rural Partners Comments at 7.

27

Importantly, the proposed narrowband WSD framework is conservative, and will

adequately protect all users of the TV bands. The proposed limit of ten seconds per hour of total

operation per transmitter is extremely strict and will, as the Commission explains, prevent

narrowband devices from being used for data-intensive applications.80 Microsoft’s experimental

licenses for narrow channel WSDs demonstrated typical applications. These included a

transmitter mounted in the ground as part of a soil moisture sensor, integrated with a water gauge

for an irrigation system to measure daily and weekly flows, and incorporated into sensors

operating in a series of greenhouses to monitor hourly temperatures. Microsoft’s expectation is

that the Commission’s adoption of rules allowing narrow channel WSDs will unlock new

applications across many vertical markets that can benefit from the unique characteristics of the

VHF and UHF bands, only limited by the imagination of the end user.

Indeed, in response to Microsoft’s Petition including these operating parameters, NAB

supported the inclusion of narrowband rules in a Notice of Proposed Rulemaking.81 And finally,

the proposed rules would also prohibit any co-channel operation between narrowband IoT and

WMTS by specifying that narrowband WSDs may only operate on frequencies below 608

MHz.82

Finally, the Commission asks whether, if it requires narrowband devices to operate as

clients of fixed devices that contact the WSDB it should increase the minimum separation

distances from co-channel and adjacent channel TV station contours to those for

personal/portable devices operating as clients.83 If the Commission does so, it should only

80 TVWS NPRM ¶ 46. 81 See NAB Comments at 5. 82 TVWS NPRM at App. A, § 15.707(c). 83 See id.

28

require this increase for clients that do not have incorporated geo-location, analogous to a

personal/portable type I device.

D. The Commission Should Permit TVWS Operations at Increased Power Levels on First-Adjacent Channels While Maintaining Conservative Protections for Broadcasters.

Finally, the Commission should enable network operators to use fixed WSDs at higher

power levels on first-adjacent channels to broadcasters and on channels with a 3-MHz offset

from broadcaster channels. The Commission can do so without causing harmful interference to

licensees by allowing the WSDB to use a more accurate, terrain-based propagation model, and

by correcting outdated assumptions regarding broadcast receiver selectivity and desired-to-

undesired signal (“D/U”) ratios. To demonstrate the efficacy of this approach, Microsoft

conducted testing in both a lab environment and in the field to determine the circumstances

under which fixed WSDs on channels adjacent to broadcast operations could operate at higher

power levels without harmful interference. The lab testing confirms that the D/U ratio of ATSC

1.0 TV receivers with integral displays is more robust than the value assumed by the FCC,

consistent with several prior studies by the FCC and third parties, and that the FCC can authorize

higher power TVWS operation on adjacent channels in certain circumstances with no

impairment to broadcast receivers. ATSC 3.0 plug-in receivers offer the promise of even greater

adjacent channel selectivity.

In response to Microsoft’s Petition, parties encouraged the FCC to update the propagation

model used to determine whether and where TVWS operations are permitted on the first-

adjacent channel. 6Harmonics, for example, asked the Commission to allow TVWS providers

the option to use the Longley-Rice model, noting that the FCC has accepted the Longley-Rice

model “as the most accurate and appropriate methodology to determine propagation” in the 600

29

MHz bands.84 The existing F-curve models, as Adaptrum explains, do “not consider local terrain

factors that can limit the reception of broadcast signals,” and using a more accurate, updated

model such as Longley-Rice would “account for local signal availability factors to improve

predictions.”85 The Dynamic Spectrum Alliance likewise encourages the FCC to allow WSDBs

to use terrain-based models such as Longley-Rice or ITU-R. P-1812 to calculate channel

availability, noting that radio propagation modeling is “rapidly becoming more granular as very

detailed GIS data on terrain, clutter and other factors enhance the algorithms used by spectrum

databases to enforce compliance with interference protection rules.”86

Subsequent to its adoption of F-curves as the model for signal strength prediction for

TVWS operations, the Commission has endorsed the use of terrain-based propagation models to

calculate interference protection and coexistence between a range of different services. In its

recent Report and Order authorizing unlicensed use in the 6 GHz band, for example, the

Commission directed Automated Frequency Coordination Systems to use either the WINNER-II

model or the Irregular Terrain Model combined with a clutter-specific model depending on the

distance.87 Spectrum Access Systems in the 3.5 GHz band also use the terrain-based Irregular

Terrain Model.88

84 6Harmonics Comments at 2. 85 Adaptrum Comments at 4. 86 DSA Comments at 11–12. 87 See Unlicensed Use of the 6 GHz Band, Expanding Flexible Use in Mid-Band Spectrum

Between 3.7 and 24 GHz, Report and Order and Further Notice of Proposed Rulemaking, FCC No. 20-51, ET Docket No. 18-295, GN Docket No. 17-183, ¶¶ 65–68 (rel. Apr. 24, 2020).

88 See Wireless Innovation Forum, Requirements for Commercial Operation in the U.S. 3550-3700 MHz Citizens Broadband Radio Service Band at 11, Doc. WINNF-TS-0112, V1.9.1 (Mar. 11, 2020), https://winnf.memberclicks.net/assets/CBRS/WINNF-TS-0112.pdf.

30

The Commission has also endorsed the use of updated propagation models in the

broadcast context. In evaluating the potential for interference between the Automated Maritime

Telecommunications System (“AMTS”) and TV stations, the FCC explained that OET used

software that implements the Longley-Rice model and provides integrated mapping.89 Further,

the FCC’s rules allow low-power TV station applicants to use “terrain shielding and Longley-

Rice terrain dependent propagation prediction methods to demonstrate that the proposed facility

would not be likely to cause interference to low power TV, TV translator and TV booster

stations.”90 The Commission’s rules also permit the use of the Longley-Rice model in

determining whether a new DTV station will cause interference in areas served by another post-

transition DTV station.91 In addition, the Commission has recently suggested the Longley-Rice

model for determining whether a television signal reaches a certain percentage of the population

in a community, in connection with updating its methodology for determining whether a

broadcast station is ‘significantly viewed’ in a community outside of its local television market.92

The use of a terrain-based propagation model combined with, as described below, a more

accurate D/U ratio for TV receivers will produce more accurate interference protection

calculations and will allow service providers to expand service to additional areas where doing

so would otherwise be precluded by outdated models and assumptions.

89 See Applications of AVISTA CORPORATION to Modify Licenses for Automated Maritime

Telecommunications System Stations WQKP817, WQKP819, and WQKP820, Order, 27 FCC Rcd. 263, ¶ 6 (2012).

90 47 C.F.R. § 74.707(e). 91 See 47 C.F.R. § 73.616(d)(1). 92 See Significantly Viewed Stations, Modernization of Media Regulation Initiative, Notice of

Proposed Rulemaking, FCC No. 20-41, MB Docket Nos. 20-73, 17-105 ¶ 11 (rel. Mar. 31, 2020).

31

Additionally, several studies conducted over the past decade support the use of a more

accurate, updated D/U ratio for TV receivers than the -33 dB D/U ratio most recently used by the

Commission in its adjacent channel separation distance calculations.93 Even in 2007, a study of

DTV receivers available in 2005 and 2006 indicated that those receivers were typically more

selective with respect to undesired signals on the first adjacent channel signals than indicated by

the -33 dB value, finding that the median adjacent channel D/U ratio for the sampled receivers

was -38 dB when the desired DTV signal was near the minimum threshold level for service.94 A

more recent 2014 study measuring LTE-to-DTV interference likewise suggested that the D/U

ratio was better than -33 dB, and that “[t]he co- and adjacent-channel D/U protection ratios

presently in FCC rules . . . may be conservative.”95 Another LTE-to-DTV field study

consistently measured D/U ratios in the range of -35 to -45 dB.96 Additionally, the 2014 CEA

Study, conducted as part of the Incentive Auction proceeding, also observed that the D/U ratios

93 See Unlicensed Operation in the TV Broadcast Bands; Additional Spectrum for Unlicensed

Devices Below 900 MHz and in the 3 GHz Band, Third Memorandum Opinion and Order, 27 FCC Rcd. 3692, ¶ 17 (2012).

94 See Unlicensed Operation in the TV Broadcast Bands; Additional Spectrum for Unlicensed Devices Below 900 MHz and in the 3 GHz Band, Second Report and Order and Memorandum Opinion and Order, 23 FCC Rcd. 16807, ¶ 177 (2008) (“TVWS 2008 R&O, M&O”) (citing Office of Engineering and Technology, FCC, Interference Rejection Thresholds of Consumer Digital Television Receivers Available in 2005 and 2006, OET Report FCC/OET 07-TR-1003, at 5-9, 15-13 (Mar. 30, 2007), https://transition.fcc.gov/oet/info/documents/reports/DTV_Interference_Rejection_Thresholds-03-30-07.pdf).

95 Office of Engineering and Technology, FCC, Measurements of LTE into DTV Interference: Tests on four ATSC DTV Receivers of OFDM 64 QAM Co- and Adjacent-Channel Interference, Report TA-2014-01, at 7421, 7428-7430 (June 17, 2014), https://docs.fcc.gov/public/attachments/DA-14-852A2_Rcd.pdf.

96 H. Stephen Berger, TEM Consulting, LP, Field Study and Technical Analysis of the Potential for Interference from LTE UE Operating in the 700 MHz A Block to Reception of DTV Channel 51, WPWR-TV, at 9 (June 2015), https://ecfsapi.fcc.gov/file/60001090656.pdf.

32

for error-free reception of DTV signals from the receivers tested were better than the ATSC A/74

recommended standard.97

To investigate the interference rejection capabilities of current TV receivers, Microsoft

contracted with DEKRA Certification and Testing, Inc. (“DEKRA”) to conduct lab

measurements to test the required D/U ratio in both the first-adjacent channel and in a channel

with a 3-MHz offset from the broadcast channel using a representative sample of ATSC 1.0

receivers, including both set-top boxes and stand-alone receivers, covering different resolutions

and price points, as well as available ATSC 3.0 receivers.98 For the lab testing of the ATSC 3.0

receivers, as well as the subsequent field testing, Microsoft also partnered with ARK

Multicasting, Inc. and Edge Spectrum, Inc. (“ESI”). As described in additional detail in

Appendix A, for each model of ATSC 1.0 and ATSC 3.0 receiver, the lab testing first determined

the threshold sensitivity for viewing a test video error free (the mean value was -85.1

+/-1.8 dBm). For the ATSC 3.0 receivers, the tests were conducted at different modulation and

code rates.

For the next lab test, a “desired” signal was generated using an ATSC 1.0 transmitter at

one of four different signal strengths: moderately strong (-43 dBm), moderate (-53 dBm), weak

(-65 dBm) and very weak (-80 dBm). The “undesired” signal was generated by a commercial

97 See Gary Sgrignoli, MSW, A Report to the Consumer Electronics Association Regarding

Laboratory Testing of Recent Consumer DTV Receivers with Respect to DTV & LTE Interference, 38 (2014) (“CEA Report”), as attached to Letter from Julie Kearney, Vice President of Regulatory Affairs, CEA, to Marlene H. Dortch, Secretary, FCC, GN Docket No. 12-26, ET Docket No. 14-14 (filed May 22, 2014), https://ecfsapi.fcc.gov/file/7521150300.pdf (showing D/U values for DTV-into-DTV interference ranging from -39.8 dB to -46.8 dB).

98 See Appendix A, DEKRA Certification and Testing, Inc., TVWS-DTV Coexistence on the First Adjacent Channel, Laboratory Measurements Test Report, at 3 (May 3, 2020) (“DEKRA Lab Results Test Report”).

33

fixed WSD operating on the first adjacent channel. For each ATSC 1.0 receiver at each desired

signal level, the power of the WSD was increased until impairment to the video was observed.

The same test sequence was repeated for the ATSC 3.0 receivers. Then, this entire process was

repeated with the WSD radio frequency tuned so that there was a 3-MHz separation between the

edge of the WSD channel and the edge of the broadcast TV channel.

The observations from the lab testing confirmed that DTV receivers operate with D/U

ratios better than the assumed ratios. The required D/U ratio between the desired DTV signal (the

“D”) and the undesired WSD signal (the “U”) in the first adjacent channel across all ATSC 1.0

DTV receivers (with an integral display) and desired signal levels, was found to range

between -40.3 to -46.7 dB.99 The required D/U ratio in the first adjacent channel for all ATSC

1.0 DTV set-top boxes (without an integral display) tested was found to range between -33.8 and

-42.4 dB.100 For the high-definition ATSC 3.0 DTV receiver (without an integral display), the

required D/U ratio in the first adjacent channel ranged between -43 dB to -64 dB, depending on

the modulation rate, error correction code, and desired signal level.101

On average, when the undesired WSD signal was shifted by 3 MHz from the edge of the

desired broadcast ATSC 1.0 TV channel, there was a 5.7 dB improvement in the required D/U

ratio observed in the first adjacent channel across all ATSC 1.0 DTV receiver models and

99 See DEKRA Lab Results Test Report at 4, 12. By contrast, the recommended D/U ratio for

DTV-into-DTV interference under the ATSC A/74 standard on the first adjacent channel for moderately strong and weak desired signals is -33 dB. See ATSC Recommended Practice: Receiver Performance Guidelines, Document A/74:2010 at 15, Tbl. 5.2 (Apr. 7, 2010), https://www.atsc.org/wp-content/uploads/2015/03/Receiver-Performance-Guidelines-1.pdf.

100 See DEKRA Lab Results Test Report at 4, 12. 101 DEKRA Lab Results Test Report at 4, 13. This experiment tested multiple modulation

schemes to evaluate the impact associated with different ATSC 3.0 technology usage scenarios.

34

desired signal levels.102 For ATSC 3.0 technologies, when the undesired WSD signal was shifted

by 3 MHz from the edge of the desired broadcast ATSC 3.0 TV channel, there was a 3.6 dB

average improvement in the D/U ratio above that observed in the first adjacent channel across

both ATSC 3.0 DTV receiver models and desired signal levels.103

The Commission derived the existing WSD power limit on the first adjacent channel at

the DTV receiver using the following formula:

WSDPOWER < = (threshold sensitivity of the DTV receiver) – (D/U ratio DTV receiver)104

Using the ATSC recommended standard for the D/U ratio of -33 dB, and a threshold sensitivity

value of -84 dBm (one dB more conservative than the ATSC standard), the WSD power limit in

the first adjacent channel received by the DTV receiver was found to be -51 dBm. This means

that if the WSD power level in the first adjacent channel to DTV receiver is less than -51 dBm,

no harmful interference to the DTV receiver is predicted.

