Spectrum pooling for next generation public safety radio systems

2008 IEEE Symposium on New Frontiers in Dynamic Spectrum Access Networks, DySPAN 2008

Volumn , Issue , 2008, Pages 639-661

  Lehr, William ^a   Jesuale, Nancy ^b  

^a MASSACHUSETTS INSTITUTE OF TECHNOLOGY   (United States)

^b NetCity Inc   (United States)

Author keywords

Cognitive radio; Economics; Policy; Public safety; Spectrum management

Indexed keywords

ANALOG DIFFERENTIAL ANALYZERS; NETWORK MANAGEMENT; RADIO SYSTEMS; SPECTRUM ANALYSIS; WIRELESS NETWORKS;

ADAPTIVE PRIORITIZATION; COGNITIVE RADIO; INFRA STRUCTURES; LEGACY TECHNOLOGIES; MARKET CHALLENGES; POLICY; PUBLIC SAFETY; RESEARCH AND DEVELOPMENTS; RF SPECTRUMS; SPECTRUM MANAGEMENT; SPECTRUM POOLING; SPECTRUM POOLS; WIRELESS SERVICES;

RADIO;

EID: 57849140777     PISSN: None     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/DYSPAN.2008.72     Document Type: Conference Paper

References (56)

1. For example, Bluetooth or UWB to connect stereo components or personal biometric monitoring sensors mounted on a patient or first-responders hazmat suit.
2. Wireless technologies for different ranges (from a few inches to thousands of miles) and use environments have very different requirements that give rise to specialized and different technologies. While technologies designed for one purpose may often be used for another (e.g., VoIP over WiFi), there is no single technology/architecture that is best in all situations.
3. See Section 8, SAFECOM, "Public Safety Statement of Requirements for Communications & Interoperability," U.S. Department of Homeland Security, Volume I, Version 1.2 (October 2006) ("SAFECOM SoR") (available at: http://www.safecomprogram.gov/NR/rdonlyres/8930E37C-C672-48BA- 8C1B-83784D855C1E/0/SoR1_v12_10182006.pdf).
4. All users will be heavy users some of the time (when they use resource-intensive applications), and there will be more users who are heavy users more of the time (and those who are light users more of the time). Examples of all such profiles/users are easy to suggest.
5. Although in some cases, wired services may be viewed as substitutes for wireless, overall, we expect the impact of growing wireless and wired services to be complementary. For example, consider how WiFi routers helped increase aggregate demand for DSL/cable modem services by making such services more valuable. And, consider how the expansion of fiber toward the edges of wired networks expand the capabilities of high-bandwidth, shorter-range wireless services.
6. Federated networking refers to the ability to traverse heterogeneous network architectures owned by others. Ad-hoc networking refers to the ability for devices to communicate without network infrastructure. Mesh networking is a specific type of ad-hoc networking wherein devices form networks by routing traffic from and to other devices nearby.
7. We refer to the recent award of the 700 MHz public safety broadband spectrum license to the Public Safety Spectrum Trust (PSST). This policy is discussed more fully in subsequent sections of this paper.
8. At this early stage of development, DSA/CR technology solutions are likely to be more expensive than legacy solutions when used for legacy applications, at least in the near term.
9. A key difference between unlicensed spectrum access and spectrum pooling is that the former does not prioritize or limit user access. With unlicensed spectrum, any device that complies with the technical access requirements may use the spectrum, and users are not protected from interference due to congestion or from other compliant uses. With spectrum pooling, the range of users may be restricted and priorities granted to enable interference protection. Assurances of such protection are likely to be essential in gaining public safety user trust and acceptance of the pooling concept, as we discuss further below.
10. This is another reason why traditional worst-case provisioning is cost prohibitive in the public safety future.
11. Oversight of public funding and non-profit status impose bureaucratic constraints on expenditures and budgets that make it difficult for public safety network operators to rapidly scale or adjust their capacity, and typically also impose tight constraints on overall spending.
12. This includes voice conferencing in infrastructureless environments (underground, remote areas, where traditional infrastructure has been destroyed) and noisy environments (high interference).
13. http://www.psst.org/publicsafetynetwork.jsp, last accessed 7-14-08
