• DOI:10.4018/jiscrm.2010070103
• Corpus ID: 33817022

Achieving Electric Restoration Logistical Efficiencies During Critical Infrastructure Crisis Response: A Knowledge Management Analysis

@article{Durbin2010AchievingER,
  title={Achieving Electric Restoration Logistical Efficiencies During Critical Infrastructure Crisis Response: A Knowledge Management Analysis},
  author={Teresa Durbin and Murray Eugene Jennex and Eric G. Frost and Robert Judge},
  journal={Int. J. Inf. Syst. Crisis Response Manag.},
  year={2010},
  volume={2},
  pages={36-50}
}

• T. Durbin, M. Jennex, +1 author R. Judge
• Published 1 July 2010
• Computer Science
• Int. J. Inf. Syst. Crisis Response Manag.

The technology used in citizen-to-authority emergency calls is based on traditional telephony, that is, circuitswitched systems. However, new standards and protocols are being developed by the Internet Engineering Task Force (IETF) to allow emergency communications over packet switched networks, such as the Internet. This architecture is known as Next Generation-9-1-1 (NG-911). In this paper, the authors present lessons learned from experiments on the IETF standard called Location to Service Translation protocol (LoST). LoST maps the user’s location to the address of the emergency call center that serves that location. After implementing the standards in a test-bed with real-world systems, spatial databases, and communication networks, the authors observed performance issues that users may experience. Based on their observations, the authors propose practical ideas to improve the performance of the NG-911 system and LoST protocol operation for mobile users. (e.g., 9-1-1, 1-1-2, etc). In order to better explain the context of this paper, next we provide a brief history of emergency calls, followed by a description of this paper’s motivation. A. A Brief History of Emergency Call Systems It was in 1937, in Great Britain, that the idea of a single emergency number was developed. DOI: 10.4018/jiscrm.2010070101 2 International Journal of Information Systems for Crisis Response and Management, 2(3), 1-24, July-September 2010 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. By dialing 9-9-9, British citizens could contact the police, fire department, or medical services. Only later, in 1958 US legislators began to consider the use of a single emergency number. The number chosen in the US was 9-1-1. It is known that one of the reasons for this choice is that the numbers were located on opposite sides of the key-pad, which reduced the chances of accidental calls. In 1968, the first 9-1-1 call was made in Alabama. Later that year, the American Telephone and Telegraph Company (AT&T) announced the 9-1-1 service. This first 9-1-1 service is also known as “traditional 9-1-1”. It directed all emergency calls to a nearby Public Safety Answering Point (PSAP) which was directly connected, via dedicated trunks, to the telephone company’s central office (CO). A typical PSAP consists of a group of call takers who are responsible for answering emergency calls as well as dispatching the appropriate emergency service (such as ambulance, or police, or fire fighters). Because the plain old telephone system (POTS) network only established and terminated calls, call takers had few resources to determine the caller’s location. As a result, traditional 9-1-1 soon gave way to landline Enhanced 9-1-1 (E-911), which greatly increased the ability to locate the caller. E-911 employs an address database called Automatic Location Identification (ALI), which uses the telephone number of the caller to determine his/her identity and location. E-911 was implemented in 1978 and is still being used today (Figure 1). Landline E-911, however, only temporarily met the new technological need. Cell phones soon made it necessary to again upgrade the system. The Federal Communications Commission (FCC) mandated that all United States wireless carriers provide the location information of the caller (Reed, Krizman, Woerner, & Rappaport, 1998). Implementation of this mandate would enable wireless carriers to “pinpoint” the location of any 9-1-1 caller. The FCC mandate was divided in two phases: Phase I required the wireless carrier to send the PSAP the location of the antenna or cell site; Phase II required the wireless carrier to send an estimate of the geographic coordinates of the mobile user, typically using some type of network triangulation scheme (Feng & Law, 2002; Sayed, Tarighat, & Khajehnour, 2005). Not all wireless providers have implemented Phase II service as of 2009, based on the National Emergency Number Association (NENA)’s recent 9-1-1 Deployment Reports (NENA, 2010). In the late 1990s, with the onset of voice over Internet Protocol (VoIP) (Goode, 2002) and other Internet services, researchers and developers have realized that a complete overhaul of the current E-911 system is necessary – proposing a Next Generation 9-1-1 (NG-911) (Hixson, Cobb, & Halley, 2007; NENA, 2007a, 2007b; Schulzrinne & Arabshian, 2002) system that Figure 1. Evolution of North American 9-1-1 