Desperately Seeking Efficient Spectrum Utilization

1
Desperately Seeking EfficientSpectrum Utilization

• Stagg Newman
• Chief Technologist
• Federal Communications Commission
• BWRC Winter Retreat
• January 10, 2000

2
Purpose

• To explore
• the forces shaping demand for spectrum, and most
  particularly the Internet
• key policy issues facing the FCC that need
  wireless solutions
• new approaches to better spectrum utilization.
• To provoke and challenge your thinking

3
Disclaimer

• The opinions expressed in this talk are those
  of the speaker and do not represent official FCC
  policy or the views of the Commissioners.

4
Outline

• Spectrum Demand Forces the Internet
• Role of wireless in the marketplace
• Disruptive Technologies and Virtuous Cycles
• Whats different about the Internet
• Key Policy Issues Wireless
• Proposals for Better Spectrum Utilization

5
Technology Framework for Wireless SolutionsWhat
wireless is and is not good for

• Wireless is the only solution for mobile
  communication
• Wireless is a good solution for nomadic
  communicators
• Wireless is good for market entry where a
  wireline solution cannot be readily exploited
• Wireless is challenged to compete with existing
  exploitable wireline solutions

6
Network Architecture - The VisionCirca 2010

• Fiber with Wireless Tails
• if it moves its a wireless communicator
• if it does not move (buildings), get a fiber to
  it
• use fiber plus wireless for LANS
• wireless silicon cockroaches
• IP Unter Alles
• all applications run over IP
• Control plane (MPLS?) controls packet flows over
  media
• Satellite for Ubiquitous reach and price cap

7
Network Architecture - The RealityThe Mass
Market is the Critical Issue

• Picture a map of the U.S. with dots representing
  the major cities. Now connect the dots with
  fiber.
• Industry has been there, and done that
• ( and continues to do so!)
• Picture flying into the SF Bay Area at night.
  Picture each light as a building or vehicle that
  needs to be connected by fiber or broadband
  wireless respectively. Now design a cost
  effective network for 20 market share.
• Thats a challenge!

8
Outline

• Spectrum demand forces the Internet
• Role of wireless in the marketplace
• Disruptive Technologies and Virtuous Cycles
• Whats Different About the Internet
• Key Policy Issues Wireless
• Toward Better Spectrum Utilization

9
Virtuous Cycle
Processor and Memory Costs Dropping
Speed and Capacity Increasing
Applications Growing
The Computer Industry
10
NETWORK DEMANDParallels with the Computer
Industry
Free MIPS is to the Computer
Industry as Free B/W will be to the
Communications Industry
The desperate search for bandwidth is generating
billions of investment in the latest technology
that will destabilize the industry.
11
Growth of Internet Users
Number of molecules in the universe
55 million users today
9 million users in 1995
Number of humans on the planet
12
NETWORK DEMANDThe Explosion in Traffic Demand
Tomorrows Traffic
800
700
600
500
400
Network Traffic in Terabits
300
200
Todays Traffic
100
0
1995
96
97
98
99
0
1
2
3
4
2005
13
UNDERLYING TECHNOLOGYEnablers Are Here
Technology
Network Technology Architectures
Optical Networking
Service Providers
New Products
New Services
Wireless
Deregulation
Digitization
Market Forces
Exploding Demand
Internet
14
Performance Improvements

• Optical Technology - Doubling every year
• Wireless - 10x to 100x improvement in 2-5 yrs
• Chip Technology - Moores Law (2X in 18 mo)

15
Virtuous Cycle
Optoelectronics, Radio, Processing Costs Dropping
Affordable Bandwidth Increasing
Applications Growing
The Communications Industry
16
Digitization Hypotheses

• The introduction of digital technology into a
  market creates competition and obsoletes
  regulation because of
• Exponential destabilization of Price/performance
• Reduced costs of entry
• Enhanced feature differentiation
• Power of microelectronics, S/W, radio, and
  photonics come together
• Regulated Communications Regimes Are All Going
  Digital

17
Reality CheckWhy dont we have 1.5 Mbps for 20
per month?

• Infrastructure Costs and Build
• First Telegraph - Baltimore to Washington
• 10s Bps, crew of 5-10
• 1 mile per day
• 1 regulator- 45 days
• First major long haul fiber - NE Corridor system
• 100s Mps, crew of 5-10
• 1 mile per day
• 100s regulators, 4 yrs
• Operations and customer acquisition
• Regulated monopoly/oligopoly industry

18
More Reality

• Cost of urban fiber run in conduit
• United States - 400,000 per kilometer
• Japan - 1,200,000 per kilometer
• LMDS may cost 1/25 that price in Japan.
• But ROWs, line of sight, building access, are
  difficult issues.