Based on the measurements of the threshold sensitivity and D/U ratio described above,

Microsoft’s lab testing also determined the WSD power limits at the DTV tuner for all the ATSC

1.0 and ATSC 3.0 receivers tested, for operations in both the first adjacent channel and in a

channel with a 3 MHz offset from the edge of the DTV channel. In all cases, the WSD power

limits were higher than those calculated based on the ATSC recommended standards.105

102 DEKRA Lab Results Test Report at 4–5, 14–15. The standard deviation of + 3.6 dB is

consistent with the variation observed across all models and desired signal strengths. 103 DEKRA Lab Results Test Report at 5, 16–17. The standard deviation of + 3.3 dB is

consistent with the variation observed across all models and desired signal strengths. 104 TVWS 2008 R&O, M&O ¶ 171, n.237. 105 See DEKRA Lab Results Test Report at 18.

35

The adjacent channel selectivity for DTV receivers as a group has improved by several

dB from the ATSC A/74 recommended D/U ratio of -33 dB. The -33 value is an average of the

D/U ratio of the first adjacent channel above and below the broadcast channel, used in lab

measurements, to account for the sideband splatter through the DTV transmitter’s emissions

mask observed in field measurements. However, the values observed in Microsoft’s lab testing

and field testing provide an updated picture of the D/U ratios that the Commission should use in

determining how to allow WSDs to operate at higher powers on channels closer to broadcast

operations. Due to the Commission’s strict conducted adjacent-channel emissions limits for

TVWS operations, WSDs do no exhibit the same splatter as DTV emissions. Consequently, the

D/U ratio measured in the lab is expected to be reproducible in the field.

Next, Microsoft’s field testing confirmed that allowing WSDs to operate at power levels

of 34 dBm EIRP on the first adjacent channel did not cause harmful interference to

broadcasters.106 The field testing was designed to measure the interactions between the selected

television receivers receiving TV signals in the presence of WSD transmissions from a base

station at varying heights above ground level. The testing was conducted in the rural community

of Grapeland, Texas, pursuant to a temporary authorization from the FCC that allowed

Microsoft’s testing partner, ESI, to conduct testing in the first adjacent channel to an actual

broadcast station, KTWC-LD, a commercial low power TV station broadcasting out of Crockett,

Texas, in the Tyler, Texas designated market area.

The test setup used a configuration designed to simulate a pair of nearby homes, one with

a TV receiver, and the other with WSD Customer Premise Equipment (“CPE”) at 24 test points

106 See Appendix B, DEKRA Certification and Testing, Inc., TVWS-DTV Coexistence on the

First Adjacent Channel, Field Measurements Test Report, at 4–5 (May 3, 2020) (“DEKRA Field Test Report”).

36

between 500 meters and 3.6 kilometers from the tower where the WSD base stations were

mounted.107 Based on the F-curves, the area of measurement received a weak to moderate

desired signal from the KTWC-TD station.

The testing measured the ability of the TV receiver to receive a broadcast test video

signal unimpaired, recorded for a one-minute test period, with a series of three different TV

receiver models, while nearby TVWS CPE receivers simultaneously received transmissions from

WSD base stations, sequentially, at three different heights. The least selective ATSC 1.0 DTV

receiver (with integral display), an ATSC 1.0 set-top box (with high measured Additive Gaussian

White Noise level), and an ATSC 3.0 plug-in receiver were taken from the lab testing to the field

for further testing. Microsoft validated the D/U ratio in the field test for each of the three DTV

receivers. A series of tests were conducted with the ATSC 1.0 DTV transmitter. Upon

completion, the ATSC 1.0 DTV transmitter was replaced by an ATSC 3.0 transmitter and the

series of test were repeated.

The test results demonstrate that in the real world, when all the signal loss mechanisms

are accounted for, a WSD base station transmitting to a WSD CPE in consumers’ homes at 34

dBm EIRP will not cause harmful interference to broadcasting operations. Even without any

form of database control or other protections, as described above, our field testing found only a

small number of impairments to broadcast standard video (dropped pixels) out of several

hundred combinations of transmitter and receiver locations, and TVWS transmitter heights.108

All but two incidents of video impairment were associated with the low-cost set-up box. For the

107 DEKRA Field Test Report at 11–12. 108 DEKRA Field Test Report at 4–5, 26–28, tbls. 8 & 9.

37

DTV receiver with the integrated display there was a total of only 2 seconds of impaired video

over 8,640 seconds of test time (0.02 percent).

Other than the persistent video impairment observed for the DTV set-up box located near

power lines, the other recorded impairments had little, if any, impact on the quality of the video,

using industry standard analytical tools and metrics. Moreover, even in these cases, it appears

doubtful that the adjacent-channel WSD was the cause of those incidences: the recorded

impairments occurred only once during the 1 minute measurement period with duration times

ranging from less than a second to three seconds.109 Overhead power lines were only an issue for

the set-top box and not the ATSC 1.0 DTV and ATSC 3.0 receivers. If the impairments were

caused by the WSD transmissions, the impairments to the video would have been persistent. This

was not the case. Because the field tests did not observe any video impairment caused by the

WSD base station signal for either the ATSC 1.0 DTV or ATSC 3.0 receiver, this phase of the

field test did not test an overly conservative 3-MHz separation from the edge of the DTV

channel, as the lab test did.

In addition to the expected terrain and clutter losses and loss due to angular misalignment

between the WSD base station antenna and the DTV antenna, significant cross polarization loss

was observed between DTV receivers, which are generally horizontally polarized, and WSD

transmissions which are vertically polarized. Our testing revealed that this loss contributed

additional isolation ranging between 9 and 27 dB.110

Measurements of the ATSC 1.0 and ATSC 3.0 transmitter power received at the DTV

receive antennas compared to the predicted values using both the F-curves and Longley-Rice

109 DEKRA Field Test Report at 26. 110 DEKRA Field Test Report at 4, 12.

38

model confirmed why the application of a terrain-based model that can take clutter into account

will allow for a more accurate assessment of whether the WSD power at the DTV receiver will

be above the WSD power limit in the first adjacent channel, or other channels. In each case, the

terrain-based prediction was substantially more accurate than the F-curve model, with an average

accuracy improvement of 6 dB. While the Commission has previously raised concerns that the

combined uncertainty of the predicted desired and undesired signal levels would be so great that

it would be impossible to authorize adjacent channel operations, our testing found that a terrain-

based model is sufficiently precise to address this issue. Although we typically found that

measured signal levels were approximately 20 dB lower than predicted using the F-curve

models, we found that this held for both the DTV signal and the WSD signal. Because the model

overestimated received signal levels for both signals, and did so by similar amounts, these two

sources of uncertainty effectively compensate for one another, resulting in de minimis net

uncertainty.

The final phase of the field test sought to implement the worst-case scenario the

Commission described in its 2008 TVWS Report and Order where a WSD is operating at 4 watts

(36 dBm), 16 meters (50 feet) away from a DTV receiver, producing a carrier level of -8.1 dBm,

assuming free space path loss.111 The Commission concluded that if the TVWS signal is “at an

azimuth in the main beam of the TV receive antenna, interference could occur to TV service at

any location where a TV signal is -51 dBm or less on a first adjacent channel.”112 However, our

real-world replication of this scenario confirmed that, even in this situation, no harmful

interference would occur. The field test closed the link between the WSD CPE and the WSD

111 See TVWS 2008 R&O, M&O ¶ 172. 112 Id.

39

Base Station mounted on a tower, with the WSD CPE operating at 33 dBm, at a very high duty

cycle. The CPE was located such that the WSD CPE was in the main beam of the DTV receiver

(with modest misalignment between the WSD CPE and DTV receiver in azimuth, but not in

elevation) and realistic polarizations of both antennas. The DTV receive antenna was separated

from the WSD CPE antenna by 16 meters. Nonetheless, the testing found no impairment of the

test video.113

Overall, the field test results demonstrated that there are a number of scenarios in which

the WSDB would be able to authorize higher-power operation on frequencies closer to the edge

of broadcast channels without causing harmful interference to broadcast services. As part of

doing so, to account for these results and more accurately reflect the actual interference

landscape, the Commission should require that WSDBs employ a terrain-based model in

performing the necessary interference calculations to determine the appropriate EIRP limit in the

first adjacent channel, with a sufficiently small cell size to ensure accuracy.114 Continuing to use

a model that does not account for terrain would be inaccurate and arbitrary. Furthermore, as

demonstrated by the lab testing, the Commission should also update its now-out-of-date

assumptions regarding broadcast receiver selectivity and D/U ratios.

113 DEKRA Field Test Report at 31–32. 114 For the appropriate grid size to determine where WSDs can safely operate, the Commission

should consider the approach it adopted for calculating interference protection in the ATSC 3.0 context under which applicants can request cell sizes of 1 km or 0.5 km instead of the default 2 km size. See Authorizing Permissive Use of the “Next Generation” Broadcast Television Standard, Report and Order and Further Notice of Proposed Rulemaking, 32 FCC Rcd. 9930, App. B (Final Rules) (2017); 47 C.F.R. § 73.616(d)(1).

40

CONCLUSION

By adopting rule changes to TVWS rules tailored to improve coverage areas and

facilitate new applications, the Commission can continue to build on the success of TVWS

technology and support a cost-effective, innovative way to expand broadband coverage. The

Commission has wide-ranging support from a broad group of stakeholders to adopt its proposed

rules. Microsoft encourages the Commission to move forward quickly with a final order that will

connect more Americans, spur innovation, and contribute to economic growth.

Respectfully submitted,

Paula Boyd Senior Director, Government and Regulatory Affairs

Michael Daum Director, Technology Policy, CELA Privacy and Regulatory Affairs MICROSOFT CORPORATION 901 K Street NW, 11th Floor Washington, DC 20001 (202) 263-5900 May 4, 2020

_______________________________ Paul Margie Paul Caritj Joely Denkinger HARRIS, WILTSHIRE & GRANNIS LLP 1919 M Street NW, Suite 800 Washington, DC 20036 (202) 730-1300

Appendix A

Page 1 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

TVWS-DTV Coexistence on the First Adjacent Channel

Laboratory Measurements

Test Report

Date: 3 May 2020

Page 2 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table of Contents 1. INTRODUCTION ............................................................................................. 3

1.1 BACKGROUND & OVERVIEW ......................................................................................... 3 1.2 SUMMARY OF OBSERVATIONS – LABORATORY MEASUREMENTS ..................................... 4

2. LAB ENVIRONMENT MEASUREMENTS ...................................................... 6

2.1 OBJECTIVE & OUTLINE ................................................................................................. 6 2.2 THRESHOLD SENSITIVITY ............................................................................................. 8 2.3 DTV-TVWS D/U RATIO............................................................................................... 9

3. APPENDIX .................................................................................................... 19

3.1 TEST EQUIPMENT ...................................................................................................... 19

IMPORTANT: NO PARTS OF THIS DOCUMENT MAY BE REPRODUCED OR QUOTED OUT OF CONTEXT, IN ANY FORM OR BY ANY MEANS, EXCEPT IN FULL, WITHOUT PREVIOUS WRITTEN PERMISSION.

Page 3 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

1. Introduction

1.1 Background & Overview

Microsoft has engaged DEKRA Certification, Inc. (“DEKRA”) to provide qualified test services during the development of an experimental test program with intent to assess whether a commercial fixed TV White Spaces (“TVWS”) network transmitter is able to operate, without causing harmful interference, in a first adjacent channel to a UHF broadcast TV station or in a channel offset by 3 MHz from either the upper or lower channel edge of a UHF broadcast TV station. For this experimental test program, the Microsoft and DEKRA engineering team members have established and applied measurement techniques described within this document to a laboratory test environment in Sterling, Virginia and a real-world field test environment in Grapeland, Texas (described in the separate TVWS-DTV Coexistence Measurements Test Report).

The objectives of the laboratory test program were: (1) measure and document receiver sensitivity thresholds for select 2018-2019 model ATSC 1.0 and ATSC 3.0 receivers; (2) document and report the calculated desired-to-undesired power ratio (D/U ratio) using select 2018-2019 model ATSC 1.0 and ATSC 3.0 digital television receivers in the presence of an undesired TVWS signal operating in the first adjacent channel of the broadcast TV channel; (3) document and report the calculated D/U ratio of select 2018-2019 model ATSC 1.0 and ATSC 3.0 receivers in the presence of an undesired TVWS signal operating in a channel that is offset by 3 MHz from the edge of the broadcast TV channel.

Based on receiver sensitivity, additive white gaussian noise (“AWGN”), and D/U ratio calculation results, two ATSC 1.0 receivers and one ATSC 3.0 receiver were selected for the field tests. Based on the calculated D/U ratio for each of the three receivers selected for the field, TVWS transmitter power limits at the DTV tuner for first adjacent channel operations and for operation in a channel with a 3-MHz offset were predicted. This information was used to determine the test point locations for the field measurements.

The simultaneous utilization of two distinct controlled test environments—one in the laboratory and one in the field—allowed the team’s test engineers to replicate and verify theoretical scenarios and compare them to real-world performance. The laboratory environment was used extensively to prove calculated models, which were then applied to a field test configuration. These two test environments were synchronized throughout the experiment to verify data being generated in each. Conditions in the field were replicated in the laboratory, while equipment and tools entering the field test environment were confirmed in the laboratory for configuration and measurement accuracy.

The scaled laboratory test environment consisted of a commercially available ATSC 1.0 DTV signal generator and an ATSC 3.0 transmitter provided by third-party Digital Television Transmitter suppliers. The engineering test team used these DTV systems to deliver controlled DTV video streams on broadcast TV channel 36 (602-608 MHz). This laboratory-generated DTV signal was used consistently for the tests determining the impact of the TVWS signal in the first adjacent channel and in the channel with the 3-MHz offset to the edge of the broadcast TV channel (593-599 MHz). A series of attenuators were used to vary the DTV signal’s power level, thus obtaining “moderately strong”, “moderate”, “weak” and “very weak” desired DTV signals for use during laboratory tests. The undesired signals were generated in the laboratory using a commercial, off-the-shelf 6Harmonics GSW4000 Series TVWS radio and an AWGN generator. The undesired signals were controlled using similar attenuation methods to obtain predicted and corresponding signal levels found in the field.

