technology leaders forum - IEEE Xplore

LYT-TECH LEADERS-Moyer

10/20/10

4:42 PM

Page 18

TECHNOLOGY LEADERS FORUM R&D FOR BROADBAND COMMUNICATIONS: A PROPOSAL TO REVITALIZE BROADBAND R&D IN THE UNITED STATES ADAM T. DROBOT, 2M COMPANIES, INC. DIANE E DUFFY AND STAN MOYER, TELCORDIA

INTRODUCTION There is widespread recognition throughout the world of the critical role that information and communications technology (ICT) and broadband networking play in driving economic activity. ICT enables cost-effective support of economic sectors, from education, healthcare, transportation, public safety, and energy to commerce, manufacturing, government, entertainment, science, and engineering. Efforts to develop policies to spur broadband deployment and use are actively underway in many countries.1 Political, economic, regulatory, and social regimes vary widely across the world, and each nation must develop its own approach to cost-effectively providing broadband capability that works within their own framework. These approaches must be supported by appropriate levels of investment, regulation, and policy, including R&D. While this column provides specific recommendations for U.S. broadband policies, many of the suggestions and insights are applicable more broadly as countries around the world develop and evolve their broadband policies. The United States has historically played a lead role in the development and adoption of telecommunications, computing, and networking. A broadband infrastructure both improves productivity and drives innovation. Ensuring global competitiveness in the 21st century requires a broadband infrastructure that is secure, scalable, evolvable, trusted, robust, interoperable, affordable, inclusive, fault-tolerant, easy to use, and easy to maintain. The underlying technological capabilities that drive ICT and broadband are evolving rapidly. At the same time, regulatory and policy regimes follow a different timescale, creating significant challenges to achieving highspeed nationwide broadband. To maintain a position of leadership in critical ICT technologies and to implement the roadmap of the U.S. National Broadband Plan,2 the U.S. must develop policies and programs, which we argue should include vigorous funding of R&D. Meeting these challenges requires a balanced, coordinated and strategically directed R&D investment of adequate size and scope. To achieve value and avoid bottlenecks, innovations must proceed in a balanced fashion: • Across the key dimensions of capital expenditures (CapEx), operating expenditures (OpEx), services, and adoption so that underlying network capability is available to deliver new services and new adopters can be supported. • Across the services value chain, which includes the totali1

See, for example, the Berkman Center’s Broadband Study, which was developed for the Federal Communications Commission (FCC) in October 2009 and titled “Next Generation Connectivity: A Review of Broadband Internet Transitions and Policy from Around the World.” This comprehensive study surveys policies and practices from many different countries around the globe and includes in-depth overviews of Denmark, France, Japan, South Korea, The Netherlands, Sweden, and Switzerland; http://www.fcc.gov/stage/pdf/Berkman_Center_Broadband_Study_13Oct09.pdf

ty of organizations and market segments that design, develop, deploy, and operate information networking technology: from physical components, devices, and equipment; to network systems, services, management, and operations; to content, media, storage, and applications; and to management and assurance of content source, identity, and security. • Across the life cycle phases of R&D from basic and applied research to development, deployment, systems management, and operational capability so that the nation reaps the rewards of research by getting better and more affordable services in the hands of a wider range of citizens and businesses, regardless of their ability, geography, or industry. • Across the short, medium, and long term so that we solve not only the problems of today, but also lay the groundwork for breakthrough innovations in the longer term. There are a number of specific challenges in the current environment for initiating and sustaining long-term highrisk research projects with uncertain but promising — and perhaps revolutionary — potential. To target the most important problems and opportunities, and to achieve rapid and successful technology transfer and commercialization, leadership, coordination, and a healthy level of competition are required among R&D funders and performers, both public and private. Stronger and more strategic leadership is necessary to avoid fragmentation, duplication of effort, and a preponderance of incremental research. Given the dynamic, complex, and heterogeneous structure of the broadband industry, coordination is needed to accelerate the impact of R&D by facilitating technology and knowledge transfer within the broadband industry and related industries. Technology transfer and commercialization are not straightforward one-way processes of ensuring that downstream organizations can implement and deploy new technologies; rather, they are multidirectional activities which also guarantee that real-world knowledge informs all types of research, from basic to applied. We propose a new entity for broadband and information networking R&D to help ensure the balanced, coordinated, and strategically directed R&D investment described above. This entity would provide leadership and coordination on a competitive basis among the following key constituencies: • Businesses, from large well established firms to small startups, in all areas of the industry • Universities, colleges, and other educational institutions • Standards organizations and industry groups • Government — including federal, regional, state, local, and municipal • Foundations and other non-government organizations (NGOs) active in ICT and digital inclusion • Healthcare, transportation, public safety, energy, commerce, manufacturing, entertainment, and other closely related industries In a competitive world, where simple measures can be misleading but nevertheless have impact, the United States is not perceived as a leader of broadband by several popular mea-