19
More Reality for BroadbandU.S. Competitive
Disadvantages

• Wireline network cost per subscriber inversely
  proportional to density. U.S. has very low
  density.
• Wireless is an alternative BUT we have the most
  encumbered spectrum in the world.

20
Outline

• Spectrum demand forces the Internet
• Role of wireless in the marketplace
• Disruptive Technologies and Virtuous Cycles
• Whats Different About the Internet
• Key Policy Issues Wireless
• Toward Better Spectrum Utilization

21
Architecture Perspective POTS
Electronic Mail
APPLICATION
Video Server
Info Browsing
Financial Services
Voice
Analog not packet not broadband not correct
separation not driven by competition
NETWORK
Wireless
LANs
DBS
ATM
PSTN
SMDS
Frame Relay
22
Architecture Perspective ISDN
Electronic Mail
APPLICATION
Video Services
Financial Services
Info Browsing
Voice
Digital barely packet not broadband not correct
separation not driven by competition
BEARER SERVICE
NETWORK
Wireless
LANs
DBS
ATM
ISDN
SMDS
Frame Relay
23
Architecture Perspective BISDN
DATA APPLICATIONS
Video Services
Voice
Financial Services
Web Browsing
Electronic Mail
IP
Non- IP Data
ATM
Digital Packet Broadband Not correct
separation Not driven by competition
SONET
DWDM
24
Advanced Internet Architecture
APPLICATION
Electronic Mail
Video Server
Voice
Web Browsing
Financial Services
Digital Packet Broadband Correct
separation Driven by competition
IP
ADSL
Wireless
LANs
Private Lines
Satellite
ATM
SMDS
Dial-up
SONET
Frame Relay
Cable Modem Access
DWDM
25
Lessons

• For IP based networks adding new technology and
  services is easier and costs less
• Internet approach will always reach more
  customers than any other network
• Internet approach will attract more developers,
  leading to more rapid service evolution

Voice and Video over the Internet are becoming
real! Convergence is happening. Its the
Internet!!
26
Question

• Why did the Internet succeed while other open
  packet networks, particularly those developed by
  PTTs and regulated common carriers fail?

27
Possible Answers

• Worldwide deregulation/unregulation
• inexpensive private lines in U.S.
• flat rate dial access in U.S.
• World Wide Web fueled demand
• open access to the world of content
• greatly reduced cost of S/W development
• Growth of data compared to voice
• Technology that rides Moores Law
• Driven by hyper-competitive industry

28
Question re Spectrum Efficiency

• Will most communications applications using
  wireless transport migrate to IP networks?
• IP over the optical layer will be the future
  network architecture for really high speed
  networking
• Virtual private IP nets will replace todays
  private nets
• IF so, what are the implications for spectrum
  policy?

29
Outline

• Spectrum demand forces the Internet
• Key Policy Issues Wireless
• Creating a competitive broadband access market
• Connecting rural America to Info Superhighway
• Meeting the exploding need for mobile data
• Toward Better Spectrum Utilization

30
(De) Regulators Dream

• IP separates Services Layer from Digital Access
• Services Layer is fully competitive
• Customers have many choices of ISP
• Customers have many choices of Digital Access

31
Assumptions

• The Internet will supplant todays PSTN.
• Email, WWW, and Broadcast radio are here
• Video and carrier class voice are coming
• A Virtuous Cycle will drive communications.
• Market forces will drive the proper
  Interconnection and open interfaces.
• Within a few years, most customers will have
  multiple choices of broadband pipes.

32
Deregulators Dream Circa 2002Freedom of Choice
World of Digital Content
Backbone Network Providers
POTS
Digital Bit Pipes
IWU
B/C
Cable
Legacy Regs
Customers
33
????(Re)regulators Dilemma Circa 2002???Will
There be a new Broadband Monopoly?
World of Digital Content
Backbone Network Providers
For now Let market work!!! monitor
PSTN
Cable TV
Customers
34
Disclaimer Reminder

• The opinions expressed in this talk are those
  of the speaker and do not represent official FCC
  policy or the views of the Commissioners.