Page 4 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

For the series of measurements conducted in the first adjacent channel, the undesired TVWS signal was tuned to broadcast TV channel 35 (596-602 MHz). For the testing in the channel with a 3- MHz offset, the undesired TVWS signal was tuned to channel 34.5 (593-599 MHz). The GWS4000 series TVWS radio was operated as close as possible to a download-only mode and configured with the highest duty cycle possible to simulate multiple TVWS devices operating in proximity.

The Device(s) Under Test (“DUT”) within the laboratory environment used multiple models of digital ATSC 1.0 television receivers spanning quality and price points for these receivers. When testing ATSC 3.0 technology as the desired signal, the DUT was commercially available; however, selection and qualities were limited.

For each measurement, the same “standard” video stream was broadcast as a continuous loop from the DTV transmitter. Industry standard analysis tools were used to assess the sensitivity threshold and the undesired power level (at a threshold of visibility) that causes impairment to the video image.

Because the FCC limits the power of fixed TVWS devices operating on the first adjacent channel to TV broadcast channels to 40 mW, during this experiment within the laboratory all testing was done in an isolation chamber which, as configured, also served the purpose of reducing ambient and unwanted RF noise, adding to the accuracy of the measurements. Field testing was conducted under an experimental license authorizing these transmissions.

During the laboratory testing, laboratory test data was collected, extrapolated, and applied to the experimental TVWS field test environment.

1.2 Summary of Observations – Laboratory Measurements • The minimum required D/U ratio between the desired DTV signal and the undesired TVWS signal

in the first adjacent channel across all ATSC 1.0 DTV receivers (with an integral display) and desired signal levels was measured to range between -40.3 to -46.7 dB. In comparison, for DTV-to-DTV interference, the D/U ratio recommended by the ATSC A/74 standard on the first adjacent channel for moderately strong and weak desired signals is -33 dB and the FCC’s DTV planning factors are -26 dB and -28 dB, respectively, for the first adjacent channel above and below the broadcast channel.

• The measured D/U ratio between the desired DTV signal and the undesired TVWS signal in the first adjacent channel for all ATSC 1.0 DTV set-top boxes tested (without an integral display) ranged between -33.8 and -42.4 dB, with a mean of -41.9. The AGWN level measured at the threshold of visibility for each of the three DTV set-top boxes ranged between 13 to 14 dB above that of the DTV receiver with the lowest AGWN level. Examination of the internal components of the RF design of the DTV set-top boxes indicated minimal isolation was considered in the RF design of these very low price point products.

• The minimum required D/U ratio between the desired DTV signal and the undesired TVWS signal in the first adjacent channel for the high-definition ATSC 3.0 DTV receiver (without an integral display) was measured to range between -43 dB to -64 dB, depending on the modulation rate, error correction code, and desired signal level. This experiment tested multiple modulation schemes to evaluate impact associated with the different ATSC 3.0 technology usage scenarios provided.

• On average, when the undesired TVWS signal was shifted by 3 MHz from the edge of the desired broadcast ATSC 1.0 TV channel, there was a 5.7 dB improvement in the minimum required D/U

Page 5 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

ratio below that observed in the first adjacent channel across all ATSC 1.0 DTV receiver models and desired signal levels. The standard deviation of + 3.6 dB was consistent with the variation observed across all models and desired signal strengths.

• On average, when the undesired TVWS signal was shifted by 3 MHz from the edge of the desired

broadcast ATSC 3.0 TV channel, there was a 3.6 dB improvement in the minimum required D/U ratio below that observed in the first adjacent channel across both ATSC 3.0 DTV receiver models and desired signal levels. The standard deviation of + 3.3 dB was consistent with the variation observed across all models and desired signal strengths.

Page 6 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2. Lab Environment Measurements

2.1 Objective & Outline

A controlled laboratory environment was created to determine the baseline threshold sensitivity for selected DTV receivers (“DTVR”) and for determining the desired-to-undesired (“D/U”) power ratio for DTV receivers in the presence of an undesired TVWS signal on the first adjacent channel to the digital television channel and at a 3-MHz offset to the edge of the digital television broadcast channel. The desired signal for testing the ATSC 1.0 DTVRs was generated by DEKTEC model DTU-315. The desired signal for the testing of ATSC 3.0 DTVRs was generated by HITACHI-COMARK model Exact V2. The undesired TVWS signal was generated by a 6Harmonics 4000 Series Fixed TV White Space transmitter (“TVWS transmitter”).

The DTV transmitters, DTVRs, and the TVWS transmitter were placed inside an isolation chamber within DEKRA’s test laboratory facility in Sterling, VA. The DTV transmitter was connected to the signal analyzer and to the DTVR via RF cables, an attenuator, a splitter, and a combiner (see Figure 1 below). The signal analyzer was used to monitor and measure the signal levels received by the DTVR from the DTV transmitter while simultaneously monitoring and measuring the TVWS transmitted signal levels presented to DTVR within this configuration. These signal levels were adjusted to specified levels by changing (1) the power output of the DTV transmitter, (2) the power of the TVWS transmitter, and/or (3) adjusting attenuator value. The goal was to establish the specified, relatively low power levels required for each measurement while minimizing the wideband noise.

DTVR performance was measured by analyzing a benchmark video stream as presented to the DTVR and determining a Threshold of Visibility (“TOV”). For this experiment, the TOV is the level at which the video image remains error free during at least 60 seconds of video transmission playback. A Difference Mean Opinion Score (“DMOS”) and the Video Multimethod Assessment Fusion (“VMAF”) metrics are used to score the quality of the standard video stream recorded from the DTVRs using an external video quality analysis tool. Table 1 below provides the DMOS and VMAF values used to determine the quality scoring for the test data generated.

Metric Video Quality Scoring

Excellent Good Fair Poor Bad DMOS 0 - 20 20 - 40 40 - 60 60 - 80 80 - 100 VMAF 100 - 80 80 - 60 60 - 40 40 - 20 20 - 0

Table 1: DMOS and VMAF Video Quality Scoring Chart

Consistent with the document ATSC Recommended Practice: Receiver Performance Guidelines, Document A/74: 2010, 7 April 2010, the D/U ratio for each ATSC 1.0 DTVR was determined to be 1) -43 dBm for a moderately strong desired signal, 2) -53 dBm for a moderate desired signal, 3) -65 dBm for a weak desired signal, and 4) -80 dBm for a very weak desired signal. Due to the magnitude of the sideband splatter associated with simple, stringent, and full-service DTV emissions masks, TVWS transmitters would not be expected to operate in the first adjacent channel where there is a strong desired DTV signal.

For the ATSC 3.0 signal, in addition to using the four desired signal strength levels, the physical layer pipe (“PLP”) was configured with four representative modulation levels (QPSK, 16QAM, 64QAM, and 256QAM), including error correction rates.

Page 7 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

A sample of ATSC 1.0 DTVRs were selected for testing, covering a range of display formats, display resolutions, and consumer price points:

• LED backlit LCD DTVRs with integral displays—two each, with resolutions of 720-P (High definition), 1080-P (Full High Definition) and 2160-P (4K)—were selected. The testing also sought to determine whether the D/U ratio varies with DTV display resolution as a proxy for DTVR price point.

• OLED DTVR with an integral display and a resolution of 2160-P (4K). • All digital-to-analog and digital-to-digital set-top box DTVRs that did not have an integral

display panel had low price points. Two of the three models included an indoor sheet antenna that could be affixed to a wall, which indicated that these receivers were intended for indoor use where there is a strong desired signal.

The ATSC 3.0 DTVRs are a relatively new technology and commercial availability is very limited. The two ATSC 3.0 DTVR models secured for the experiment were in the form of USB “dongle” plug-in receivers. One ATSC 3.0 dongle was advertised as providing a high definition image and the other dongle was advertised as providing an ultra-high definition image. The USB dongles were connected to a high-resolution PC and monitor, where the ATSC 3.0 receiver suppliers delivered a specific, PC-based software player and drivers to play the received reference test transport video stream.

Figure 1 provides a block diagram of the laboratory test environment.

Figure 1: Block Diagram of Laboratory Test Set-up

A select group of commercially available DTVRs was identified and later technically characterized for this test experiment. The intent with the DTVR selection identified in “Table 2: DTV Receiver Information” is to diversify test samples over different consumer price points, display resolutions, physical display dimensions, and DTVR form factors.

Table 2 provides a summary of DTVR information used during this test experiment.

DTVR ID Manufacturer Model Standard Profile Resolution Display

Size RX1 Samsung UN65NU8000 ATSC 1.0 TV 4K 65 inch

RX2 TCL 55S517 ATSC 1.0 TV 4K 55 inch

RX3 Hisense 40H3080E ATSC 1.0 TV 1080p 40 inch

RX4 Samsung UN32N5300AFXZA ATSC 1.0 TV 1080p 32 inch

Page 8 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

RX5 Insignia NS-24DF310NA19 ATSC 1.0 TV 720p 32 inch

RX6 Toshiba 32LF221U19 ATSC 1.0 TV 720p 32 inch

RX7 LG OLED55B8PUA ATSC 1.0 TV 4K 55 inch

RX8 Leelbox1 ATSC DZ001 ATSC 1.0 Set-top Box 1080p -

RX9 iDOO1 Converter-0A ATSC 1.0 Set-top Box 1080p -

RX10 Mediasonic1 HOMEWORX HW130STB ATSC 1.0 Set-top Box 1080p -

RX11 Lowasis LDA 1100/1200 ATSC 3.0 USB Dongle UHD -

RX12 RedZone TVXPLORER BUNDLE ATSC 3.0 USB Dongle HD -

Table 2: DTV Receiver Information

Note 1: The Leelbox and the iDOQ set-top box performs digital-to-analog conversion. The Mediasonic set-top box performs digital-to-digital conversion.

2.2 Threshold Sensitivity Threshold sensitivity, or Minimum Signal Level (“MSL”), defines the nominal carrier power at the receiver input terminals required for the proper operation of DTVR.

Measurement Procedure for establishing DTVR Threshold Sensitivity

• An ATSC DTV signal was generated on channel 34 (590-596 MHz) using a DTV transmitter. • DTVR was connected to the experiment setup as shown above in Figure 1: Block Diagram

of Laboratory Test Set-up. • The path attenuation between DTV transmitter and DTVR was adjusted in steps of 1 dB. • The video quality on the DTVR was quantified.

o If video quality was at TOV, the current level was reported as the sensitivity of the DTVR.

o If the video is error free, the path attenuation was varied. • Repeat steps for all DTVR in the experiment test bed.

Table 3 provides the receiver sensitivity threshold and AGWN level (at the TOV for a -80 dBm desired signal) of the ATSC 1.0 DTV receivers used for the subsequent D/U measurements.

DTVR ID Standard Sensitivity (dBm) AGWN Level at TOV (dBm) RX1 ATSC 1.0 -83.79 -87.25 RX2 ATSC 1.0 -87.78 -86.20 RX3 ATSC 1.0 -83.46 -87.20 RX4 ATSC 1.0 -82.52 -83.65 RX5 ATSC 1.0 -87.00 -83.65 RX6 ATSC 1.0 -83.45 -85.57 RX7 ATSC 1.0 -84.09 -87.26 RX8 ATSC 1.0 -86.94 -69.78 RX9 ATSC 1.0 -85.24 -86.32 RX10 ATSC 1.0 -87.18 -70.94

Table 3: Threshold Sensitivity and AGWN Level of ATSC 1.0 DTVRs

Page 9 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Observations: The data suggests that threshold sensitivity of the ATSC 1.0 DTVRs varies from -82 to -87 dBm. The ATSC A/74 recommended value is -83 dBm. Contributing factors impacting sensitivity threshold levels include: 1) DTV RF isolation within the hardware design, e.g., shielding and grounding, 2) tuner gain-control, and 3) RCVR filtering characteristics. The AGWN levels of two of the three set-top boxes were 13 to 14 dB above that of the DTV receiver with the lowest AGWN, indicating a higher noise figure.

Table 4 provides the measured threshold sensitivity of ATSC 3.0 standard DTV Receivers operating at different modulation standards.

DTVR ID Standard Modulation Sensitivity (dBm)

RX11 ATSC 3.0

QPSK_7-15_CodeRate -84.78 16QAM_7-15_CodeRate -81.31 16QAM_11-15_CodeRate -79.55 64QAM_11-15_CodeRate -76.45 256QAM_7-15_CodeRate -78.45

RX12 ATSC 3.0

QPSK_7-15_CodeRate -83.01 16QAM_7-15_CodeRate -83.76 16QAM_11-15_CodeRate -82.56 64QAM_11-15_CodeRate -79.43 256QAM_7-15_CodeRate -82.56

Table 4: Threshold Sensitivity of ATSC 3.0 DTVRs

Observations: The data suggests that threshold sensitivity of ATSC 3.0 DTVR varies from -78 to -85 dBm. Lower modulation schemes provided better sensitivity, while higher modulation schemes provided lower sensitivity. The devices tested, RX11 and RX12, are the first commercially available ATSC 3.0 DTV receivers.

2.3 DTV-TVWS D/U Ratio The D/U ratio is the ratio of desired to undesired signal power required for artifact-free

operation of the DTVR. For this experiment, the D/U ratio is a measure of DTVR adjacent channel selectivity. In 2008, the FCC adopted rules allowing personal/portable TVWS devices to operate at a radiated power of up to 40 mW e.i.r.p on a first adjacent channel to television broadcast stations.1 In its analysis, the FCC used the D/U ratio for a DTV broadcast station operating in a first adjacent channel to another DTV broadcast station as a proxy for the D/U ratio of a TVWS transmitter operating on a first adjacent channel to a broadcast television station. The ATSC A/74 document recommended a D/U ratio of -33 dB, on a first adjacent channel for DTV-to-DTV interference. The A/74 recommended value averages the D/U ratio on the first adjacent channel above and below the broadcast channel and takes into account the DTV transmitter sideband splatter in the field. Even though the TVWS transmission does not have the sideband splatter, the FCC adopted the -33 dB value for TVWS-to-DTV interference. The FCC applied the same methodology in 2015 when it permitted a fixed TVWS transmitter to operate at up to 40 milliwatts in the first adjacent channel to a

1 See Unlicensed Operation in the TV Broadcast Bands, Second Report and Order and

Memorandum Opinion and Order, 23 FCC Rcd. 16807, ¶ 176 (2008).