2

Preview the Plan, FCC, http://blog.broadband.gov/?category=Preview percent20the percent20Plan

18

0163-6804/10/$25.00 © 2010 IEEE

(Continued on page 20)

IEEE Communications Magazine • November 2010


10/20/10

4:42 PM

Page 20

TECHNOLOGY LEADERS FORUM (Continued from page 18) sures.3, 4 We have also seen a considerable migration of the manufacturing base, talent, and innovation to other regions of the world. While migration creates diversity and is an engine for advancing the world economy, increased funding and investment are still needed for the United States to be a global leader in broadband and reap the resulting benefits. Specifically, we recommend that the United States: • Increase the amount of federal funding for ICT and broadband, which includes networking and information technology R&D5 • Expand the R&D tax credit to a multiyear program with provisions for a graduated credit, which provides incentive for companies to increase both the size of their total R&D effort in the United States and the fraction of the effort that addresses basic research and high-risk projects6 • Create and federally fund an entity for broadband and information networking R&D which will oversee and manage the research portfolio • Direct investment into the critical and “gap” areas • Invest strategically to achieve breakthroughs in functionality, cost, usability, scale, and security • Encourage coordination of efforts across the public and private sector and across ICT and key industries • Support science and engineering to develop the human resources to create innovations

CURRENT BROADBAND RESEARCH ENVIRONMENT The U.S. National Broadband Plan will provide a roadmap to building the telecommunications infrastructure needed for America’s competitiveness in the 21st century. Broadband communications and information technology underpin all sectors of the economy and their advancement is critical to national growth and productivity. To paraphrase a recent report by the National Academy of Engineering, 7 ICT “has transformed, and continues to transform, all aspects of our lives: commerce, education, employment, health care, manufacturing, government, national security, communications, entertainment, science, and engineering. [ICT] also helps drive the economy — both directly (the [ICT] sector itself) and indirectly (other sectors that are powered by advances in 3

“Strategy Analytics: US Ranks 20th in Global Broadband Household Penetration,” http://www.strategyanalytics.com/default.aspx?mod=PressReleaseViewer&a0=4748, 18 June 2009. 4

OECD Broadband Portal, http://www.oecd.org/document/54/0,3343,en_2649_34225_38690102_1_1 _1_1,00.html, 19 Jan. 2010

[ICT]).” The underlying capabilities that comprise the ICT industry and our broadband infrastructure are evolving at a very fast rate, creating a number of significant challenges that must be met to cost-effectively realize the U.S. National Broadband Plan roadmap. • The demand for broadband is exploding with an annual growth rate for IP traffic of 40–50 percent.8 Meeting this demand requires innovation in the fundamental physical layer technologies for broadband media to achieve capacity increases, while keeping infrastructure costs the same or lower. R&D investment is needed from basic hardware to operational capability to keep pace with the evolving cost-capacity horizon. • Broadband systems, networks, and services are growing rapidly in complexity and scale. Affordably designing, installing, operating, interworking, maintaining, repairing, upgrading, and evolving these systems is a growing problem. R&D investment is needed to develop architectures and systems that are adaptable, flexible, scalable, and evolvable more automatically and at lower costs. • Network security, reliability, threat, and trust models have not kept pace with the evolution of infrastructure functionality, adopters’ concerns, or the capabilities of hackers and criminals. Security and reliability are currently addressed reactively, in a piecemeal fashion, and through a seemingly endless series of patches and updates. R&D investment is needed to develop and implement resilient architectures and security mechanisms that are versatile and proactive to threats. • The digital divide has left many Americans disenfranchised, particularly the elderly, disabled, those with lower income, and non-native speakers of English. Broadband applications, interfaces, and devices are often designed with early adopters in mind. R&D investment is needed to develop new paradigms for usability and to lower costs for universal design. Fundamentally, R&D aims to do things “faster, cheaper, better” through the creation and improvement of concepts, techniques, processes, methodologies, tools, services, and products. It is critical that advancements be balanced to avoid roadblocks and ensure that the benefits of the innovations can be achieved. One dimension that requires balance is the four areas depicted in Figure 1: CapEx, OpEx, services/capabilities, and adoption. All four areas must advance evenly as they are interdependent. Balance is also needed across entities in the broadband technology provider supply chain, depicted in Fig. 2. The number of technology organizations involved in creating, deploying, and operating the broadband infrastructure and providing services and information is vast. Not all organizations have the same resources, incentives, and capabilities for innovation. As the FCC noted in an Interim Report, “Rapid innovation in some sectors must not be limited by bottlenecks (Continued on page 22)