35
Bold Ideas for a Deregulatory Future

• Competitive Broadband access
• Interconnection through market determination
• Universal access instead of Universal Service
• Sunset legacy regulation
• (The latter three ideas will not be discussed in
  this talks. Viewgraphs are in your material.)

36
Competitive Broadband AccessFCC Actions

• Wireless Spectrum Availability
• 2-Way MMDS, Auctions, Unlicensed Rules,
  satellite. UNE Rulemakings
• Competition rulemaking
• ROWs, roof rights, etc.
• Access to MDUs, inside wiring, etc.
• Advances Services Rulemakings
• Collocation, Spectrum Mgmt, Line sharing, etc
• Forbearance and monitoring of cable

37
Broadband Wireless Internet Access

• Make enough spectrum available for competition
  and multiple solutions
• Permit flexible use of the spectrum
• Enable competing technologies
• Permit multiple, evolving standards
• Minimal number of rules
• to minimize interference
• to encourage interconnection
• to promote open interface standards

38
Interconnection

• ISPs that either self-provide or privately obtain
  their telecommunications input will not be
  subject to common carrier regulation provided
  they follow a general policy of fair and open
  interconnection. The Commission will carefully
  monitor this. Only if market forces fail to
  work for Internet interconnection, setup an
  "Interconnection Commission" as a tribunal of
  last resort to arbitrate any interconnection
  disputes through a process of binding
  arbitration.

39
Universal Access

• Universal access or universal service
• Base subsidy to people/institutions in need
• All areas should have a least one evolable
  broadband access network
• Wireless providers can receive funds via
  customers
• Work with Joint Conference to develop national
  policy with local administration

40
Sunset Legacy Regulation

• Challenge
• Develop a blueprint to sunset legacy regulation
  that encourages investment by incumbents while
  not permitting anti-competitive exploitation of
  todays dominant market position.

41
Wireless Internet AccessThe Mass Market Question

• Economics of Market Entry
• Right Coverage is key
• The Chicken and Egg problem
• 6000 ISPs threatened by ILECs and cable
• Proposed solutions no clear winners
• MMDS - cost effective coverage a question mark
• WCS - out of band limits are stringent
• UNII - trees and roof clutter are major issue
• LMDS - limited range, costly equipment
• Unlicensed - can it work for a common carrier
• Not enough clear spectrum at lower frequencies

42
Wireless Internet AccessNascent, Huge -
Innovation Needed

• Distinct markets, Different Solutions
• Mobility market - vehicle speed in lieu of bit
  speed
• Nomadic market - coverage is key
• Fixed wireless access - the ISP dilemma
• In lieu of Inside Wiring - solving the last foot
  problem

43
Wireless Internet AccessIdeal Requirements -
Mass Market

• Initially gt100Kbs, soon gt1Mbs Good building
  penetration (no IW costs)
• Can serve gt90 of potential customers
• Wireless Modem - 100 PCMIA like
• Build using existing tower infrastructure
• Engineered for IP traffic
• Competitive vis a vis xDSL, Cable
• Wide area nomadic access

44
Wireless Internet AccessEmbedded Internets
Solving the IW Problem

• Inside Wiring
• Ugly Realities create opportunity
• Vision of Embedded Internets (Wired Article)
• Computing in Peoples Hands
• Computing Embedded in Appliances
• Unlicensed Spectrum Exploitation
• Roles of 900 ISM, 2400 ISM, UNII, 60 GHz

45
Wireless Internet AccessCommission Encouragement

• CMRS Flexibility
• FWA Spectrum and New Spectrum
• MMDS (300 Mhz near 2.6)
• DEMS (400 Mhz near 24 Ghz)
• LMDS ( 1.3 Mhz near 30 Ghz
• 39 Ghz ( 1.4 Mhz near 39 Ghz)
• Unlicensed (100 Mhz near 5.7GHz, 83.5MHz -
  2.4GHz)
• Competitive Networks Proceeding
• Ultrawideband waivers and NOI
• Spectrum focus group,Technology Advisory Council

46
Wireless Internet AccessPolicy Issues

• Spectrum policy and rules were not developed with
  this market in mind
• packet world
• wideband two-way communications
• Jurisdictional Issues
• International policies
• implications of Europe/Japan planned 3G
  commitment
• technology mandates for spectrum use
• standards as trade barriers
• Universal service obligations?