14. Radios will need to be dynamically reconfigurable spatially to work within buildings, on-site, and in conjunction with wider-area coverage systems. Radios will need to support out-of-home area roaming.
15. Radios will need to be dynamically flexible over varying time-scales, including being able to address congestion issues in real-time during an event and at longer time scales covering equipment reconfiguration (setting up for an event) and investment (upgrading and expanding system capabilities).
16. Radios will need to be capable of adapting to special circumstances such as destruction/failure of existing infrastructure and special needs for interoperability (e.g., using commercial facilities). Longer-term, this implies graceful scaling of infrastructure.
17. SAFECOM is program of Department of Homeland Security focused on interoperability issues among state, local, and Federal public safety users. See Statement of Requirements for Next Generation Public Safety Wireless Communications & Interoperability, the SAFECOM Program, Department of Homeland Security, Version 1.0, March 10, 2004 (available from: http://www.safecomprogram.gov/NR/rdonlyres/3FFFBFBA-DC53-440E-B2EF-ABD391F13075/ 0/SAFECOM_Statement_of_Requirements_v1.pdf).
18. Project MESA (Mobility for Emergency and Safety Applications) is an international collaborative effort to coordinate the development of next-generation mobile wireless data systems (see http://www.projectmesa.org/ MESA_SoR/mesa_sor_executive_summary.pdf).
19. st Century Report").
20. These are the rules advanced by the Public Safety Trust for potential D-block bidders (see http://www.psst.org/documents/BID2_0.pdf).
21. The NRF and NIMS are discussed further below.
22. Since RPC's are composed of "all" public safety users in a region, and must reach agreement by consensus, they can and do adopt policies which regulate the priority of uses for spectrum. These policies are generally followed by mutual agreement.
23. The hyper-fragmentation of the LMR market produces small levels of buyer demand for very specialized equipment. A 20-channel digital trunked radio system in any band with three transmit towers and three repeater towers (typical mid-size city configuration) will cost around $20 Million. Portable radios will cost $3,000 to $6,000 each, depending on features - or 1000 times more than an i-Phone!
24. Quoted from Jesuale, Nancy and Bernie Eydt (2007), "A Policy Proposal to Enable Cognitive Radio for Public Safety and Industry in the Land Mobile Bands," in IEEE International Symposium on New Frontiers Dynamic Spectrum Access Networks 2007 (DySPAN2007), April 2007.
25. Essentially, spectrum portability is very similar to "roaming" and involves the ability to use spectrum access rights of the user on infrastructure owned and operated by another party. Under the current architectures and regimes, this is impossible. I cannot go to another state or another network, and have my public safety radio negotiate access to a channel in spectrum I am licensed to use for use on another licensee's network. Under a DSA paradigm, spectrum portability will be common, acceptable and necessary. The concept of Spectrum portability (the ability to be served the best frequency available by virtue of one's role at any place and time) is discussed further in the following sections.
26. We refer to this as "artificial scarcity" because it results from regulatory constraints as opposed to technical or market constraints. More spectrum would be available for all users if regulatory policies allowed the cumulatively available spectrum to be shared more efficiently.
27. The business/industrial pool now includes Power, Petroleum, Forest Products, Film & Video Production, Relay Press, Special Industrial, Business, Manufacturers, and Telephone Maintenance Radio Services and the Land Transportation Radio Services (Motor Carrier, Railroad, Taxicab, and Automobile Emergency Radio Services).
28. The NRF describes the national framework for responding to all hazard events, including describing who is responsible for what. The NIMS is the system/framework under the NRF for managing the reporting and tracking of domestic hazard incidents across all Federal, state, and local agencies. See National Response Framework (NRF), U.S. Department of Homeland Security, January 2008 (available at: http://www.fema.gov/emergency/nrf/) and National Incident Management System, U.S., Department of Homeland Security, March 1, 2004 (available at: http://www.nimsonline.com/docs/NIMS-90-web.pdf). The ICS is a management tool, originally conceptualized in the 1970s, intended to assist in emergency response. It identifies best practices and is an important element of NIMs (see http://www.training.fema.gov/EMIWeb/IS/ICSResource/index.htm or Incident Command System Review Materials (2005), available at: http://www.training.fema.gov/EMIWeb/IS/ICSResource/assets/reviewMaterials.pdf).