5

http://www.nitrd.gov/pubs/2010supplement/FY10Supp-FINALFormatWeb.pdf, the 2009 NITRD budget was $3.9B, which is only a few percent of the total 2009 federal R&D budget. 6

See this web memo from The Information Technology and Innovation Fund for more information (http://www.itif.org/index.php?id=319). 7

National Academy of Engineering, Assessing the Impacts of Changes in the IT R&D Ecosystem: Retaining Leadership in an Increasingly Global Environment, 2009.

20

8

See the “ATLAS Internet Observatory 2009 Annual Report” by Labovitz, Iekel-Johnson, McPherson, Oberheide, Jahanian, and Karir at http://www.nanog.org/meetings/nanog47/presentations/Monday/Labovitz_Obse rveReport_N47_Mon.pdf, pg. 29, which shows annual growth rates from four sources ranging from 35 percent to 60 percent. See also Cisco’s white paper at http://www.cisco.com/en/US/solutions/collateral/ns341/ns525/ns537/ns705/ns8 27/VNI_Hyperconnectivity_WP.html, which predicts that Internet traffic will grow at a compound annual growth rate of 40 percent from 2008 to 2013.



10/20/10

4:42 PM

Page 22

TECHNOLOGY LEADERS FORUM

CapEx

OpEx

Services

Adoption

• Lower production costs for network equipment and devices • Decrease time frame and expense of installation • Lengthen equipment replacement cycles • Improve fundamental capacity to meet core network demands

• More adaptable systems to changes in conditions • Improved security and privacy protection requiring less operator intervention • Autonomic operation for “lights-out” operation • More automated monitoring and alarming

• Service creation platforms enabling rapid development and deployment of new services • More flexible services enabling longer service life • Service portability (across networks and providers) • Better personalized services improving efficiency

• Improve ease-of-use to decrease time and expense of training • Expand usability of services and devices to support disabled users • Services aware of user “context” requiring less input from users • Machine-to-machine communications

Figure 1. Four areas where innovation and advancement is required to achieve the national broadband plan — note that the list of items is only representative of the types of innovation and advancement needed.

FUNDING, LEADERSHIP, AND COORDINATION

(Continued from page 20) in others.”9 Broadband infrastructure is fundamentally complex, and it is this complexity and the many interdependencies that create the challenge of advancing the broadband infrastructure. To ensure that inventions are successfully deployed, balance must also be ensured across the product and services lifecycle. The U.S. Department of Defense recognizes this need and categorizes R&D activities into seven phases 10 shown in Fig. 3. Without appropriate investment in the latter phases, inventions, whether in end user devices, networking, or applications, may fail to achieve widespread deployment due to excessive cost and difficulty in management and operations. While the amount of resources required in each phase increases as you move from left to right in the diagram, successful outcomes will not occur without proper investment in the early phases. In addition to ensuring that each phase receives adequate investment, it is necessary to address interconnections between phases and to implement feedback mechanisms. For example, if technology leapfrogs from advanced component development and prototypes directly to operational systems development, issues related to scalability, performance, and efficient management may not be adequately addressed. To deliver solutions to current near-term problems and create breakthrough discoveries, the broadband R&D effort should comprise a balanced portfolio of projects that range from longer-term higher-risk projects to near-term incremental projects. The research must also range from core technology areas such as security architectures and advanced wireless to new and emerging topics. The portfolio should be regularly reviewed and rebalanced to ensure that it contains the proper mix of low-risk projects and those projects that may deliver unexpected and/or major discoveries.

