The FCC formed a Spectrum Policy Task Force (SPTF) in June 2002 to identify and evaluate changes in spectrum policy that will increase the public benefits derived from the use of radio spectrum. This was the first time that there was a comprehensive and systematic review of FCC spectrum policy.
in new spectrum access models FCC has not kept up with the pace of communication tech and accelerating demand for spectrum Current “Command + Control” policy and procedures micromanage spectrum Thus it is outmoded and obsolete Need “out-of-the-box” ways to allocate and maximize spectrum access Find ways to maximize public benefits delivered through spectrum based services and devices
Demand for Spectrum- Based Services and Devices Consistently underestimated demand 1994 projected 54M mobile phone users for 2000 Actual number of users in 2000 was 110M Unlicensed band (2.4Ghz) spurred explosion of new devices and services $2.9B IN 2002 New tech allows for devices paid for & controlled by millions of end users Old policies based on small number of licensees (broadcasters)
Advances: Enabling Changes in Spectrum Policy Digital Signal Processing allows for radical new modulation techniques Wideband Spread Spectrum Ultra-Wideband pulse Very low power per hertz Cognitive / Software Defined Radios Dynamically and Intelligently utilize and share spectrum Moore’s Law makes it practical and affordable (and inevitable)
Access: Mitigating Scarcity of Spectrum Resources Measurements of actual spectrum utilization in Metro areas showed: Nearly 100% of spectrum allocated, but only 30% actually used Looking for new ways to better utilize spectrum Underlay Spectrum Commons (UWB, Spread Spectrum) Cognitive Radios dynamically sensing and releasing spectrum Secondary Markets
vs. Frequency Time Domain Wavelength or Pulse Width duration Viewed with Oscilloscope Frequency Domain Cycles / Second Viewed with Spectrum Analyzer (Spectrograph) Time Frequency
Spread Spectrum Trades off Spectrum for power Wider spectrum produces more sharing and bandwidth Called Process Gain Can underlay legacy narrowband users Graphics from Spread spectrum communications by Jay Fitzsummons, Troy Morris and Tony Parezanovic http://murray.newcastle.edu.au/users/staff/eemf/ELEC351/ SProjects/Morris/project.htm
Sequence Spread Spectrum Pseudo-noise (PN- code) mixed with Data to produce the coded signal to modulate a carrier Looks like noise source centered around the carrier with a bandwidth of the Pseudo-noise Receiver knows PN- code to demodulate signal Graphics from Spread spectrum communications by Jay Fitzsummons, Troy Morris and Tony Parezanovic
Hopping Spread Spectrum Data is conventionally modulated on carrier SS Bandwidth is carved up to many narrow channels PN-code selects which channel is utilized as the carrier Hopping rate is in order of milliseconds / hop thus minimizing interference with legacy narrowband users Graphics from Spread spectrum communications by Jay Fitzsummons, Troy Morris and Tony Parezanovic
nano-pulses Extremely short pulses instead of carrier waves 10 - 1000 of picoseconds (trillionths of a second) wide in time 1 - 10 Gigahertz wide in frequency Picowatts of power per hertz (in the noise floor) Radios can create output signal directly with digital techniques High precision timing, but low complexity Graphics from Scientific American: Wireless Data Blaster by David G. Leeper
Modulations Many ways to modulate pulse streams No Multipath fading Main issue is precision synchronization Applications Communications Sub-centimeter positioning Thru-wall/ground radar Graphics from Scientific American: Wireless Data Blaster by David G. Leeper
Status Feb 2002: US FCC allowed limited use Less than Part 15 levels below 3.1Ghz Some restrictions on applications Several Chip Vendors Some samples XtremeSpectrum Most announcements for mid to late 2003
/ Software Defined Radios Cognitive radio “understands” local conditions and user requirements Will aggregate bands of spectrum that may be allocated but not being used locally Software Defined Radio (SDR) Radio signal modulated/demodulated in software Can create arbitrary signals Could be Spread Spectrum, UWB or traditional
of SDRs Some commercial implementations Very Limited Applications like multi-band / multi- standard cell phones Vanu Inc. SDR Software Developers Kit Gnu-Radio Open Source SDR Military most advanced DARPA NeXt Generation Communications
Networks End points can relay through other user nodes Low Power Route around obstacles Cooperation Gain Total Capacity of Mesh increases with added users / relay nodes Matches low power / high process gain tech like Spread Spectrum & UWB 0 12.5 25.0 37.5 50.0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 3.3 6.5 9.8 13.0 Mesh Network Capacity vs Station Density Total Capacity Number of Stations Per Station Capacity
of Mesh Networks Limited commercial deployment Nokia Rooftop First generation product very limited throughput, proprietary and expensive. For residential / infrastructure use only http://www.wbs.nokia.com/ Mesh Networks Inc. Initial product proprietary Promising an 802.11 based product that supports infrastructure & end user relaying http://www.meshnetworks.com/ Long history of Military development