Page 10 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

TV broadcast station.2 In its 2017 First Report and Order on Next Generation Television, the FCC established that the primary ATSC 3.0 stream is protected on its first adjacent channel using a D/U ratio of -33 dB, concluding that there was no reason to believe that ATSC 3.0 receivers will perform any differently than DTV receivers, but explaining that it would revisit whether “either the co-channel or adjacent channel interference protection criteria for ATSC 3.0 should be any different from the interference protections provided for DTV in OET Bulletin No. 69,” if it received additional information or conducted receiver tests.3

DEKRA’s laboratory measurements were intended to validate the following hypotheses: (1) DTVRs’ required D/U ratios in the first adjacent channel have improved since the D/U ratio of -33 dB for TVWS-to-DTV interference in the first adjacent channel was adopted, (2) the D/U ratio of an ATSC 1.0 DTVR will show further improvement if the TVWS transmitter operates with a 3-MHz offset to the DTV station’s channel edge, and (3) ATSC 3.0 transmissions using an OFDM standard should allow for a more robust D/U ratio in both the first adjacent channel and in a channel with a 3-MHz offset than that of an ATSC 1.0 receiver transmitting using the 8-VSB standard.

For these tests, the default configuration of the 6Harmonics GSW4000 TVWS system was used, in which the modulation and coding scheme (“MCS”) rate is set to Auto and the transmitter is operating with a fully allocated duty cycle (>95% utilization). The reason for testing with a fully allocated duty cycle is to simulate a scenario where there are multiple fixed TVWS transmitters in the vicinity of a DTVR, each operating with bursts of data transmission.

Measurement Procedure for establishing DTV to TVWS D/U Ratio

• An ATSC DTV signal was generated (1) on channel 36 (602–608 MHz) for the adjacent channel and 3-MHz offset test conditions, and (2) on channel 34 (590–596 MHz) for co-channel conditions using a DTV transmitter.

• The DTVR was connected to the experiment setup as shown above in Figure 1: Block Diagram of Laboratory Test Set-up.

• The TVWS wireless system was configured to operate in the desired test configuration: (1) channel 35 (596 – 602 MHz) for first adjacent, (2) channel 34 (590–596 MHz) for cochannel, and (3) channel 34.5 (593–599 MHz) for 3-MHz offset.

• The TVWS signal level received at the input port of the DTVR was increased in steps of 1dB using path attenuation.

• The video quality on the DTVR was quantified. o If the video quality was at TOV, the sensitivity of the DTVR was reported at the current

level. o If the video was error free, the path attenuation was varied.

• The measurement was repeated for moderately strong (-43 dBm), moderate (-53 dBm), weak (-65 dBm), and very weak (-80 dBm) DTV signals at the DTVR (see Figures 2 through 5 below).

• The measurement was repeated across all DTV receivers supporting ATSC 1.0 and ATSC 3.0 standards (and MCS).

2 See Amendment of Part 15 of the Commission’s Rules for Unlicensed Operations in the Television

Bands, Repurposed 600 MHz Band, 600 MHz Guard Bands and Duplex Gap, and Channel 37 Reference, et al., Report and Order, 30 FCC Rcd. 9551, ¶ 64 (2015).

3 See Authorizing Permissive Use of the ‘Next Generation’ Broadcast Television Standard, Report and Order and Further Notice of Proposed Rulemaking, 32 FCC Rcd 9930, ¶ 112 (2017).

Page 11 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figures 2 to 5 are images of the DTV signal at test-specified desired signal levels, operating with a test-specified undesired TVWS signal transmitting in the first adjacent channel.

Figure 2 Figure 3

Strong DTV Transmitter Signal and Adjacent TVWS Signal Moderate DTV Transmitter Signal and Adjacent TVWS Signal

Figure 4 Figure 5

Weak DTV Transmitter Signal and Adjacent TVWS Signal Very Weak DTV Transmitter Signal and Adjacent TVWS Signal

Page 12 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table 5 provides the D/U ratio measured for each DTVR when the undesired TVWS signal was operating in the first adjacent channel.

DTV Receiver Type Threshold (dBm)

TV Signal

-43 dBm -53 dBm -65 dBm -80 dBm

D/U (dB)

RX1 -83.79 -40.55 -41.42 -42.19 -40.43

RX2 -87.78 -45.10 -45.01 -43.95 -42.30

RX3 -83.46 -45.97 -45.89 -45.39 -44.32

RX4 -82.52 -41.14 -42.06 -41.93 -40.28

RX5 -87 -47.46 -46.75 -45.75 -45.96

RX6 -83.45 -46.58 -45.94 -43.44 -45.87

RX7 -84.09 -44.91 -45.91 -42.78 -46.21

RX8 -86.94 -33.82 -38.75 -42.37 -40.91

RX9 -85.24 -35.76 -36.82 -38.21 -40.64

RX10 -87.18 -38.28 -40.64 -40.38 -40.88

Table 5: D/U Ratio for ATSC 1.0 DTVRs with the TVWS Signal Operating in the First Adjacent Channel for Different Desired Signal Strengths

Table 6 summarizes the D/U ratio by DTVR type when an undesired TVWS signal was received in the first adjacent channel. The D/U ratio below is an average value for the number of receivers within that category (e.g., each LED 720-P DTVR result was summed and divided by the total number of LED 720-P DTVRs). Based on the limited data set, there does not appear to be any discernible pattern of variation across DTV receiver type.

DTV Receiver Type Threshold (dBm) TV Signal

-43 dBm -53 dBm -65 dBm -80 dBm D/U (dB)

LED 720-P (2) -85.23 -47.02 -47.12 -45.10 -45.92 LED 1080-P (2) -82.99 -43.56 -44.14 -44.90 -42.30 LED 2180-P (2) -85.79 -42.83 -47.06 -45.22 -41.37

OLED (1) -84.09 -44.91 -45.91 -42.78 -46.21

Table 6: D/U Ratio for ATSC 1.0 DTVR with the TVWS Signal Operating in First Adjacent Channel by Display Resolution

Page 13 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table 7 summarizes the D/U ratio for ATSC 3.0 receivers when the undesired TVWS signal was operating in the first adjacent channel.

DTV Receiver Type

Threshold (dBm) Modulation

TV Signal -43

dBm -53

dBm -65

dBm -80

dBm D/U (dB)

UHD

-84.78 QPSK_7-15_CodeRate -55.78 -56.10 -58.99 -59.61

-81.31 16QAM_7-15_CodeRate -53.41 -54.06 -55.95 -48.56

-79.55 16QAM_11-15_CodeRate -49.71 -49.74 -50.02 -39.84

-76.45 64QAM_11-15_CodeRate -44.50 -45.07 -43.98 -36.49

-78.45 256QAM_7-15_CodeRate -48.01 -46.84 -47.87 -38.83

HD

-84.78 QPSK_7-15_CodeRate -61.03 -62.85 -64.18 -63.91

-81.31 16QAM_7-15_CodeRate -55.07 -57.68 -58.58 -59.45

-79.55 16QAM_11-15_CodeRate -43.25 -47.12 -47.85 -51.56

-76.45 64QAM_11-15_CodeRate -45.24 -46.07 -47.29 -49.39

-78.45 256QAM_7-15_CodeRate -42.91 -49.26 -52.08 -56.35

Table 7: ATSC 3.0 DTVR D/U Ratio with the TVWS Signal on First Adjacent Channel

Observations: (1) The D/U ratio of all the ATSC 1.0 receivers tested appears to be better than the -33 dB value used in the FCC rules. (2) Set-top box DTVRs present an inferior D/U ratio compared to other ATSC 1.0 TV receivers. This inferior performance may be related to the quality of the tuner and shielding. Figure 6 and Figure 7 below show pictures of set top box RX10, disassembled to the board level. (3) The required D/U ratio of ATSC 3.0 receivers tested appears to be approximately 10 dB lower for lower modulations and similar to ATSC 1.0 receivers while operating at 64QAM and 256QAM.

Figure 6 Figure 7

Page 14 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table 8 shows the required D/U ratio measured for each DTVR when the TVWS signal was operating in a channel with a 3-MHz offset from the edge of the desired broadcast TV signal.

DTV Receiver Type Threshold (dBm)

TV Signal -43 dBm -53 dBm -65 dBm -80 dBm

D/U (dB)

RX1 -83.79 -46.35 -45.71 -45.01 -42.93 RX2 -87.78 -48.78 -48.40 -45.43 -46.07 RX3 -83.46 -49.56 -49.42 -47.87 -47.95 RX4 -82.52 -42.54 -42.06 -41.76 -40.24 RX5 -87.00 -47.46 -48.29 -46.75 -45.96 RX6 -83.45 -54.19 -55.36 -57.62 -56.54 RX7 -84.09 -53.23 -53.47 -52.78 -51.67 RX8 -86.94 -36.95 -46.13 -48.64 -45.58 RX9 -85.24 -47.38 -45.64 -49.03 -46.16 RX10 -87.18 -39.24 -43.21 -47.68 -49.80

Table 8: D/U Ratio of ATSC 1.0 DTVRs for the TVWS Signal Operating with a 3-MHz Offset to the Edge of the Desired Broadcast DTV Signal

Table 9 summarizes the required D/U ratio measured for ATSC 1.0 DTVRs, by display type, when the TVWS signal was operating at a 3-MHz offset channel from the edge of the desired broadcast DTV signal.

DTV Receiver Type Threshold (dBm) TV Signal

-43 dBm -53 dBm -65 dBm -80 dBm D/U (dB)

LED 720-p (2) -85.23 -54.84 -55.33 -57.05 -56.45 LED 1080-p (2) -82.99 -46.05 -45.74 -44.82 -44.10 LED 2180-p (2) -85.79 -47.57 -47.06 -45.22 -44.50

OLED (1) -84.09 -53.23 -53.47 -52.78 -51.67

Table 9: ATSC 1.0 DTVR D/U ratio, by DTV Receiver Resolution, with TVWS Operating at 3-MHz Offset Channel From the Edge of the Desired Broadcast TV Signal

Page 15 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table 10 provides the calculated D/U ratio delta values for each DTVR when the TVWS transmitter is operating on the adjacent channel versus when the TVWS transmitter operates on a 3-MHz offset channel from the edge of the desired broadcast TV signal. (see Figures 8 through 12, illustrating the traces and delta measured).

DTV Receiver Type Threshold (dBm) TV Signal

-43 dBm -53 dBm -65 dBm -80 dBm Delta D/U (dB)

RX1 -83.79 -5.80 -4.29 -2.82 -2.00 RX2 -87.78 -3.68 -3.39 -1.48 -3.77 RX3 -83.46 -3.59 -3.53 -2.48 -3.63 RX4 -82.52 -1.40 0.33 0.17 0.04 RX5 -87.00 -8.03 -7.00 -9.72 -10.39 RX6 -83.45 -7.61 -9.42 -14.18 -10.67 RX7 -84.09 -8,32 -7.56 -10.00 -5.46 RX8 -86.94 -3.13 -7.38 -6.27 -4.67 RX9 -85.24 -11.62 -8.82 -10.82 -5.52 RX10 -87.18 -0.96 -2.57 -6.85 -8.92

Table 10: Change in ATSC 1.0 DTVR D/U Ratio Between the TVWS Signal Operating in a Channel with a 3-MHz offset to the DTV Signal Edge and Operating in the First Adjacent

Channel

Observation: On average, across all DTVRs, the D/U ratio improves (is more selective) by 5.7 dB when the TVWS signal is offset by 3 MHz from the edge of the desired broadcast TV signal as compared to operating on the first adjacent channel. If RX4 is removed from the average calculation, the average D/U ratio change would be even greater.

Page 16 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table 11 presents the required D/U ratio for the two ATSC 3.0 receivers when the undesired TVWS signal was operating in a channel with a 3-MHz offset from the edge of the desired broadcast DTV signal. As in the case of the undesired TVWS signal operating in the first adjacent channel, the D/U ratio was measured for different ATSC 3.0 transmitter modulation levels and code rates.

Table 12 illustrates the difference in the D/U ratio for both ATSC 3.0 receivers when the TVWS signal was operating in a channel with a 3-MHz offset from the edge of the desired broadcast DTV signal and when it was operating in the first adjacent channel.

DTV Receiver Type

Threshold (dBm) Modulation

TV Signal -43

dBm -53

dBm -65

dBm -80

dBm D/U (dB)

UHD

-84.78 QPSK_7-15_CodeRate -59.92 -60.53 -60.36 -63.21

-81.31 16QAM_7-15_CodeRate -57.09 -58.04 -57.72 -52.84

-79.55 16QAM_11-15_CodeRate -53.76 -53.86 -54.51 -35.25

-76.45 64QAM_11-15_CodeRate -49.95 -49.89 -49.48 -40.39

-78.45 256QAM_7-15_CodeRate -49.87 -51.09 -47.86 -51.16

HD

-84.78 QPSK_7-15_CodeRate -66.10 -69.00 -68.08 -70.13

-81.31 16QAM_7-15_CodeRate -59.32 -59.18 -60.20 -61.13

-79.55 16QAM_11-15_CodeRate -46.88 -47.60 -47.96 -49.05

-76.45 64QAM_11-15_CodeRate -53.28 -55.07 -52.93 -49.94

-78.45 256QAM_7-15_CodeRate -54.11 -54.06 -56.33 -52.37

Table 11: D/U Ratio for ATSC 3.0 DTVR with TVWS Signal Operating With a 3-MHz Offset Below the Edge of the Desired Broadcast DTV Signal

DTV Receiver Type

Threshold (dBm) Modulation

TV Signal -43

dBm -53

dBm -65

dBm -80

dBm D/U (dB)

UHD

-84.78 QPSK_7-15_CodeRate -4.14 -4.43

-1.37 -3.60

-81.31 16QAM_7-15_CodeRate -3.68 -3.98 -1.77 -4.28

-79.55 16QAM_11-15_CodeRate -4.05 -4.12 -4.49 4.59

-76.45 64QAM_11-15_CodeRate -5.45 -4.82 -5.5 -3.90

-78.45 256QAM_7-15_CodeRate -1.86 -4.25 0.01 -12.33

HD -84.78 QPSK_7-

15_CodeRate -5.07 -6.15 -3.90 -6.22

-81.31 16QAM_7-15_CodeRate -4.25 -1.50 -1.62 -1.68

Page 17 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

DTV Receiver Type

Threshold (dBm) Modulation

TV Signal -43

dBm -53

dBm -65

dBm -80

dBm D/U (dB)

-79.55 16QAM_11-15_CodeRate -3.63 -0.48 -0.11 2.51

-76.45 64QAM_11-15_CodeRate -8.04 -9.00 -5.64 -0.55

-78.45 256QAM_7-15_CodeRate -11.2 -4.80 -4.25 3.98

Table 12: Change in ATSC 3.0 DTVR D/U ratio between the TVWS operating 3 MHz Below the Edge of the Desired Broadcast DTV signal and Operating in the First Adjacent

Channel

Observations: (1) Although current FCC rules do not define the D/U ratio in a 3-MHz offset configuration, the D/U ratio of all ATSC 1.0 receivers tested using a TVWS undesired signal offset at 3 MHz appears to have an even lower impact than a first adjacent channel interferer. (2) A similar result was observed with the D/U ratio of ATSC 3.0 receivers tested using a TVWS undesired signal offset at 3 MHz, as shown in Figures 8 to 11.