Given the criticality of broadband information networking to all facets of the economy and society, the total level of investment must be adequate to support and maintain a vibrant and worldleading sector. The United States invests a substantially smaller percentage of resources in telecommunications and networking R&D compared to other parts of the world. For example, the European Union in its 7th Framework Program made the ICT theme its single largest with over €9 billion in funding.11 Current U.S. investment levels in ICT and related content and information storage and retrieval technologies are moderate at best. ICT investment relative to the national GDP has remained flat since 2002, and venture capital investments in the communications technology sector have declined in the last two years12 (Fig. 4). Broadband is a competitive global market, and regaining a leadership position in this industry requires that it receive higher priority and a larger quantity of federal investment funds. Long-term investment has played a role in some of the highest performing (in terms of broadband metrics) countries. For example, “leaders in fiber deployment — South Korea, Japan, and Sweden — are also the leading examples of large, long term capital investments through expenditures, tax breaks, and low cost loans that helped deployment in those countries.”13 In addition to increasing the federal budget for broadband R&D, we strongly support tax breaks in the form of an expansion of the R&D tax credit. The R&D tax credit is a proven mechanism for spurring discovery via industry investment and, as Adrian Slywotzky wrote in a Business Week article, “Discovery drives innovation, innovation drives productivity, (and) productivity drives economic growth.” 14 Specifically, we (Continued on page 24) 11

See http://cordis.europa.eu/fp7/ict/.

12 9

FCC, September Commission Meeting, http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-293742A1.pdf, Sept. 29, 2009, p. 136.

FCC, September Commission Meeting, http://hraunfoss.fcc.gov/edocs_public/attachmatch/DOC-293742A1.pdf, Sept. 29, 2009. 13 Berkman

Center Broadband Study, http://www.fcc.gov/stage/pdf/Berkman_Center_Broadband_Study_13Oct09.pdf.

10 Beyond

Sputnik: U.S. Science Policy in the Twenty-First Century by Homer A. Neal, Tobin L. Smith, Jennifer B. McCormick, University of Michigan Press, 2008.

22

14

Business Week, Aug. 27, 2009, “Where Have You Gone Bell Labs?” by A. Slywotzky.



10/20/10

4:42 PM

Page 24


Figure 2. Influencers involved in broadband, particularly entities in the broadband technology provider supply chain.

(Continued from page 22) believe the R&D tax credit should be enhanced as follows: • Multiyear program: Without the certainty of multiyear tax benefits, it is difficult for industry to embark on projects with a longer time horizon. • Graduated: Increase the tax credit for companies that invest a higher fraction of their resources in R&D. • Preferential: Build in a preference for more exploratory (early phase) R&D targeted at addressing specific industry needs that also include effective technology transition paths and for domestically sourced R&D. We are all aware that the United States does not lead globally in advanced telecommunications and information networking. As the Knight Center for Digital Excellence notes, “Our European and Asian counterparts have been investing in broadband robust enough to support new products and services, creating a test bed for innovation. This strategy is fostering and attracting companies seeking the 21st century infrastructure lacking in the U.S.” 15 In many respects the United States has not had a leadership position since the industry restructuring that led to the dismemberment of AT&T’s Bell Labs. The Bell Labs era was remarkable for the top-notch, large, and tightly knit research and engineering organization that developed, deployed, and operated vertically integrated telecommunications networks (and won six Nobel Prizes along the way). Bell Labs, like RCA Labs (now Sarnoff) and Xerox PARC, was a single organizations with researchers and engineers delivering a balanced R&D effort that spanned all the areas and phases of R&D, from basic science to operational systems development, in an integrated fashion. This breadth and balance of R&D — from basic to operational — is the key construct that needs to be addressed in our policies now. Today’s information networking industry is dynamic,