emergency calling system International Journal of Information Systems for Crisis Response and Management, 2(3), 1-24, July-September 2010 3 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. enables both voice and data communications for emergency calls. Recently, an NG-911 proof-of-concept was developed (Kim, Song, & Schulzrinne, 2006; Song et al., 2008). This development represents a critical step and an important building block to validate IP-based emergency calling. The proof-of-concept allowed the testing of IPbased emergency calls to IP-based PSAPs on five main scenarios: (i) calls from traditional landline phones, (ii) calls from cellular phones, (iii) text messages (SMS), (iv) telematics service messages, and (v) multimedia calls from fixed/ enterprise VoIP devices. In addition, the NG-911 proof-of-concept tested the idea of Emergency Services Router Proxies (ESRPs) for policy-based routing. ESRPs improve the call routing capability of 9-1-1 calls: they can re-route emergency calls to any PSAP (e.g., PSAPs that are not overloaded, or PSAPs with special capabilities such as the ones with sign-language capable call takers). With IP-based emergency calling, a PSAP can now be anywhere, even in a mobile device such as a laptop. The NG-9-1-1 proof-of-concept and all the technology behind it mean that a PSAP does not have to be a centralized call center as before, but can be a much more flexible, portable, and less expensive system. The NG-911 proof-of-concept concludes our brief history of the 9-1-1 evolution. However, IP-based emergency calling still faces new challenges. Standards are being developed, security vulnerabilities must be addressed, and the transition from PSTN-based PSAPs to IPbased PSAPs is being investigated. B. Motivations for This Paper One important NG-911 research challenge is related to the location of the caller, which can be very dynamic. Enterprise VoIP phones have the flexibility to be connected to any network port. Home VoIP subscribers can be connected to any access network (e.g., an Internet service provider). Also, mobile devices such as dualmode cell phones (with 802.11 radio), laptops with broadband wireless coverage (using 3G, or 4G cellular data cards) can be constantly changing its location. Moreover, the architecture for IP-based emergency services will have different entities: the access infrastructure network and the application service provider (such as a voice service provider (VSP)) (Tschofenig, Schulzrinne, Shanmugan, & Newton, 2007). This poses unique challenges to current IP-based emergency systems. For instance, it is the subscriber’s responsibility to provide his/her location to the VoIP service provider; however, the subscriber may not update this information when the VoIP device is moved to a different location (i.e., a different access network). As a result, VoIP service providers may not have the most up to date civic or geographic address of the user. New Internet Engineering Task Force (IETF) standards have been created to allow the addition of location information (Peterson, 2005; Polk & Rosen, 2009) to the Session Initiation Protocol (SIP) call setup messages (known as SIP Invites) (Rosenberg et al., 2002). In addition, the Location to Service Translation (LoST) protocol (Hardie, Newton, Schulrzrinne, & Tschofenig, 2008; Schulzrinne, Tschofenig, Hardie, & Newton, 2007; Schulzrinne, 2008) has been created to help both end users or SIP proxies to obtain the SIP address of the PSAP where the emergency call must be routed to. This SIP address is known as SIP Uniform Resource Identifier (URI) (e.g., sip:psap1@domain). Initial experiments using the LoST protocol were performed in (Schulzrinne, Tschofenig, Hardie, & Newton, 2007), which presents the concept of the LoST protocol, introduces the idea of the Emergency Services IP Network (ESInet), and how the protocol queries are performed. The authors conclude the paper with a few performance results, mainly delays of the LoST queries. Additionally, the use of emergency services by mobile users is described by an IETF standard on best current practices (Rosen & Polk, 2009; Schulzrinne & Marshall, 2008). It provides a set of guidelines on obtaining the location of 4 International Journal of Information Systems for Crisis Response and Management, 2(3), 1-24, July-September 2010 Copyright © 2010, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited. mobile users, querying LoST servers, and storing the PSAP URI at a cache at the SIP/LoST client (Wu & Schulzrinne, 2004). In Chintapatla, Goulart, and Magnussen (2010), we presented preliminary results from an initial work with the NG-911 testbed, where preliminary performance results on the LoST queries and processing delay at the mobile client were presented. Now, we present in this paper more detailed experiments, lessons learned, and a complete overview of the operation and performance of the LoST protocol, including description of techniques in computational geometry for processing location information and service boundaries data (both at the LoST client and server).