29. The Industrial, Scientific, and Medical (ISM) bands include frequencies at 900MHz used by cordless phones and 2.4GHz and 5GHz used by wireless LAN technologies like Wi-Fi. Use of these bands is unlicensed, subject to the requirements of complying with the FCC's Part 15 rules for unlicensed devices.
30. Access among co-primary users may be prioritized according to whatever policies the pool managers deem appropriate, with guidance provided by he NIMS/NRF framework.
31. The public safety 700 MHz license and the D-block are discussed further below. Under the current proposal, the FCC's plans call for a commercial licensee to construct a common national infrastructure that would be tied to both the D-block frequency band and the PSST's band. This overlay network would sell "wholesale" services to public safety users across the nation. An alternative perspective might call for this spectrum to be assigned to a common pool that could be shared by public safety infrastructure across the nation, which need not be integrated into a single network.
32. According to Jesuale and Eydt (2007): "Rather than just considering how to prevent interference in isolation, the new paradigm needs to organize spectrum access on the basis of co-existence, adopting, from the field of artificial intelligence, policy-based rules. These rules would govern priority use when contention for access exists, but otherwise maximize access to any available channel across a wide swath of frequencies."
33. See "DC public safety network selected to monitor effectiveness in sharing radio spectrum with federal, state, and local governments," Government Technology, May 1, 2006 (available at: http://www.govtech.com/gt/ articles/993555).
34. See Alaska Land Mobile Radio: "Our commitment is to provide a wireless, cost effective public safety interoperable communication system for all of Alaska" (http://www.ak-prepared.com/almr/); or, Clifton, Kristine (2007), "Alaska Land Mobile Radio (ALMR) Update," powerpoint presentation, November 28, 2007 (available from: http://www.fea.gov/Documents/ Spec%20Briefings/ALMR%20Update%20to%20FEA%2028%20Nov%2007.ppt)
35. See, for example, Public Safety Spectrum Trust (PSST), "Public Safety Spectrum Trust Public/Private Partnership Bidder Information Document," November 2007, available from: http://www.psst.org/documents/ BID2_0.pdf).
36. The 700 MHz public safety band is actually composed of two sets of spectrum; half is reserved for narrowband licenses, and the other half is reserved for broadband uses. The broadband allocation has been licensed to the PSST, and is intended to be shared by the D-Block licensee, should one be awarded. Currently, its status remains in limbo as the FCC reconsiders.
37. In comments to the FCC, Coleman Bazelon identified a number of problems with the D-Block auction rules which contributed to its failure. For example, he estimated that the restrictive rules imposed on D-Block bidders, constraining them to share spectrum with public safety providers and be subject to public safety pre-emption substantially reduced the value of the license. See Comments of Coleman Bazelon, In the Matter of Service Rules for the 698-746, 747-762 and 777-792 MHz Band (WT Docket No. 06-150), Implementing a Nationwide, Broadband, Interoperable Public Safety Network in the 700 MHz Band (PS Docket No. 06-229), Development of Operational, Technical and Spectrum Requirements Permitting Federal, State and Local Public Safety Communications Requirements through Year 2010 (WT Docket No. 96-86), Federal Communications Commission, June 20, 2008 (see http://fjallfoss.fcc.gov/prod/ecfs/retrieve.cgi? native_or_pdf=pdf&id_document=6520030998).
38. See FCC, Second Further Notice of Proposed Rulemaking In the Matter of Service Rules for the 698-746, 747-762 and 777-792 MHz Bands (WT Docket No. 06-150), Implementing a Nationwide, Broadband, Interoperable Public Safety Network in the 700MHz Band (PS Docket No. 06-150), Adopted May 14, 2008 (see http://hraunfoss.fcc.gov/edocs_public/attachmatch/FCC-08-128A1.pdf)
39. Repeaters are needed for operation at 4.9GHz because operation at these higher frequencies requires line-of-sight and has poor penetration properties through building materials. The density of repeaters needed substantially increases the up-front infrastructure investment cost required to enable wireless broadband services in this band.