complex, varied, and dramatically-changed from the days of Bell Labs. Innovations and innovators are found around the globe, and re-creating a single monolithic R&D entity for telecommunications, whether public or private, is neither logical nor advisable for meeting today’s needs. A Business Week article titled “Where Have You Gone Bell Labs?” noted the following about the state of the U.S. technology industry: “The venture-capital and initial public offering components… are still in place; we just have to rebuild the upstream labs that focused on basic research, the headwaters for the whole innovation ecosystem.” The article goes on to say, “Today’s challenges require the government to unleash a series of highly focused, aggressively managed projects supported by a growing research investment in a dozen or more leading companies that in the aggregate reproduce the cumulative impact of Bell Labs, RCA Labs, Xerox PARC, and others.” 16 We agree strongly with these remarks, and believe that a combination of increased federal funding and an expanded R&D tax credit will provide a foundation for reinvigorating upstream basic and applied research. Since the burst of the telecommunications bubble in 2001, the telecommunications sector has underinvested in this type of research. We also believe a new organization is needed to provide coordination, direction, and oversight for R&D in broadband and information networking. This organization — which we call an entity for broadband and information networking R&D simply for purposes of discussion — could be jointly federally and commercially funded, and chartered with maintaining the nation’s competitiveness and productivity in broadband by managing and coordinating the broadband research portfolio. The entity would not have its own laboratories nor perform research projects, but would oversee the nation’s investment in broadband R&D by directing (Continued on page 26)

15

Doug Adams, “It’s ‘patent-ly’ obvious: U.S. needs better broadband,” Knight Center of Digital Excellence, http://www.knightcenter.org/webapparticle-action/do.articleDetail/id.720/topicID.31, Sept. 28, 2009.

24

16

Business Week, Aug. 27, 2009, “Where Have You Gone Bell Labs?” by A. Slywotzky.



10/20/10

4:42 PM

Page 26


Feedback

Basic research

Applied research

Advanced System Advanced component development Management Operational technology development systems and support development and development demonstration prototypes People/Ideas

Figure 3. R&D progresses (from left-to-right in the diagram) though seven phases in order to mature basic research into real-world deployable systems. (Continued from page 24) funding to private and public institutions and collaborative teams in a competitive manner based on the national broadband roadmap. This entity would be ideally suited for ensuring that R&D for broadband communications is properly balanced across the many dimensions described previously in this paper. When R&D falls out of balance, the entity should recommend appropriate measures for restoring balance through some of the mechanisms described in this column (e.g., tax credits, multiyear commitments, federal R&D funding). Current drivers and incentives for both industrial and university researchers can favor less risky projects of an evolutionary nature, as opposed to potentially ground-breaking projects. Proactive leadership can limit fragmentation, set visionary goals, ensure coverage of all areas and phases of the R&D spectrum, and actively guide funding into important areas. With the wide variety and diversity of organizations involved, leadership and coordination are required for directing investments into promising areas, and for stimulating interactions and synergies to accelerate commercialization. The need for coordination among federal government funders of networking research has already been recognized and is currently addressed by the Federal Networking and Information Technology Research and Development Program.17 This program is but a small step in the right direction, as much wider coordination and more proactive leadership and direction are needed among all of the critical constituencies, as shown in Fig. 5. The proposed coordination entity should provide strong and active leadership to define important research directions. Additionally, the entity should provide support for program implementation, contracting, and administration and address roadblocks to full and open collaboration and knowledge sharing. One of the difficulties that can hamper research work is a lack of data and information on which to test new ideas, and identify and modify the most promising concepts. Developing a workable contracting framework, including confidentiality, intellectual property protection, and limitation of liability, is necessary. For example, organizations may need to share information about security threats, attacks, and possible response strategies without either increased risk of lawsuits or exposing vulnerabilities to criminal elements. The entity 17

See http://www.nitrd.gov/about/about_nitrd.aspx.