Figure 8 Figure 9

Strong DTV Transmitter Signal and 3 MHz Offset TVWS Signal Moderate DTV Transmitter Signal and 3 MHz Offset TVWS Signal

Figure 10 Figure 11

Weak DTV Transmitter Signal and 3 MHz Offset TVWS Signal Very Weak DTV Transmitter Signal and 3 MHz Offset TVWS Signal

Page 18 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

TVWS Power Limit on DTVR Tuner When Operating in the First Adjacent Channel

The TVWS power limit in the first adjacent channel that would not be detected by a DTV receiver in the desired channel under the current FCC rules is given by:

(-84 dBm) [sensitivity threshold in desired channel] – (-33 dB) [D/U ratio] = -51 dBm [TVWS power limit at DTVR tuner]

Applying the same analysis to RX4, RX10, and RX12 (at 256QAM) to determine the TVWS power limit in the first adjacent channel at the DTVR yields the values shown below. The D/U ratios below are based on the average values for the moderate and weak desired signals:

• RX4: (-82.5 dBm) – (-42 dB) = -40.5 dBm • RX10: (-87.2) – (-40.5) = -46.7 dBm • RX12: (-78.4) – (-47.3) = -31.2 dBm

Note that these values are considerably lower than the appropriate corresponding regulatory power limit for a TVWS transmitter because these results measure only the limit at the DTVR and therefore do not take effects such as propagation loss into account.

A similar analysis can be performed for the TVWS radio operating in a channel with a 3-MHz offset from the edge of the DTV channel. The corresponding predicted TVWS power limits are as follows:

• RX4: -41.5 dBm • RX10: -41.7 dBm • RX12: -29.0 dBm

Page 19 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

3. Appendix

3.1 Test Equipment Table 13 provides details about the DTV Transmitters used as a part of this experiment.

Standard Manufacturer Model Serial No. ATSC 1.0 DEKTEC DTU-315 315.001.267 ATSC 3.0 HITACHI-COMARK Exact V2 00121-H106

Table 13: DTV Transmitter Information Table 14 provides additional equipment used for measurements.

Manufacturer Model Reference

Keysight/Agilent MXA 9202 Spectrum

Analyzer

https://www.keysight.com/en/pdx-x202266-pn-N9020A/mxa-signal-analyzer-10-hz-to-265-

ghz?cc=US&lc=eng

Anritsu MS2720T Spectrum

Analyzer https://www.anritsu.com/en-us/test-

measurement/products/ms2720t

AccepTv Video Quality

Analyzer http://www.acceptv.com/en/products_vqa.php

6Harmonics TVWS 4000 Series

Radios

http://www.6harmonics.com/wp-content/uploads/2017/02/GWS-4000-Series-

Datasheet-FCC-certified-July-2016.pdf

Rohde Schwarz SMU200A Vector Signal Generator

https://www.rohde-schwarz.com/us/product/smu200a-productstartpage_63493-7555.html

Table 14: Additional Test Equipment Information

Page 20 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Disclaimer © 2020, DEKRA Certification, Inc. Please note that use of this document is conditioned by the following:

The information in this document is subject to change by DEKRA Certification, Inc. without notice.

No warranties express or implied, are given by DEKRA Certification, Inc. All implied warranties, including implied warranties of merchantability, fitness for a particular purpose, and non-infringement are disclaimed and excluded by DEKRA Certification, Inc.

The document does not constitute an endorsement, recommendation or guarantee of any of the products (hardware or software) mentioned. The document does not guarantee that there are no errors of defects in the products or that the products will meet the reader’s expectations needs or specifications, or that they will operate without interruption.

This document does not imply any endorsement, sponsorship, affiliation, or verification by or with any organizations mentioned in this document.

Under no circumstances shall DEKRA Certification, Inc. be liable for any direct, indirect, incidental, special or consequential damages arising from any error or omission in this document.

Appendix B

Page 1 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

TVWS-DTV Coexistence on the First Adjacent Channel

Field Measurements

Test Report

Date: 3 May 2020

Page 2 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Table of Contents

1. INTRODUCTION ............................................................................................. 3

1.1 BACKGROUND & OVERVIEW ..................................................................................... 3

1.2 SUMMARY OF OBSERVATIONS .................................................................................. 4

2. FIELD MEASUREMENTS ............................................................................... 5

2.1 DTV-TVWS MEASUREMENT ................................................................................... 6

2.1.1 BROADCAST TELEVISION STATION……………………………………………………....6

2.1.2 SELECTING THE TEST AREA LOCATION…………………..……………………...……...7

2.1.3 ESTABLISHING THE TEST POINTS WITHIN THE TEST AREA…………………………....11

2.1.4 CHANNEL OCCUPANCY SCAN WITHIN THE TEST AREA……………………….….…... 13

2.1.5 TEST SITE DOCUMENTATION PROCEDURE………………………….…………….….. 14

2.1.6 MEASUREMENT PROCEDURE……………………..…………………………………...15

2.1.7 TEST POINT LOCATIONS RELATIVE TO TRANSMITTERS AND BACKGROUND NOISE……15

2.1.8 PREDICTED AND MEASURED POWER AT EACH TEST POINT……..……………………16

2.1.9 IMPACT OF TERRAIN ON MEASURED TVWS DEVICE POWER…………….…………...18

2.1.10 DESIRED AND UNDESIRED SIGNAL POWER AT THE DTV RECEIVER………………..…21

2.1.11 IMAGE QUALITY ASSESSMENT…………………………………………….…………. 23

2.1.11.1 IMPAIRMENT TO THE RX4 DTVR………………………………………………………23

2.1.11.2 IMPAIRMENT TO THE RX10 DTVR……………….……………………………….......24

2.1.11.3 ASSESSMENT OF IMPAIRMENTS TO RX4 AND RX10 DTVS…………………………..26

2.1.11.4 IMPAIRMENT TO THE RX12 DTVR…………………………………………………… 27

2.2 BREAKPOINT MEASUREMENT ............................................................................... 28

2.3 UPLINK TEST USING A SEPARATION DISTANCE OF 16 METERS..…………………….31

3. APPENDIX .................................................................................................... 33

3.1 TEST EQUIPMENT ............................................................................................... 33

IMPORTANT: No parts of this document may be reproduced or quoted out of context, in any form or by any means, except in full, without the previous written permission

Page 3 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

1. Introduction

1.1 Background & Overview Microsoft has engaged DEKRA Certification, Inc. (“DEKRA”) to provide qualified test services

during the development of an experimental test program to assess whether a commercial fixed TV White Spaces (“TVWS”) transmitter can operate in a first adjacent channel to a broadcast TV station or in a channel with a 3-MHz offset to the edge of a broadcast TV station without causing harmful interference. For this experimental test program, the Microsoft and DEKRA engineering team members established a laboratory test environment in Sterling, Virginia and a real-world field test environment in Grapeland, Texas. The field test was conducted in partnership with Edge Spectrum Inc., a licensee of the adjacent-channel broadcast TV station, and under Special Temporary Authority granted to Edge Spectrum Inc. The report on the laboratory test phase is provided under separate cover.

At the conclusion of the laboratory testing, three DTV receivers (“DTVRs”) were selected for the field test. The ATSC 1.0 models were manufactured by RX4 (Samsung) and RX10 (Mediasonics). The ATSC 3.0 model, RX12, was manufactured by RedZone.

The simultaneous utilization of two distinct controlled test environments—one in the laboratory and one in the field—allowed the team’s test engineers to verify theoretical scenarios and compare them to real-world performance. The laboratory environment was used extensively to prove calculated models, which were then applied to a field test configuration. These two test environments were synchronized throughout the experiment to verify data being generated in each. Conditions in the field were replicated in the laboratory, while equipment and tools entering the field test environment were confirmed in the laboratory for configuration and measurement accuracy.

The scaled field test environment operating in the area of Grapeland, Texas consisted of a commercial digital television transmission station broadcasting on KTWC-LD (channel 34) in Crockett, Texas. The field test environment also incorporated three TVWS transmit/receive radios, mounted on a tower just west of Grapeland, TX at heights of 30 meters, 60 meters, and 85 meters above ground level (“AGL”). For the series of field measurements conducted in the first adjacent channel, the undesired TVWS radios were tuned to operate in TV channel 35 (596-602 MHz). This TVWS network in Grapeland, TX was designed and established as a virtual point-to-multipoint TVWS network. The generated TVWS undesired signal was used to achieve field measurements at 24 distinct individual test points. These 24 distinct test points were selected based on predicted DTV signal strength values in relation to the predicted TVWS signal strength values. The desired signal to undesired signal ratio (“D/U ratio”) data values from laboratory measurements were used in predicting the location and operating range of the TVWS devices, as well as power limits on the first adjacent channel.

The objectives of the field test were to: (1) validate the laboratory-derived D/U ratio for each of the three DTVRs, (2) compare the predicted versus measured received power for the DTV and TVWS transmission at each test point, (3) at each test point determine whether the full-power simulated TVWS network operating on channel 35 would cause impairment to a video broadcast in both ATSC 1.0 and ATSC 3.0 format on channel 34, and if so, characterize any impacts of the impairment using industry standard tools, and (4) measure the impact of adjacent-channel TVWS operations in a worst-case scenario, where the DTV antenna is in line with TVWS link and separated from the TVWS transmitter by 16 meters to simulate a TVWS radio placed outdoors on a rooftop in proximity to a neighbor’s DTV.

Page 4 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

A continuous test video stream was broadcast in a looped feed by KTWC-LD, first using an ATSC 1.0 television transmitter, then an ATSC 3.0 television transmitter. At each test point, two vans (“Van 1” – TVWS and “Van 2” – DTVR), were placed end-to-end to simulate residences. Van 1 and Van 2 each were configured with trailers supporting a 10 meter AGL high antenna. At each test point, the TVWS customer premise equipment modem (“CPE”) within Van 1 always pointed the 10-meter AGL antenna toward the tower where TVWS Base Stations (BS) were mounted at heights of 30, 60 and 85 meters AGL. Each TVWS BS operated at 34 dBm EIRP (23 dBm conducted power with an 11 dBi antenna). The TVWS CPE operated at 33 dBm (23 dBm conducted power with a 10 dBi antenna). The 10-meter AGL DTV antenna on Van 2 was always pointed toward the KTWC-LD transmitter. In Van 2, industry-standard diagnostics software was used to assess any degradations in image quality over the one-minute measurement period at each test point when each of the three TVWS transmitters were operated individually and in sequence.

The test environments for lab and field allowed efficient measurement of DVTR performance under test conditions showing best, average, and worst case for the signal transmissions mentioned above. The lab and field test platforms were configured using emulated and real world commercial Digital Television transmission equipment, respectively. The laboratory test environment provided data to characterize the impact to the DTVR when it is exposed to signal levels beyond existing regulatory limits.

1.2 Summary of Observations • This field test environment provided real world empirical test data relevant to the objectives stated

above. The lab and field test platforms were configured to provide user experience data for DTVR reception while exposed to TVWS transmitters emitting adjacent and co-channel RF signals.

• There were significant differences observed between the predicted and measured value of the received ‘desired’ DTV power and received ‘undesired’ TVWS signal at the DTV receiver. The measured values were lower than the predicted values. The difference in predicted and measured value at any one location was likely due to a combination of terrain, clutter, angular misalignment between antennas, and cross-polarization loss. Measured cross-polarization loss was significant and ranged between 9 and 27 dB. Consequently, the measured TVWS power limit on the first adjacent channel at the DTVR predicted in the lab for RX4, RX10, RX12 was never reached in the field.

• The D/U ratios for RX4, RX10, and RX12 established through laboratory experiments were validated in the field measurements. In one of the test scenarios, the DTVR and TVWS CPE antennas were brought within meters of each other and depolarization was introduced to characterize the amount of isolation provided between the horizontal and vertical polarization of the two different antennas.

• For the ATSC 1.0 DTV receiver (RX4), there was one instance of impairment observed at two

test locations, each lasting a second, which represents 0.02 percent of total measurement time of 8640 seconds. The cause of the impairments could not be determined. The impairments could not be reproduced.

• For the ATSC 1.0 Set-Top box, essentially a circuit board housed in a plastic case designed for

indoor use, a total of 20 incidences of impairments were observed at 10 of 24 test points. At six of these test points, only one instance of impairment was observed, each for one second or less. Three impairments were observed at two test points each. The cause of these impairments could

Page 5 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

not be determined. At one test point, located in the shadow of overhead power lines, the measured background noise was greater than the receiver sensitivity threshold and impairment was observed for all six measurements. Excluding the high background noise test site, the total time of impairment was less than 19 seconds in a measurement time of 8280 seconds (0.23 percent). Outside of the test point with the high background noise, the recorded video for 10 of the occurrences registered image quality in the good-to-excellent range and 4 of the occurrences registered image quality in the fair-to-good range.