Incredible Document from a Federal Bureaucracy 9 months from start to finish Introduces and promotes several concepts that would have been unthinkable only a year ago Not perfect, several contradictory positions Will mark a milestone for new regulatory thinking
Rights Models Command and Control Current style of regulatory policy Exclusive Use Spectrum as Private Property Spectrum Commons Technology used to share and manage spectrum
and Control Model Traditional management of spectrum for the last 80 years Government agency micromanages all spectrum allocation FCC National Telecommunications and Information Administration (NTIA) Spectrum allocated to a specific entity for a specific use and specific technology Little or no flexibility how licensees can utilize spectrum
Use Model Allocates spectrum as property Spectrum holder can do whatever they want with it Within the power and interference technical requirements Can lease/resell all or portions to create secondary markets Economists believe this will evolve spectrum to its “Highest Value”
Commons Model Unlimited unlicensed users share spectrum via technological mechanisms 802.11 WLANs proved the value Already US$2B Industry and growing rapidly Still rules and limitations on how Spectrum is used Power per hertz, freq range, geographical, etc. Marketplace of devices, services and technology
Spectrum thru Technology, Not Politics Spectrum is infinitely divisible Tech determines physical access to usable spectrum Tech possibilities are just beginning Most legacy systems are based on early 20th century technology Cell phones and 802.11 use primitive sharing Multi-dimensional real-time sharing by space, frequency, time, coding, mesh has no comparison to today’s limited capacity Data Link (MAC) Network Transport Session Presentation Application Physical Economics Politics 802.11 IP TCP/UDP 2 3 4 5 6 7 1 8 9
SPTF Commons Suggestions A Low Power Underlay Spread Spectrum, UWB, and other tech to utilize new capacity in the noise floor of legacy systems Interference Temperature defines noise floor Dynamic Reuse of idle spectrum Sense and utilize local spectrum being unused by primary, release as soon as primary uses it. Geographical / Guard Bands Max power determined by local conditions + rules set by FCC and/or primary licensee
Temperature A paradigm for assessing the interference in an environment A quantitative measurement that allows for technology based access control to spectrum Measures the RF power available at the receiving antenna per unit bandwidth.
Temperature Metric to establish maximum permissible levels of interference Characterizes the worst case environment in which a receiver would be expected to operate. Different threshold levels could be set for each band, geographic region or service,
Radio Enables Dynamic Sharing Also known as Cognitive or Software Defined Radio (SDR) Dynamically discovers + utilizes local unused spectrum in real-time Takes advantage of “White Space” in spectrum allocations Releases slices of spectrum if primary licensee starts to use it also Scales power based on application, local condition and rules set by FCC and/or Primary Licensee Combine with Spread Spectrum and UWB
Agile Use of Spectrum Channel 6 Channel 7 Channel 5 Legacy uses such as TV do not allow adjacent channels in the same geographical location due to primitive receivers Agile Radio could use low power channel 5 inside of channel 6 and 7 coverage areas Could use any channel at higher power outside of their and adjacent channels in areas where they are not allocated
Recommendations Designate additional bands for unlicensed use Pursue secondary markets for use of licensed spectrum Government granted easements to licensed spectrum to enable a commons for low power non-interfering users Promote spectrum flexibility in rural areas Promote experimental spectrum allocations
Bang Auction of Spectrum Championed by FCC Economists Evan Kwerel and John Williams Existing spectrum licensees incentivized to put “their” spectrum up for auction Not required, but if they don’t, their use of the spectrum continues to be bound by old rules Get to keep all the proceeds from their sale Government (including military) puts all its spectrum in the auction Government can “buy back” spectrum for government (military, public safety) or public “Spectrum Parks” Purchasers can aggregate spectrum Corporations or Organizations can buy spectrum for “unlicensed” uses WiFi Alliance could buy spectrum for 802.11 for instance
& Farber Proposal Place all spectrum into the market, using Big Bang Ownership model: Fee simple with non-interference easement I own the spectrum and have absolute use priority; others can use it but only if they don’t interfere with this absolute use priority UWB, agile radio, mesh networks OK; “virtual commons” Monitoring and enforcement = transactions costs Gov’t (at all levels) and private groups can own spectrum and make it available for commons use: “spectrum parks”
with Faulhaber & Farber Proposal Better than a pure Private Property Model Still assumes spectrum as private property should be the dominant model Does not show that the benefits outweigh its costs Treats the Commons as a hedge Still constrains Commons opportunity Promotes permanent grant of private property Provides no revisability when technology or applications change Not enough information to make such final and irrevocable decision