26

should have a modest-sized permanent staff and a body of rotating personnel on limited term assignments from constituent and stakeholder organizations. The rotating staff model is particularly important for effecting technology transfer, which is, as the oft-quoted phrase says, “a contact sport.” In fact, one key industrial R&D executive went on to say, “It (technology transfer) depends upon collaboration, teamwork, and interpersonal networking much more than fundamental research does.”18 The entity should be directed by a high-level expert advisory council along the lines of the national technology council recommended by the Telecommunications Industry Association (TIA). 19 This “Broadband Technology Council” would consist of senior executives and scientists, and would provide direction to “define and guide strategic areas in [broadband technologies] that require further research critical to the future growth of the U.S. economy. Such a Council should include representatives from different sectors, such as government, academia, and industry.” This Council would also serve as a forum for communication and the dissemination of ideas, information, developments, and solutions. The entity and Council should establish high-level linkages to related industries, standards bodies, and other international organizations. Broadband is, by its very nature, an enabling infrastructure that benefits smart grids, e-health, education, law enforcement, intelligent transportation, and many other industries. To fully realize the benefits of a high-performance communications infrastructure for these verticals, collaborative research programs should be funded to bring together diverse participants in ICT and other industries to address critical challenges through precompetitive research projects. Providing strong connections at a senior level among these industries will help avoid bottlenecks and ensure that advancements in one industry spur those in others. Government support for precompetitive collaborative R&D offers the (Continued on page 28)

18

Quote from Dr. Jim Mitchell, Director of Sun Labs; see http://research.sun.com/features/tenyears/F3ttJB.html. 19

TIA, “Investing in Telecom for Tomorrow’s Innovations: The Case for Increased Telecommunications Research Funding,” http://futureofinnovation.org/PDF/TIA percent20Telecom percent20Research percent20Funding percent20Final.pdf



10/20/10

4:42 PM

Page 28

TECHNOLOGY LEADERS FORUM 2.0%

ICT R&D (% of GDP)

1.8% 1.6%

U.S. venture capital investments Millions of dollars

Private sector R&D (% of GDP)

1.4%

700 600

1.2% Millions

1.0% 0.8% 0.6%

500 400 300

Telecommunications

Networking and equipment

2Q 09

1Q 09

4Q 08

3Q 08

2Q 08

1Q 08

4Q 07

07 20

06 20

05 20

04 20

03 20

02 20

01 20

3Q 07

0

0.0%

1Q 07

100

0.2%

2Q 07

200

0.4%

Semiconductors

Figure 4. Information and communications technology R&D has remained flat and is modestly less than in 2002 and U.S. venture capital investments across communications sectors has declined. (Continued from page 26) advantages of reducing replication and promoting interoperability. Appropriate areas for this work include: • Addressing wide-ranging challenges that impact multiple industries and sectors • Advancing the industry in ways that provide significant societal benefit but minimal market benefit • Performing the sometimes unglamorous research necessary for ensuring scalability and end-to-end performance • Reducing costs and sharing risks for expensive high-risk R&D ventures

SUMMARY Successful implementation of the National Broadband Plan is essential for driving the U.S. economy and society, and for meeting the national challenges. It is a critical enabling infrastructure for other industries such as healthcare, energy, and transportation, and in fact is a key component of the value chain across all sectors. Improving the cost effectiveness and impact of ICT will enable a higher-performance nation and boost the overall economy. As this column has emphasized, research, development, and innovation must be supported across the entire technology supply chain from devices, components, and systems to networking, services, applications, content, and media. R&D investment must also span all phases of the product and services lifecycle to avoid bottlenecks and achieve technology transition success. Furthermore, these advancements must be balanced; for example, as new networking capabilities are deployed, we must also develop new and enhanced policies for ensuring their security, privacy, and trust in order to realize the value and benefits. We recommend the creation of an objective, trusted organization to provide coordination, management, and oversight

28

Commercial R&D Government R&D organizations at all levels

Start-ups and VC-funded organizations

University R&D

R&D coordination and leadership

Foundations, NGOs

Standards and industry groups

Related industries (energy, transportation, etc.)

Figure 5. R&D coordination and leadership is necessary to maximize the output of all organizations involved in R&D for broadband information networking.

for broadband R&D investment funding. With R&D investment directed to key gaps, and the development of public-private partnerships and cross-sector coordination through an entity for broadband and information networking R&D, we can ensure improved international competitiveness and productivity improvements for all citizens, sectors, and organizations, public and private.