• There were no instances of visual impairment observed with the ATSC 3.0 receiver.

• DEKRA reproduced the uplink scenario in the FCC 2008’s Report and Order, which the

Commission cited in support of its decision not to allow higher power TVWS devices to operate on a first adjacent channel to a broadcast TV station. The DTV antenna was situated in the path between the TVWS transmitter and the TVWS CPE, separated by a horizontal distance of 16 meters from the TVWS CPE operating at 33 dBm EIRP. Both the DTV antenna and the CPE antenna were 10 meters above ground level. No video impairment was observed.

2. Field Measurements Following on the baseline data collection from the laboratory environment, three DTVR were selected for the field measurements. RX4 was an ATSC 1.0 DTV receiver. RX10 was an ATSC 1.0 set-top box. RX12 was an ATSC 3.0 receiver. Each receiver was the least selective receiver in its cohort. The objectives of the field measurement program were: (1) measure the TVWS and DTV power received at the DTV antenna and compare them with predicted values; (2) characterize TVWS impact on DTVR operation at multiple locations within the test area; (3) validate the D/U ratio laboratory measurements in the field; and (4) conduct a 16 meter separation uplink test.

A test environment was created to measure (1) Background Noise measurement, (2) DTV signal level at DTVR, (3) TVWS signal level, and (4) DTV-TVWS impact.

Figure 1 below is a block level view of field environment created.

Page 6 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 1: Block Diagram of Field Test Set-up

2.1 DTV-TVWS Measurement

2.1.1 Broadcast Television Station The ATSC 1.0 and ATSC 3.0 signals were transmitted over KTWC-LD, a low-power TV station broadcasting on channel 34. Its community of license, Crockett, TX, is located in the Tyler, TX DMA. The field measurements were performed in partnership with the licensee, Edge Spectrum, Inc (“ESI”). ESI obtained special temporary authority from the FCC to conduct these measurements.

The KTWC-LD transmitter was located at a latitude of 31 degrees, 17 minutes, 37 seconds north and a longitude of 95 degrees, 28 minutes, 54 seconds west (NAD 83 coordinates). The transmitter operated at 9 kW ERP with a horizontally polarized non-directional antenna and used a stringent out-of-channel emissions mask. The height of the radiation center above mean sea level (RCAMSL) was 280.9 meters and its Height Above Average Terrain (HAAT) was 182 meters. KTWC-LD’s protected contour is shown in Figure 2 below.

Page 7 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 2: Coverage Area of KTWC-LD

A series of measurements were performed using the existing ATSC 1.0 transmitter. Upon completion of the ATSC 1.0 measurements, the licensed transmitter was replaced by an ATSC 3.0 transmitter. Once the measurements using the ATSC 3.0 transmitter were completed, the ATSC 1.0 transmitter was restored to its original location.

2.1.2 Selecting the Test Area Location Grapeland, Texas was established as the measurement area based on: (1) a willing broadcast partner and available channels, (2) the DTV coverage map using F-curves and Longley-Rice models predicted field strength in the moderate to weak signal range, (3) availability of a suitable tower within KTWC’s protected contour, and (4) a representative community for fixed TVWS use.

(1) The testing was a collaborative effort between Microsoft, ESI, and DEKRA. Candidate communities were ones where (a) ESI held a broadcast license and (b) there are at least three contiguous vacant channels to its broadcast channel, either above or below in frequency. The three contiguous vacant channels on one side of KTWC allow for field measurements using a TVWS transmitter operating in the first adjacent channel and then in the center 6 MHz of the first two adjacent channels. The vacant third adjacent channel ensures there is at least a channel and a half separation between the TVWS transmitter operating on the center 6 MHz channel and the nearest incumbent broadcaster (in the event that an ESI station is not the nearest incumbent).

(2) The predicted desired DTV signal strength was in the moderate to weak range. The coverage map of each of the shortlisted communities was examined. KTWC was selected.

Page 8 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 3 below is KTWC’s coverage map obtained from the RabbitEars website that uses the terrain-based Longley-Rice model.1 Note that RabbitEars is a website accessible via the desktop. The purple line represents the station’s protected 51 dBu contour.

Figure 3: KTWC-LD Coverage Map Using the RabbitEars Web Site

The coverage map in Figure 4 below was derived using the Nominet WAVEDB TVWS database tool depicting received signal levels in dBm. Note that the solid magenta line represents the 41 dBu contour. Under the FCC’s rules, low power TV stations and translators are protected from other television stations to their 51 dBu contour but are protected to its 41 dBu contour with respect to TVWS devices. There is a roughly 11 km wide annulus between the 41 dBu and 51 dBu contours where TVWS devices cannot operate.

1 Rabbit Ears, KTWC-LD Coverage Map,

https://www.rabbitears.info/contour.php?map=Y&appid=e47cefa85d714c09bd2fcf6212b6c0ea&site=1 (last visited May 1, 2020).

Page 9 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 4: KWTC Coverage Map Using the Nominet WAVEDB Tool

(3) Availability of a suitable tower

Within KTWC’s protected contour, identifying a suitable tower to place the TVWS radios for the field measurement campaign proved challenging. Issues included tower availability, tower height above ground level, representative communities that would be in range of a TVWS transmitter operating on the tower, other service(s) broadcasting from the tower (and in the vicinity), and structural integrity, tower loading, and required remediation. One aspect of the testing was to determine whether a terrain-based computer model can be used to permit higher power TVWS operations on the first adjacent channel. Obtaining different received TVWS signal strengths at different TVWS base station transmitter heights would indicate that terrain and clutter are factors that needed to be taken into account in the FCC protection criteria.

(4) Representative community

The search for a representative community was performed in parallel with looking for a suitable tower for placement of the TVWS transmitter(s). The ideal representative community was determined to be a small town, located within 11 km from a suitable tower, where there are homes in some regular pattern and where some residents are expected to use an outdoor TV antenna to receive their DTV signal. In planning the field test, there is less concern over high power operations in the first adjacent channel causing harmful interference to the DTVR at the solitary farmhouse many kilometers away from the TVWS tower and more concern about a homeowner with an outdoor TVWS CPE causing harmful interference to her or his neighbor’s DTVR by way of his or her rooftop antenna.

Page 10 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Grapeland, Texas was selected as the location for the field measurements. Grapeland is at the intersection of U.S. Highway 287 and Farm Road 227, twelve miles north of Crockett in northern Houston County.2 The town is roughly 2 square miles. The 2010 census put its population at 1,489.

The TVWS transmitter tower site is approximately 1 km due west of Grapeland. The maximum height for placing a TVWS transmitter on the tower was 85 meters above ground level.

Figure 5 below overlays the TVWS tower site with the DTV coverage plot. Grapeland is roughly due north of the KTWC transmitters and due east of the TVWS transmitter.

Figure 5: Relative Location of the TVWS Base Station Within the KTWC Protected Contour

Figure 6 provides an expanded view of the predicted DTV field strength in Grapeland immediately surrounding the area, taking into account terrain but not clutter or any other losses. The area met our criteria for a moderate-to-weak received DTV signal.

2 Grapeland.com, http://www.grapeland.com/History/history.html (last visited May 1, 2020).

Page 11 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 6: Expanded View of DTV Coverage Map in Grapeland, Texas

2.1.3 Establishing the Test Points Within the Test Area The TVWS base station site was a tower located at the coordinates (31°29'7.80"N, 95°29'48.84"W). There were three 6Harmonics 4000 Series TVWS radios placed on the tower at heights of 30 meters, 60 meters, and 85 meters above ground level, corresponding to HAATs of 162 meters, 192 meters, and 217 meters. Under the FCC rules, the UHF band in Grapeland is considered a less congested area.

Each TVWS transmitter operated at 23 dBm conducted power. The antenna was a 6Harmonics BTS (90 degree) Panel Antenna (SL12948B) with a gain of 11+1 dBi. For the front lobe, the antenna’s 3 dB points was 35 +/- 6 degrees and its 1 dB points was approximately 15 degrees. Each TVWS antenna was pointed to the horizon (elevation angle = 0 degrees). The transmitted beam of each TVWS transmitter was vertically polarized.

The test plan identified 24 test points, covering an area approximately 19.5-24 kilometer north of the KTWC transmitter and 0.5-3.6 km east of the WSD tower. The 24 Test Points are shown in Figure 7 below.

Page 12 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 7: Grapeland TVWS Tower and Test Points

At each test point, to simulate the length of a typical residence, DEKRA oriented two vans lengthwise, each towing a small trailer containing an antenna at a height of 10 meters above ground level. In one trailer was a 10 dBi DTV (Yagi) antenna, linearly polarized horizontally. At each test point, the DTV antenna was pointed toward the DTV tower to maximize the received signal. This van housed equipment to capture and analyze the video broadcast from the DTV tower. In the other trailer was the TVWS CPE attached to a 10 dBi (LPDA) antenna. At each test location, the TVWS CPE was pointed toward the TVWS base station to maximize the received signal. Figure 8 shows a typical orientation of the two vans and the respective DTV and TVWS CPE antennas.

Due to the relative geometry of the TVWS base station (due east) and the DTV tower (due south), and cross polarization loss, the DTV antenna was essentially isolated from the TVWS CPE, although it may not look this way from the picture taken at one of the Test Points. A measured cross-polarization isolation varied between 9 dB and 27 dB depending on the azimuth and alignment of the antennas.

Figure 8: Typical Test Point Set Up

Page 13 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2.1.4 Channel Occupancy Scan Within the Test Area Based on the initial test plan, background scans were taken using a spectrum analyzer at the highest point within the 8 km circumference with the receiver antenna oriented at 60 degrees, 180 degrees, and 300 degrees from true north. The KTWC transmitter was approximately 180 degrees due south of the spectrum scan testing point. The measurements captured the occupied channel and the received power in both horizontal and vertical polarizations. Figure 9 shows the location where the spectrum scan was performed relative to the TVWS BS tower.

Figure 9: Location of Spectrum Scan Relative to the TVWS Tower

The spectrum scans for the horizontal and vertical polarization due south (180 ͦ azimuth) are presented respectively in Figure 10 and Figure 11. The KTWC-LD signal was horizontally polarized. While not confirmed, presumably the vertically polarized signal received is due to depolarization of the broadcast television signal. The signals on channel 49 and channel 50 are from mobile operations in the lower 700 MHz band.

Horizontal Polarization – 180 ͦ azimuth

Figure 10: Spectrum Occupancy and Received Power for Horizontally Polarized Antenna

Vertical Polarization – 180 ͦ Azimuth

Figure 11: Spectrum Occupancy and Received Power for Vertically Polarized Antenna

Page 14 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2.1.5 Test Site Documentation Procedure

The following steps were performed at each selected Test Point prior any test in the field environment to ensure the test location selected met the test requirements:

• Documented transmitter site hardware and propagation characteristics:

o Transmitter site location coordinates (e.g., latitude and longitude);

o Hardware (e.g., block diagram components such as data sources, STLs, exciter, intermediate power amplifier, high power amplifier, harmonic filter, emission mask filter, feedline, antenna, antenna gain, azimuth and elevation patterns/polarization, etc.); and

o Signal (e.g., data modes and configurations, TPO, ERP, adjacent channel emission mask compliance, signal SNR quality, etc.).

• Deployed Test Setup

o Confirmed proper operation of transmitter and field test vehicle equipment, calibrated field test truck.

o Confirmed feasibility of raising antenna to 30 feet AGL without encountering obstructions such as tree limbs or overhead wires.

o If site was not suitable for testing or was unreachable, moved to closest suitable location (within 0.5-mile radius).

o Attached antenna to mast, connected feedline, and raised antenna to 30 feet AGL.

o Using a spectrum analyzer, oriented antenna for maximum average ATSC signal strength (in dBm) and recorded this azimuth angle (in degrees) and recorded the antenna orientation sensitivity.

• Documented receive site hardware characteristics

o Test site location coordinates (e.g., latitude and longitude), including street names, building names/addresses for fixed outdoor.

o Took pictures of the test site surroundings.

o Hardware description (e.g., block diagram of components such as calibrated test antenna, impedance matching devices, coaxial feedlines, step attenuators, amplifiers, splitters, receivers, GPS devices, manual and automated test equipment, computers, etc.).

o Computed the received power at the household antenna (10m AGL) of each TV station, using terrain data and the Longley-Rice point-to-point model.

o Computed the path loss between the TVWS base station (assumed to have an omnidirectional antenna) and the household, using terrain data and the Longley-Rice point-to-point model.

o Computed the maximum tolerable nuisance power, co-channel and in all adjacent channels, so as to keep the CNR above the threshold plus a margin.

Page 15 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2.1.6 Measurement Procedure Once the site was documented, the following steps were performed to conduct the necessary field measurements:

• DTV and TVWS system installed as shown in Figure 1: Block Diagram of Field Test Set-up.

• Disabled ATSC 1.0 or ATSC 3.0 DTV signal transmission. o Performed background noise measurement using a spectrum analyser on channel 34

(590 – 596 MHz) and channel 35 (596 – 602 MHz).

• DTV broadcast station operated on channel 34 (590 – 596 MHz). Measured the received DTV signal at DTVR receiver.

• Configured TVWS base station (30m height AGL) on channel 35 (596 – 602 MHz) and associated CPE to complete application level connectivity.

• Initiated traffic from between CPE and base station in uplink and downlink standalone. Measured the TVWS signal level at the input of DTVR.

• Quantified video quality on the DTVR.

• Repeated the test combination for antenna height of 60m AGL and 85m AGL.

• Repeated measurement across all DTV receivers supporting ATSC 1.0 and ATSC 3.0 standards.

2.1.7 Test Point Locations Relative to Transmitters and Background Noise The KTWC transmitter was located at coordinates (31°17'37.00"N, 95°28'54.01"W) and was configured to broadcast an industry standard video test stream within the FCC licensed parameters. This was the same video test stream that was used in the laboratory environment. The TVWS tower is located at coordinates (31°29'7.80"N, 95°29'48.84"W). For each test point, the azimuth and horizontal distance from each test point to the DTV tower and the TVWS base station was determined. The TVWS CPE azimuth relative to the TVWS base station was determined using the Nominet WAVEDB tool for TVWS devices operating on channel 36.