47
Outline

• Spectrum demand forces the Internet
• Key Policy Issues Wireless
• Creating a competitive broadband access market
• Connecting rural US to InfoSuperhighway
• Meeting the exploding need for mobile data
• Toward Better Spectrum Utilization

48
Rural America Rapid Market Entry - The Hope

• Cost Effective Solutions for small market share
• Avoidance of Digging, etc.
• By air, avoiding the first mile bottleneck of
  cable and telco
• but the costs are real of
• spectrum acquisition
• digital backhaul
• radio site acquisition and installation, customer
  installation
• and customer acquisition and retention.

49
Cost Effective Access for Rural AmericaKey Issues

• Good New - Low utilization of spectrum
• Bad News - Low density, market size
• Issues to address
• Vendors build for world markets, applicability to
  US?
• Universal service funding, help or hindrance?
• Individual markets are small, but collectively?
• Political attention?
• Potential synergy with low market share solutions
• Build out requirements?
• Operations???

50
Cost Effective Access for Rural AmericaPolicy
Questions

• Universal service
• obligations (state funding, E911, QOS)
• eligibility for support
• Rate Rebalancing
• More spectrum flexibility for low density areas?
• Jurisdicational Issues

51
Outline

• Spectrum demand forces the Internet
• Key Policy Issues Wireless
• Creating a competitive broadband access market
• Connecting rural US to InfoSuperhighway
• Meeting the exploding need for mobile data
• Toward Better Spectrum Utilization

52
Proposed Growth for 3G
Number of molecules in the universe
Number of humans on the planet
53
Perspectives on IMT 2000/3GAre there parallels
with ISDN

• ISDN (Innovations Subscribers Didnt Need)
• ISDN (Investment Still Doing Nothing)
• Did a lengthy standards process in a fast moving
  world mean better technologies will supplant G3?
• Are there compelling new applications?
• Did compromises to solve everything mean nothing
  is solved well? In particular is 3G well suited
  for the Internet?
• Or will 3G be the foundation for wireless data?

54
Mobile dataDifficult Policy Issues

• How to resolve tension between US and EU
  positions?
• How much spectrum is enough?
• How to free currently encumbered spectrum?
• How to meet needs of commercial and private
  markets? (Will private virtual nets on
  commercial nets be the answer)?
• Where possible let the market, not regulators
  answer the questions!

55
Outline

• Spectrum demand forces the Internet
• Key Policy Issues Wireless
• Toward Better Spectrum Utilization

56
Spectrum UsageNew questions, old questions

• Three fundamental approaches to spectrum
• Service/ Use rules
• Property rights
• Use of commons with etiquette
• Commercial, government and/or shared usage
• Separating hype from reality
• different markets/different issues (radar apps,
  LANs, WANs)
• Competing technical claims, hard scientific data
  needed

57
Policies for Wireless Innovation

• Make enough spectrum available for competition
  and multiple solutions
• Permit flexible use of the spectrum
• Enable competing technologies
• Permit multiple, evolving standards
• Minimal number of rules to minimize interference,
  to encourage interconnection, to promote open
  interface standards
• Base on good science (e.g noise floor issue)
• Role for Technology Advisory Council
• Research Community

58
Why Researchers Should Care

• Regulations do prevent technology deployment.
• The spectrum and public networks are regulated.
• Regulations and policy take years to change.
• Incumbents have a lot to lose!
• Incumbents have near term PL objectives!
• Incumbents play the DC game well.
• Investors do not like regulatory uncertainty!
• Start-ups have limited runways.
• Research w/ implementation is best fulfillment!!!
• Need to pay to play!
• Shaping policy is the pay!!

59
Conclusion

• The bandwidth virtuous cycle will fuel next
  explosive growth for wireless.
• Convergence and the Internet will fundamentally
  reshape the industry.
• The chaos of the market is the best way to meet
  the thirst for bandwidth.
• OUR CHALLENGE
• Finding ways to encourage efficient spectrum
  utilization with minimal regulation

60
For More Information

• FCC Web Site
• http/www.fcc.gov