Search RESCUE:

 

RESCUE Login:
User Name:
Password:

 

 

ITR-RESCUE is part of the California Institute for Telecommunications and Information Technology (Calit2) and its IT infrastructure is provided by Responsphere

GLQ - Gas Lamp Quarter Testbed

The Gas Lamp Quarter (GLQ) testbed consists of a rapidly deployable mobile networking, computing, and geo-localization infrastructure in the context of incident-level response to spatially localized disasters such as the World Trade Center attack. The testbed focuses on situations where the crisis site either does not have an existing infrastructure, or alternatively, the infrastructure is severely damaged. This testbed focuses on supporting basic services essential to the first responders that can be brought over to crisis sites for rapid deployment. Such services include communication among the first responders, accurate geo-localization both inside and outside of buildings, in urban as well as rural areas, computation infrastructure, incidence level command center, and technology to support information flow from/to crisis sites to/from regional emergency centers.

The primary demonstration study will be in the context of localized incident management in the Gas Lamp district in downtown San Diego . The Gas Lamp Quarter will be instrumented in partnership with the civic authorities and the San Diego Police Department. This 10 block historic region adjoins the San Diego convention center, sports stadiums as well as city, county and federal facilities and is described as Southern California 's premier dining, shopping and entertainment district. Some of the key issues that will be addressed in this testbed are discussed below.

Deploying a Network as a Living Lab

For this testbed, there is a need to have a reliable and controlled wireless network that covers the Gas Lamp Quarter area in downtown San Diego . Although there are several wireless ISPs in that area, most of them provide patchy coverage and their systems are not always reliable.

The RESCUE project is in a good position to design and deploy its own network to cover that area. This will give researchers the opportunity to have a more reliable network that is fully under their control. The following issues are being addressed as we study possible solutions for this deployment:

Frequency Band: The system can be deployed in an unlicensed band and use standard off-the-shelf equipment. This will reduce costs significantly. Our approach is to have a hybrid network design. While using the standard 802.11b for reaching end-users, 5.2 GHZ and 5.7 GHz will be used for backhauls to increase capacity. The other opportunity is using Ensemble (a San Diego based company)/XO partnership. Recently, Ensemble and XO announced that they can provide equipment and spectrum to support any wireless deployment in the Downtown San Diego area. This can increase the reliability of the system since this system can be operated in the license band.

Site Acquisition: Sentre Partners, one of the leading real estate companies in San Diego , has already committed itself to a related project to promote the city. They have three of the tallest buildings in downtown area. By using their rooftops, it will be easy to have a large footprint and cover the downtown area.

Backhauls: The cost of the backhaul is the most important factor that should be considered for the long-term. It is possible to involve some of the telephone companies, i.e., Verizon or SBC, to provide the backhaul. However, there is also a possibility for using under-utilized bandwidth in the buildings above to connect our base stations to the network. Sentre Partners, the owner of NBC building, is committed to providing access and enough bandwidth for this testbed.

Network Architecture:

As it is shown in Figure 1, the GLQ is divided into three different zones. Each zone has its central post that has LoS to the top of NBC building. The transmitter on top of NBC building provides the broadband access to these three lampposts via a 5.2 - 5.7 GHz backhaul (the red points). By using Tropos Networks 5110 outdoor units (the green points), the coverage of three zones will be expanded and we are able to provide the support for standard 802.11b users. 5110 units will find each other and mesh together using standard ISM band and they will also act as an 802.11b access point for the end users. Therefore, it will be possible (by connecting three locations to the network) to cover a large area. Also by having three entry points, the reliability of the system will increase in case of outage for one of the backhauls. Three lampposts with backhaul connectivity (5 th and Market, 5 th and E, and 4 th and G) will have a Motorola Canopy receiver and a Tropos 5110 unit installed, see Figure 2. The others will just have a Tropos 5110 unit installation. Provision of power is the only major requirement for the operation and it has been tested on these same lampposts as part of a Super Bowl experiment. The 802.11b cells communicate with each other wirelessly through a mesh routing algorithm implemented within the access points. The control protocol is part of Tropos Sphere operating and management tool.

The network topology is based on standard centralized routing. All the routing and switching is managed centrally. The backhaul links will act as pure second layer channels, whereas the 802.11b access points will do the DHCP server functionality locally. The system will support mobility based on TCP and VPN session persistent roaming without client software upgrade. The system will be managed remotely through a web-based management tool for the Tropos and standard SNMP.

Figure 1 - GLQ Setup including Location of Three Major Zones

 

Figure 2 - Central Posts to be used in GLQ Testbed

 

The deployment plan is divided into three phases:

•  Phase 1 (Proof of concept): In this phase, the inter-operability of the system will be tested in a controlled environment (lab). This includes setting up one zone and using the backhaul radio link and networking gear.

•  Phase 2 (Test deployment): In this phase, the tested zone will be deployed and real measurements will be performed to check and confirm the operational parameters of the testbed. The monitoring software will be also tested in the real environment.

•  Phase 3 (Final deployment): In this phase, the other two phases will be deployed. Final tests and optimization will be done to adjust the network.