Additionally, with the KTWC transmitter turned off, background noise measurements were taken on channel 34 at each of the 24 test points. The TVWS receiver can detect a signal down to -98 dBm in a 6 MHz channel. Note that there was a relatively high background noise level at Test Point 110, which is in the range of DTVR sensitivity threshold.

Table 1 below provides: (1) the latitude and longitude for each Test Point, (2) the azimuth angle (from the North) to the Test Location looking towards the DTV tower and the TVWS BS tower and (3) the measured background noise.

Page 16 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Test Point Latitude Longitude

RX Azimuth to TVWS BS

(⁰)

Distance to TVWS BS

(km)

RX Azimuth to DTV Tower

(⁰)

Distance to DTV Tower (km)

Background Noise (dBm)

1 31.5066 -95.4788 210 2.804 181 23.5 -94.6

4 31.5073 -95.4716 220 3.430 182 23.8 -96

9 31.5001 -95.4788 220 2.361 181 23.0 -96.1

10 31.5001 -95.4749 220 2.645 184 23.0 -96.2

67 31.4787 -95.4685 285 2.82 185 20.6 -96.4

68 31.4683 -95.4651 300 3.588 180 19.5 -96.3

77 31.4969 -95.4820 220 1.906 181 22.5 -93.4

78 31.4970 -95.4789 230 2.139 181 22.5 -96.5

81 31.4955 -95.4789 230 2.048 181 22.5 -93

83 31.4913 -95.4794 250 1.789 181 22.0 -92.1

86 31.4901 -95.4777 255 1.906 179 21.9 -90.8

89 31.4967 -95.4855 220 1.653 178 22.5 -94.2

95 31.4937 -95.4885 220 1.219 180 22.2 -93.6

97 31.4945 -95.4817 225 1.758 178 22.4 -92.3

98 31.4911 -95.4885 225 1.029 179 21.9 -94.2

100 31.4898 -95.4864 240 1.112 180 21.9 -96.1

101 31.4918 -95.4827 240 1.526 180 22.0 -94.8

102 31.4902 -95.4815 250 1.565 180 22.9 -93.6

104 31.4862 -95.4893 270 0.797 178 21.4 -93.0

105 31.4858 -95.4904 250 0.629 178 21.4 -91.0

106 31.4855 -95.4923 270 0.462 177 21.4 -91.7

110 31.4850 -95.4827 270 1.363 180 21.2 -84.5

111 31.4870 -95.4789 270 1.728 181 21.6 -96.1

114 31.48291 95.4836 280 1.304 180 21.1 -94.9

Table 1: Test Point Locations Relative to Transmitter and Background Noise

2.1.8 Predicted and Measured Power at Each Test Point Table 2 below compares the predicted DTV field strength using the F-curves and the measured field strength at the DTV antenna at a height of 10 meters above ground level. The predicted field strength of KTWC’s signal on channel 34 at each Test Point was determined using the FCC website for calculating FM and TV propagation curves3 and converted from dBu to dBm.

3 FCC, FM and TV Propagation Curves, https://www.fcc.gov/media/radio/fm-and-tv-propagation-

curves (last visited May 1, 2020).

Page 17 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Test Point

Distance to DTV Tower

(km)

Predicted Field Strength

at DTV Antenna using F-Curves

(dBm)

Predicted Field Strength

at DTV Antenna using Longley-Rice4

(dBm)

Measured Field Strength

at DTV Antenna (dBm)

1 23.5 -38.73 -42.22 -56.6 4 23.8 -38.99 -43.23 -56.3 9 23.0 -38.29 -44.04 -56.1 10 23.0 -38.29 -44.10 -65.3 67 20.6 -36.15 -43.32 -62.4 68 19.5 -35.15 -57.92 -61.8 77 22.5 -37.84 -43.36 -56.8 78 22.5 -37.84 -43.33 -58.7 81 22.5 -37.84 -42.93 -66.7 83 22.0 -37.4 -43.14 -50.1 86 21.9 -37.31 -41.74 -57.2

89 22.5 -37.84 -43.54 -65.8 95 22.2 -37.58 -43.61 -59.3 97 22.4 -37.76 -43.62 -59.7 98 21.9 -37.31 -41.61 -61.7 100 21.9 -37.31 -41.60 -58.3 101 22.0 -37.40 -42.74 -66.8

102 22.9 -38.20 --43.23 -59.0 104 21.4 -36.86 -41.92 -60.9 105 21.4 -36.86 -43.69 -50.7 106 21.4 -36.86 -44.25 -54.2 110 21.2 -36.69 -40.04 -54.8 111 21.6 -37.04 -41.37 -60.8 114 21.1 -36.60 -41.06 -51.4

Table 2: Comparison of Measured vs Predicted (F-Curve) KTWC Field Strength at Each Test Point

The measured values of the DTV power were consistent with the field strength ranges in the Nominet coverage map and were significantly less than those predicted by the F-curves.

Next, the field strength of the TVWS signal operating on channel 35 was measured at both the TVWS receiver and DTV receiver, then compared with the value(s) and prediction(s) made using the free space path loss model and the Nominet WAVEDB’s line-of-sight model for the TVWS base stations mounted at 30 meters AGL, 60 meters AGL, and 85 meters AGL. The measurements shown in Table 3 are for the TVWS transmitter mounted at 85 meters AGL.

4 Rabbit Ears, https://www.rabbitears.info/ (last visited May 1, 2020) (note that RabbitEars

coverage maps use the Longley-Rice prediction method and assume a receive height of 13 feet).

Page 18 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Test Point

Distance to TVWS

Tower (km)

Predicted TVWS Field

Strength Using FSPL

(dBm)

Predicted TVWS Field Strength at TVWS CPE

Nominet Tool on Ch. 36

(dBm)

Measured Field

Strength at TVWS CPE

(dBm)

Measured TVWS Field

Strength At DTV Receiver

(dBm)

1 23.5 -52.95 -57.4 -57.6 -76.6 4 23.8 -54.70 -57.5 -57.9 -76.3 9 23.0 -51.46 -54. -66.1 -82.3 10 23.0 -52.64 -54.8 -66.5 -83.4 67 20.6 -51.16 -53.4 -70.1 -85.0 68 19.5 -55.09 -56.5 -73.7 -85.2 77 22.5 -49.60 -52.0 -63.7 -80.0 78 22.5 -50.60 -52.4 -60.3 -76.1 81 22.5 -50.22 -51.8 -63.8 -81.2 83 22.0 -49.05 -49.7 -53.5 -80.9 86 21.9 -49.60 -50.0 -46.9 -80.9

89 22.5 -48.36 -51.3 -64.1 -81.7

95 22.2 -45.72 -48.6 -63.5 -81.1

97 22.4 -48.90 -50.8 -60.1 -81.2

98 21.9 -44.24 -46.2 -66.1 -81.4 100 21.9 -44.92 -45.9 -49.6 -76.3 101 22.0 -47.67 -48.7 -58.7 -76.6

102 22.9 -47.89 -48.4 -53.2 -76.6

104 21.4 -42.03 -42.3 -45.0 -69.4

105 21.4 -39.97 -40.1 -43.2 -64.7

106 21.4 -37.29 -37.0 -44.2 -66.7

110 21.2 -46.69 -46.9 -52.5 -76.3

111 21.6 -48.75 -49.0 -52.3 -76.3

114 21.1 -46.30 -46.7 -52.1 -75.3

Table 3: Comparison of Measured and Predicted Field Strength of the TVWS BS at each Test Point.

The difference between the last two columns reflects TVWS signal loss due to the combination of path loss between the DTV and TVWS CPE antennas (10-15 meters), angular misalignment between the DTV and TVWS BS antennas, beam splitter between DTV antenna and the DTVR, and the cross-polarization loss.

2.1.9 Impact of Terrain on Measured TVWS Devices Power Due to terrain (and clutter) loss, some variation was observed in the received TVWS based station power at the TVWS CPE and the DTV receiver for the three different TVWS transmitter heights.

Page 19 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

To better understand whether terrain is a significant loss mechanism, the commercial Nominet WAVEDB tool was used to examine the terrain profile of TVWS links operating on channel 36 at select test sites at each of the 3 TVWS transmitter heights.

Figures 12, 13, and 14 show the terrain profile for the links between the TVWS BS transmitter and the TVWS CPE at Test Point 1 (Farm-to-Market Road 228 and Maple Street) for TVWS transmitter heights of 30 meters, 60 meters, and 85 meters above ground level. Test Point 1: FM 228 and Maple

Figure 12: Terrain Profile of TVWS Link at Test Point 1 for Base Station at 30 m AGL

Figure 13: Terrain Profile of TVWS Link at Test Point 1 for Base Station at 60 m AGL

Figure 14: Terrain Profile of TVWS Link at Test Point 1 for Base Station at 85 m AGL

For the TVWS base station that operated at 30 meters above ground level, based on the link profile, there was a reduction in received signal at the Test Point above that predicted by the free

Page 20 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

space path loss model due to terrain. Based on the terrain profile of the link, improvement in the received signal strength can be expected as the height of the TVWS radio is increased. The measurements did show a stronger received signal from the TVWS base station mounted at 85 meters AGL than those mounted at 60 meters AGL and 30 meters AGL.

There were other Tests Points where the terrain profile of a comparable link on channel 36 indicated that terrain should be taken into account in any propagation model used for TVWS devices.

The terrain profile of Test Point 77 shown below in Figures 15, 16, and 17 is an example where a line-of-sight propagation model provides good accuracy.

Test Point 77 – CR 2345 & Church Street

Figure 15: Terrain Profile of TVWS Link at Test Point 77 for Base Station at 30 m AGL

Figure 16: Terrain Profile of TVWS Link at Test Point 77 for Base Station at 60 m AGL

Figure 17: Terrain Profile of TVWS Link at Test Point 77 for Base Station at 85 m AGL

The received power for all three TVWS base station heights was within measurement error. The presumption in these cases is that terrain was not a factor.

Overall, the difference in TVWS power received at the DTVR between the TVWS transmitter operating at 30 meters height AGL and the TVWS transmitter operating at 85 meters height AGL ranged between 0 and 3.5 dB. In general, the higher the TVWS transmitter, the greater the measured power. The exceptions were for the Test Points closest to the TVWS tower. The hypothesis is that the slightly lower power at higher transmitter heights in these cases was due to the antenna pattern. Each TVWS base station was pointing to the horizon (0 degrees elevation) and in the vertical plane, the full beam width was 35 degrees.

Page 21 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2.1.10 Desired and Undesired Signal Power at the DTV Receiver In Table 4, for each test point, the desired ATSC 1.0 power on channel 34 and the undesired TVWS power on channel 35 is presented. These values are applicable to measurements for DTVR RX4 and RX10.

Test Point Distance to TVWS

Tower (km)

Desired DTV Signal at the RX4 and RX10 DTVR

(dBm)

Undesired Downlink TVWS Signal at the RX4 and RX 10

DTVR (dBm)

1 2.804 -58.6 -76.6

4 3.43 -64.3 -76.3

9 2.361 -64.1 -82.3

10 2.645 -73.3 -83.4

67 2.82 -70.4 -85.0

68 3.588 -69.8 -85.2

77 1.906 -64.8 -80.0

78 2.139 -66.7 -76.1

81 2.048 -74.7 -81.2

83 1.789 -58.1 -80.9

86 1.906 -65.2 -80.9

89 1.653 -73.8 -81.7

95 1.219 -67.3 -81.1

97 1.758 -67.7 -81.2

98 1.029 -69.7 -81.4

100 1.112 -66.3 -76.3

101 1.526 -78.4 -76.6

102 1.565 -67 -76.6

104 0.797 -68.9 -69.4

105 0.629 -58.7 -64.7

106 0.462 -62.2 -66.7

110 1.363 -62.8 -76.3

111 1.728 -68.8 -76.3

114 1.304 -59.4 -75.3

Table 4: Desired Signal (Channel 34) and Undesired Signal (Channel 35) Power at RX4 and RX10

Page 22 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

In Table 5, for each Test Point, the desired ATSC 3.0 power on channel 34 and the undesired TVWS power on channel 35 is presented.

Test Point Distance to TVWS

Tower (km)

Desired DTV Signal at the RX12 DTVR

(dBm)

Undesired Downlink TVWS Signal at the TX12 DTVR

(dBm)

1 2.804 -58.8 -76.5

4 3.43 -57.6 -76.3

9 2.361 -62.2 -85.5

10 2.645 -82.6 -85.6

67 2.82 -76.9 -85.8

68 3.588 -55.7 -85.8

77 1.906 -63.1 -80.7

78 2.139 -70.2 -76.4

81 2.048 -65.7 -81.1

83 1.789 -62.9 -80.6

86 1.906 -62.1 -80.9

89 1.653 -81.1 -81.8

95 1.219 -67.8 -80.5

97 1.758 -73.5 -81.9

98 1.029 -64.8 -81.2

100 1.112 -76.1 -75.4

101 1.526 -74.2 -77.1

102 1.565 -66.9 -76.4

104 0.797 -73.1 -69.5

105 0.629 -58.5 -66.3

106 0.462 -64.5 -66.2

110 1.363 -64 -75.6

111 1.728 -65 -75.9

114 1.304 -54.9 -75.4

Table 5: Desired Signal (Channel 34) and Undesired Signal (Channel 35) Power at RX12

Page 23 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2.1.11 Image Quality Assessment

At each Test Point, the desired test video signal broadcast on channel 34 was recorded in the presence of the undesired TVWS signal on channel 35 operating from transmitters at 30 meters AGL, 60 meters AGL, and 85 meters AGL, each for one minute. DEKRA staff visually noted incidents of impairment to the broadcast video for RX4, RX10, and RX12. The video feeds captured by the RX10 and RX12 DTVRs were recorded for later playback and analysis. After the field testing was completed for the day, the video was analyzed using industry standard diagnostic tools to confirm visual observations. Tables 6 and Table 7 show, for RX4 and RX10 DTVRs, respectively, the TVWS BS heights where impairment was observed and whether the impairment was observed on the downlink or uplink transmissions. 2.1.11.1 Impairments to the RX4 DTVR

Test Location

TVWS Signal at CPE (dBm)

TVWS Signal at DTVR (dBm)

RX4 Receiver DL UL

1 -57.9 -76.7

4 -58.8 -76.8

9 -74.2 -85.3

10 -68.1 -86.8

67 -70.1 -86

68 -79.8 -86.4 Impairment at 60 m

77 -70.2 -80

78 -70.7 -76.3

81 -69.3 -81.8

83 -70.5 -80.7

86 -66.7 -80.8

89 -72.4 -82.8

95 -65.1 -81.2

97 -69.7 -81.2

98 -68.3 -81.9 Impairment at 60 m

100 -60.9 -76.5

101 -66.6 -76.9

102 -62.3 -76.7

104 -41.3 -68.4

Page 24 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Test Location

TVWS Signal at CPE (dBm)

TVWS Signal at DTVR (dBm)

RX4 Receiver DL UL

105 -39.8 -65.5

106 -53.6 -66.2

110 -64.7 -76.9

111 -64.1 -76.7

114 -57.9 -76.6

Table 6: Impairments to the RX4 DTVR

Observations: (1) TVWS signal measured at the CPE was higher than signal at DTVR due to cross-polarization, as well as alignment and azimuth of the antennas. (2) Two test points showed impairments on the DTV video reception with the operation of TVWS signal. (3) Impairment duration was around 1 second in a duration of 60 seconds of video. (4) No impairments were observed at test points while doing a retest. (5) Potential causes of impairments include multi-path reflection of TVWS CPE from nearby equipment.