40. This suggests the importance of cost/ease of deployment in public safety provisioning decisions, and the willingness to trade-off strong reliability requirements to save money or enable advanced capabilities.
41. For example, a static policy might include notching out frequencies that are never to be included in the pool. A regional policy might specify different sets of pooling frequencies for different locales, which could be identified/enforced by the GPS capability of the radio system. A more dynamic policy might be a pre-emption notification indicating that certain frequencies are no longer available in a specific locale at a specific time.
42. For a discussion of how time-limited certificates might be used in a CR architecture to enforce dynamic policies, see Chapin, John and William Lehr (2007), "Time-limited Leases for Innovative Radios," IEEE Communications Magazine, June 2007.
43. See Chapin, John and William Lehr (2007a), "The path to market success for dynamic spectrum access technology," IEEE Communications Magazine, Special Feature on Cognitive Radios for Dynamic Spectrum Access, May 2007.
44. See Chapin, John and William Lehr (2007b), "Time Limited Leases for Innovative Radios," IEEE Communications Magazine, June 2007.
45. Among the technologies being discussed to manage preemption are listen-before-talk, reliance on a control channel, and time-limited certificate leases.
46. nd IEEE International Symposium on New Frontiers in Dynamic Spectrum Access Networks (DySPAN2007), 17-20 April 2007, pages 331-339.
47. Even in the absence of uncertainty, there is a dynamic programming problem that individual networks must solve: the optimal time to adopt will be when the net present value of the future stream of benefits is larger than the net present value of the costs of adoption, after accounting for the foregone benefits/costs you would have incurred if you had adopted earlier. The relative rates of change in benefits and costs (and heterogeneity across potential adopters due to local circumstances) will result in a distribution of optimal adoption times. When you introduce uncertainty, there is a real option effect which might cause you to wait to adopt (i.e., learn which is the best technology before committing investment to a technology which you might have learned is inferior). This uncertainty is likely to make early adoption even more expensive.
48. See David, P. and Greenstein, S., "The Economics of Compatibility Standards: an Introduction to Recent Research", Economics of Innovation and New Technology, vol 1 (1990) p3-41.
49. See Farrell, Joseph and Garth Saloner (1986), "Competition, Compatibility and Standards: The Economics of Horses, Penguins and Lemmings", Technological Innovation Project/Political Economy of Technological Standards Seminar, Stanford University, October 1986. Farrell and Saloner describe how pioneers used to tie their horses together on the plains to keep them from wandering off. The horses could not agree on a direction to go so they stayed put. They also mention how penguins accumulate on the edge of the ice before jumping in, eventually getting so crowded that a few are pushed in. The penguins do this because they want to see if there is a predator waiting. If the first penguin in is not eaten, then the rest can jump in safely. Finally, the example of lemmings provides an image of the difficulties of stopping a bandwagon once started.
50. The case for imposing usage fees on commercial users who might share public safety spectrum in the future is more ambiguous.
51. The question of whether it is technically feasible to pre-empt secondary users has already been addressed.
52. Operators may elect to adopt prioritization that gives higher priority to secondary users (e.g., mutual aid, first responders who are roaming) over primary users. Such a policy is elective and so does not represent an involuntary loss of spectrum access.
53. Similarly, ad hoc/mesh networking that enables telephony in places not served by traditional infrastructure (on a mountain fighting a forest fire or in a collapsed subway tunnel) provides enhanced reliability with respect to the challenge of getting communication services where they are needed.
54. With limited service choices, making decisions is easier.
55. If every gadget carried could be a phone or if they all use the same batteries then reliability of phone service no longer depends on the likelihood of failure of any single device.
56. Unlike commercial customers who can choose where to locate their operations to minimize costs, public safety providers need to be where the need is.