based Commons Advocated by many technologists and consumer advocates David Reed (MIT), Yochai Benkler (NYU School of Law), Lawrence Lessig (Stanford) Dewayne Hendricks (Dandin Group) among others New Tech utilizes spectrum more efficiently Works best with large swaths of spectrum Underlays & Agile radios can allow for transition from legacy Industry Standards and technology manage most sharing issues
Association of Broadcasters (NAB) Mostly concerned with protecting their existing “rights” Maintain that they are defenders of public interest with “free” Television Against auctions because they undervalue the public interest benefits Claim they are already “setting new standards in spectral efficiency” Against commons for fear of interference with old radios and TVs Consider the Cellular industry their biggest threat in terms of spectrum
Telecommunications & Internet Association (CTIA) Represents the Cellular / Mobile Phone Industry Aggressively supports spectrum policy reform Cellular industry wants more spectrum for more capacity Against “giving non-viable incumbents flexibility to provide any service” I.E. allowing TV stations to compete with Mobile Phone Companies Against underlay in already allocated bands Ok for new licenses where it can be explicitly stated
Go slow, mostly licensed, some unlicensed ok Offered two technical papers Good analysis of mesh networks Raises questions on ability of cognitive radios to release spectrum fast enough Still seems more FUD to slow down new tech Supports R&D in mesh and cognitive radios But not immediate roll out or licensing Supports more unlicensed spectrum Some in 5Ghz, but most in 10Ghz and above
Broadcasting & Communications Association Protect DBS receivers from Terrestrial Interference Concerned with sharing spectrum with terrestrial Multi-channel Video Distribution and Data Service (MVDDS) Against underlays and unlicensed spectrum Current DBS receivers are easily interfered with GPS is very sensitive to interference Fear of near channel overlap on satellite radio
Association of Amateur Radio (ARRL) Amateur Operators use to be radio innovators Since the IC and Digital revolutions Amateurs have not been as involved Now are mostly legacy users FCC should use the SPTF for planning No Big Bang / privatization of spectrum Unless Amateur’s get their own “Public Park” Concerns of unlicensed spectrum Proper policing of power and other tech constraints Supports the need for regulation of receivers
Federation of America Pro-consumer advocacy organization Considers spectrum to be a “First Amendment (Freedom of Speech) asset of citizens Privatization of spectrum would limit free speech Selling of spectrum would accelerate consolidation of media Supports Spectrum Commons
Unlicensed Spectrum will Unleash Broadband Unlicensed wireless can break the broadband bottleneck Allows the Internet to “route around” incumbents Telcos who are slow to build broadband Allows end users to finance broadband builds Supports Spectrum Commons and additional unlicensed spectrum Believe there should be “rules of the road” to facilitate sharing of spectrum
Unlicensed Spectrum for the Network Revolution Similar to Microsoft’s points Some additional points of Cisco: More commons, less private spectrum ownership Don’t get stuck on international spectrum harmonization
UWB; new understanding of Interference Manufacturer of UWB technology Current vague definition of interference be replaced by explicit definition The new “Interference Temperature is a good start. Specify minimal capabilities of receivers to reject interference Translate that into max permissible emissions levels for underlay technology
Recommendations Avoid irrevocable and difficult to change policies Privatizing large amounts of spectrum would be difficult to reverse Technology is just beginning Enable some form of “easements” on existing and new licenses to allow for underlays and agile radios. Be explicit with legacy incumbent licenses Much of the issues of the report were concerning incumbents, but were not openly discussed as such Treat incumbents as a transition issue, not assume that they should be incumbents forever
Checks Spectrum Utilization Audits See how spectrum is really being used in various regions Analyze capacity of spectrum Consider several dense usage scenarios using data from the spectrum audits and demand growth profiles Calculate various spectrum utilizations with different technology assumptions See how often there is really a “tragedy of the commons”
to develop Open Spectrum Technologies Open Spectrum will be a boon for hardware device manufacturers Japan could lead in Open Spectrum device technology A consortium to develop core technologies could be highly leveraged Radio Haven in a secondary market would be an excellent test environment
Major Fork in the Road FCC & Industry split Economists & Incumbents like Property Models Technologists & Internet types like Commons Surprising support for commons though Boxer/Allen Senate Bill Pronouncements from Chairman Powell considering making unused TV bands unlicensed