2.1.11.2 Impairments to the RX10 DTVR

Test Location

TVWS Signal

at CPE

(dBm)

TVWS Signal at

DTVR (dBm)

RX10 Receiver

DL UL

1 -57.9 -76.7 Impairment at 60m

4 -58.8 -76.8

9 -74.2 -85.3

10 -68.1 -86.8 Impairment at 60m Impairment at 60m, 85m

67 -70.1 -86.0

68 -79.8 -86.4 Impairment at 60m, 85m

77 -70.2 -80

78 -70.7 -76.3

81 -69.3 -81.8

83 -70.5 -80.7

86 -66.7 -80.8

89 -72.4 -82.8

Page 25 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Test Location

TVWS Signal

at CPE

(dBm)

TVWS Signal at

DTVR (dBm)

RX10 Receiver

DL UL

95 -65.1 -81.2 Impairment at 60m

97 -69.7 -81.2

98 -68.3 -81.9 Impairment at 60m

100 -60.9 -76.5

101 -66.6 -76.9

102 -62.3 -76.7 Impairment at 85m

104 -41.3 -68.4

105 -39.8 -65.5

106 -53.6 -66.2 Impairment at 30m

110 -64.7 -76.9 Impairment at 30m, 60m, 85m Impairment at 30m, 60m, 85m

111 -64.1 -76.7 Impairment at 85m

114 -57.9 -76.6 Impairment at 60m, 85m Impairment at 60m

Table 7: Impairments to the RX10 DTVR

Observations: (1) TVWS signal measured at the CPE was higher than the signal at the DTVR due to cross-polarization, as well as alignment and azimuth of the antennas. (2) 20 test conditions showed impairments in the DTV video reception with the operation of TVWS signal. Three Test Points accounted for 12 of the 20 occurrences. (3) Impairment duration was around 1-2 seconds in a video duration of 60 seconds apart from test point 110. (4) Test 110, accounting for 6 of the 20 test conditions showing impairment, had overhead power lines nearby, resulting in a much higher noise floor. The noise floor was approximately 3 dB above the TOV for RX10. (5) If impairment was observed at lower TVWS transmitter height, it was expected to also be observed at greater heights – but this was often not the case. (6) Potential impairments could be due to poor shielding on the DTV receiver (installed in a plastic casing with no additional shielding).

Page 26 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

2.1.11.3 Assessment of Impairments to the RX4 and RX10 DTVR Table 8 below provides information about the frequency, duration, and impact on video image quality arising for each occurrence of impairment observed.

Test Point Direction Antenna

Height Impairments

DMOS3 VMAF4 Duration (s) Frequency Visible on Screen

RX4

68 UL 60m <1 Once Yes - -

98 UL 60m <1 Once Yes - -

RX10

1 UL 60m <1 Once Yes 5.52 93.26

10

DL 60m 1 Once Yes 19.34 80.53

UL 60m <1 Once Yes 9.26 89.55

UL 85m <1 Once Yes 8.23 91.08

68 UL 60m 1 Once Yes 7.68 92.04

UL 85m 1 Once Yes 9.11 89.88

95 UL 60m 2 Once Yes 24.58 71.38

98 UL 60m <1 Once Yes 11.55 87.56

102 UL 85m <1 Once Yes 11.55 87.55

106 DL 30 m 2 Once Yes 24.56 79.82

110

DL 30m >30 Whole Video Yes 83.97 6.88

DL 60m >30 Whole Video Yes 85.21 6.32

DL 85m >30 Whole Video Yes 84.73 6.59

UL 30m 5 Once Yes 69.55 10.23

UL 60m 5 Once Yes 65.22 11.05

UL 85m 4 Once Yes 62.89 11.83

111 DL 85m <1 Once Yes 12.56 84.17

114

DL 60m 3 Once Yes 31.88 72.56

DL 85m 2 Once Yes 29.62 75.17

UL 60m <1 Once Yes 6.55 93.57

Table 8: Frequency, Duration, and Video Impact of RX4 and RX10 Impairments 3DMOS Score: 0-20 Good to Excellent (Green), 20-40 Fair to Good (Yellow), 40-100 Fair to Bad (Red) 4VMAF Score: 80-100 Good to Excellent (Green), 60-80 Fair to Good (Yellow), 0-60 Fair to Bad (Red)

Observations: (1) All six measurements at Test Point 110 showed significant degradation of video quality. This location was near overhead power lines, resulting in a much higher noise floor, approximately 3 dB above the TOV for RX 10. (2) Test Points with impairments occurring once and lasting 2 to 3 seconds over the 60 second measurement period were considered fair to good image

Page 27 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

quality. Four of the impairment conditions were in this category. (3) The rest of the test conditions provided an excellent video quality rating based on the DMOS and VMAF analysis standard.

Figure 18 below captures a frame of the impaired video at Test Point 95. The impairment occurred once, lasted for two seconds, and the video quality is considered ‘good-to-fair’.

Figure 18: Impairment observed at Test Point 95 on RX10

Observation: Impairment is visible on the screen at several pixels (denoted by the green arrows), but the quality is considered to be excellent based on video analysis metrics.

2.1.11.4 Impairments to the RX12 DTVR After the completion of ATSC 1.0 data collection, KTWC was configured to operate on the ATSC 3.0 standard with same transmission antenna and with an ERP of 9 W. ATSC 3.0 DTV transmission was configured to operate at 256QAM PLP modulation scheme with an average data rate of 17.9 Mbps.

Table 9 shows the impairments observed at each test point with the desired ATSC 3.0 signal.

Test Location

TVWS Signal at CPE (dBm)

TVWS Signal at DTVR (dBm)

RX12 Receiver DL UL

1 -57.9 -76.7

4 -58.8 -76.8

9 -74.2 -85.3

10 -68.1 -86.8

67 -70.1 -86

Page 28 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Test Location

TVWS Signal at CPE (dBm)

TVWS Signal at DTVR (dBm)

RX12 Receiver DL UL

68 -79.8 -86.4

77 -70.2 -80

78 -70.7 -76.3

81 -69.3 -81.8

83 -70.5 -80.7

86 -66.7 -80.8

89 -72.4 -82.8

95 -65.1 -81.2

97 -69.7 -81.2

98 -68.3 -81.9

100 -60.9 -76.5

101 -66.6 -76.9

102 -62.3 -76.7

104 -41.3 -68.4

105 -39.8 -65.5

106 -53.6 -66.2

110 -64.7 -76.9

111 -64.1 -76.7

114 -57.9 -76.6

Table 9: Impairment to the RX12 DTVR

Observations: (1) Similar TVWS measurements were observed between ATSC 1.0 and ATSC 3.0 standards validating repeatability. (2) No impairments were observed at any test location or configuration.

2.2 Breakpoint Measurement A separate experiment was conducted to identify the minimum TVWS signal level required to completely impact video transmission of DTVR. As the TVWS base station was separated by a larger distance from the DTVR, due to higher path loss and other losses it was challenging to increase the TVWS signal level from the base station at the DTVR. To overcome this limitation, (1) separation distance between TVWS CPE and DTV receiver was reduced from 10 meters to 3 meters, (2) the

Page 29 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

LPDA antenna of the CPE was replaced with an omni-directional antenna, (3) the orientation of the Omni antenna was rotated by 90º to match the polarization of the DTV antenna avoiding cross-polarization loss between antennas, (4) the TVWS CPE was placed between the link between the DTV broadcast station and the DTVR, (5) test point 105 was selected to have a clear line-of-sight (“LOS”) between the TVWS base station and the CPE antenna, and (6) the uplink traffic flow was initiated between the TVWS base station and the CPE at close to 100% duty cycle. Figure 19 shows the test set up for the breakpoint measurement.

Figure 19: Breakpoint Measurement Set up

Modification in the installation and set-up resulted in 30 dB higher TVWS signal level at the DTVR for ATSC 1.0 and ATSC 3.0 measurement. From the laboratory data, it is estimated that the breakpoints would be -40.59 dB and -46.53 dB for ATSC 1.0 receivers RX4 and RX10 respectively. With ATSC 3.0 receiver for 256QAM configuration, it is estimated that the breakpoints would be between -29.3 dB and -28.1 dB. Table 10 shows the “breakpoint” power for the RX4, RX10, and RX12 receivers. Figures 20 and 21 provide a signal trace at the breakpoint for the RX4 and RX12 DTVRs.

Page 30 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

DTV Receiver

DTV Signal

at DTVR (dBm)

TVWS at DTVR (dBm)

Visual Impact

Observed in the Field

DTV Receiver D/U as

measured (dB)

Maximum TVWS Power

on DTVR without

causing visual artifacts (dB)

Difference Between

Measured Field and Lab

Maximum TVWS Level (dB)

RX4 -68.9 -39.9 No IMPACT

-41.93 -40.5 0.6

-36.4 IMPACT -41.93 -40.5 4.1

RX10 -68.9 -48.3 No Impact -40.83 -46.7 -1.6

-45.4 IMPACT -40.83 -46.7 1.3

RX12 -65.4 -29.3 No Impact -47.87 -31.2 1.9

-65.4 -28.1 IMPACT -47.87 -31.2 3.1

Table 10: RX4 and RX10 Breakpoint Measurement

Figure 20: ATSC 1.0 Break Point Signal Trace

Page 31 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

Figure 21: ATSC3.0 Break Point Signal Trace

Observations: (1) Breakpoint data captured in the field validated the estimation from laboratory data. (2) RX10 displayed impact at 9 dB lower TVWS signal than RX4 pointing to the limitation with tuner performance, filtering, and shielding of the DTVR.

2.3 Uplink Test Using a Separation Distance of 16 Meters At Test Point 105, a link was established between the TVWS base station transmitter mounted at 30 meters AGL and the TVWS CPE. The link was configured so that over 90 percent of the data traffic was in the uplink direction from the TVWS CPE to the TVWS base station. The DTV antenna was placed in-line within the antenna’s +1 dB point, separated by 16 meters from the CPE. The measurement period for the first test was 15 minutes. The results are presented below.

Test Location

DTV Signal at DTVR (dBm)

TVWS CPE Conducted

Power (dBm)

TVWS CPE Radiated Power (dBm)

TVWS Signal at

DTV Receiver

(dBm)

TVWS Throughput

(Mbps)

Visual Impact Observed in

the Field

105 -68.7 23 33 -70.8 9.6 No impact

-68.8 23 33 -79.5 2.4 No impact

Table 11: Uplink Test – DTV Antenna Between TVWS Base Station and CPE – 16 Meters Away (RX4)

DEKRA reduced the DTV and TVWS CPE antenna separation to 10 meters. We observed random impairments. Most of these impairments were observed when the RX4 DTVR was replaced by the RX10 receiver.

Page 32 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

• Impairments were random (once every 2-3 minutes for a duration on 1-2 seconds). • There was no increase in signal strength of TVWS at the DTV receiver. • It appears most of the TVWS signal was coming from BS rather than CPE. Impairments

were mostly happening in UL traffic from CPE to BS. • It may be multipath reflections from CPE radios that were causing these impairments

randomly.

Observation: In summary, if we place the antennas of TVWS CPE and DTV not in path with each other, we are not able to see any impairments. When they are in path, we do not observe impairments when they are separated by 16 meters. When they are in path and the separation is reduced to approximately 10 meters, we observe random impairments.

Page 33 Code: 2502.Report 3-May-20 DEKRA Certification and Testing, Inc. © 2020

3. APPENDIX

3.1 Test Equipment

- Keysight /Agilent MXA 9020 Spectrum Analyzer

o https://www.keysight.com/en/pdx-x202266-pn-N9020A/mxa-signal-analyzer-10-hz-to-265-ghz?cc=US&lc=eng

- Anritsu MS2720T Spectrum Analyzer o https://www.anritsu.com/en-us/test-measurement/products/ms2720t

- AccepTv Video Quality Analyzer

o http://www.acceptv.com/en/products_vqa.php

- 6Harmonics TVWS Radios and Accessories o 4000 Series

Page 34 3-May-20Code: 2502.Report

DEKRA Certification and Testing, Inc. © 2020

Disclaimer © 2020, DEKRA Certification, Inc. Please note that use of this document is conditioned by the following:

The information in this document is subject to change by DEKRA Certification, Inc. without notice.

No warranties express or implied, are given by DEKRA Certification, Inc. All implied warranties, including implied warranties of merchantability, fitness for a particular purpose, and non-infringement are disclaimed and excluded by DEKRA Certification, Inc.

The document does not constitute an endorsement, recommendation or guarantee of any of the products (hardware or software) mentioned. The document does not guarantee that there are no errors of defects in the products or that the products will meet the reader’s expectations needs or specifications, or that they will operate without interruption.

This document does not imply any endorsement, sponsorship, affiliation, or verification by or with any organizations mentioned in this document.

Under no circumstances shall DEKRA Certification, Inc. be liable for any direct, indirect, incidental, special or consequential damages arising from any error or omission in this document.