WiFi repeater deployment for improved communication in confined-space urban disaster search Alexander Ferworn1, Nhan Tran1,2, James Tran1, Gerry Zarnett1, Farrokh Sharifi2 1 Network-Centric Applied Research Team Department of Computer Science 2 Department of Mechanical Engineering Ryerson University Toronto, ON, Canada monitored at the perimeter of the space being Abstract entered. We examine the use of WiFi repeaters to extend the range of wireless devices employed If a robot or similar device is to be used, in confined space search in an Urban Search communication and control of the devices are always a challenge. Typically, robots are and Rescue (USAR) environment. Confined equipped with a tether that allows spaces are often encountered by emergency communication between a responder and the responders at disaster sites. They typically are too small or hazardous to be searched by device. Another option that avoids tether humans but may be accessible by rescue robots management issues is the use of wireless control. controlled and monitored from outside the space. In this study we examine the use of a WiFi network used for control and monitoring We employed a purpose-built confined employing relaying units called “pucks” to space training area constructed of steel extend the communication range of interaction reinforced sewer tubing to determine strategies possible. We employ a purpose-built confined for maintaining a usable signal for devices space training area constructed of steel deployed within spaces with similar reinforced sewer tubing to simulate the effects of rebar and other debris often encountered in characteristics at disaster sites. USAR environments. Keyword: USAR, WiFi, repeater, rescue robots, confined space 2 Tethered Teleoperation 1 Introduction Most Urban Search and Rescue (USAR) Emergency responders are often faced robots currently in operation fall into two with confined spaces. Soon after the initial entry categories, tethered or analog radio frequency of the search site searchers find themselves communication [1, 2]. Tethered robots benefit blocked due to debris or other physical barriers from having the possibility of a continuous present that are due to the events of the disaster. power feed, dependable communications,  While there may be much debris, there are and relatively easy retrieval from the search usually many voids or other confined spaces that space. The problems with tethered robots are may contain casualties or lead to larger spaces typically associated with tether management. and access paths. Confined spaces are small and Hard cable tethers require large cumbersome usually hazardous spaces typically inaccessible spools, and the tether itself is present as a load by humans but may be accessible by means such on the robot Lighter tethers may be broken, as rescue robots. These robots are often quite cause snagging or become tangled in the robot small and can be remotely controlled and itself . All these factors are further exacerbated with multiple robots in one confined The use of analog and digital wireless search space. communication for USAR robots has been examined by many investigators in the literature 2.1 Analog Radio Control [1,4-6, 9] Tunnel rat was deployed for sewer reconnaissance with the use of analog RF. It suffered from significant breakup and signal Robots communicating via wireless degradation. Similar implementations in means are used when possible because they are confined space proved that it was nearly often much easier to deploy. Most of these impossible to maintain communications beyond robots use analog radio communication This two meters into a search tunnel. There have form of teleoperation eliminates many problems been example of using robots deployed as relays encountered by tethered robots, but wireless to maintain steady communication when entering robots encounter a completely different set of buildings and other relatively open and problems. Analog Radio Frequency (RF) signals unobstructed hazardous environments [1,4,5] but suffer from interference, scatter, and this technique may not be applicable in more attenuation. Security is also a major problem. demanding settings. Any other device capable of reproducing the same RF signal can take control of the system or Player, primarily a protocol inadvertently “jam” the signal so that control and implemented as a TCP socket server that allows monitoring are lost. devices on the network to access other sensors on the network. Player does not have any direct 2.2 WiFi Selection application to confined space search. However, it may be adapted to provide a middle layer to the An alternative to analog RF ad hoc network to distributed control for robots communication is the use of digital wireless and sensory feedback. communication. Perhaps most popular among the many alternatives is the use of WiFi Confined space search imposes different communications. challenges than other environments. Spaces are always unstructured, usually very small and WiFi (IEEE 802.11) systems provide a debris filled which causes havoc to RF signals very attractive framework for use within an both analog and digital. USAR environment because of their simple architecture, robustness, expandability and their Low-cost and effective repeater “pucks” can be cost effectiveness. By using WiFi we inherit a easily be fabricated from WiFi components and reliable network architecture, and the ability to conveniently powered by batteries. set up an ad hoc network using inexpensive off- the-shelf equipment. The use of relaying pucks to extend signal range may be ideal for confined space search due to The use of WiFi’s Digital-spread- their low maintenance and small size. We are spectrum (DSS) mitigates interference, aware there are many robots systems that have intentional or unintentional jamming, and may been developed with WiFi communication and automatically resolve many problems that plague some even use relay units to extend their range, analog RF communication. but we are not aware of one that is design to address issues encountered in USAR An additional advantage that we explore environments in confined space search. in this paper is that the range of a WiFi networks may be easily extended through the use of repeating transmitters that share all the 4 Confined Space Testing characteristics of the other WiFi components of the network. With the collaboration of the Ontario Provincial Emergency Response Team (PERT) of the Ontario Provincial Police (OPP), we 3 Related Work employed a purpose-built confined space training and testing facility as a simulated disaster site exhibiting the radio characteristics often meters directly in front of the access point associated with confined space. indicated (figure 1). This acted as a base where all communication would be received. The facility is constructed of steel reinforced sewer tubing meant to simulate The measurement of the signal strength structures and materials often found in urban and quality was done starting at the access point areas and typically at disaster sites. The facility to the facility. The signal quality was recorded at is composed of a series of interconnecting pipes 50 cm increments. The analog RF signal testing forming a rectangle shape 13.45 meters by 11.4 was accomplished using 2.4Ghz analog meters, depicted by figures 1 and 2. We have transmitter/receiver pair transmitted the signal repurposed the facility to provide a suitable from an infrared camera. For the WiFi signal location for analog and digital RF testing. tests (with repeater) we employed commercial routers and repeaters. These were D-Link DI-624 and D-Link DWL-2100AP respectively. Two laptops with IEEE802.11b capability were used to transmit video. 4.2 Analog Video Test When a robot is sent into an environment instead of a person, often, the video signal is the only means of monitoring the robot’s situation. Therefore it is critical that the video feedback is of a high quality. For our RF signal tests, we used a small analog RF wireless camera system. The wireless camera transmitted video signals via RF to a receiver that output the Figure 1. Top view of confined space video to a monitor. The video receiver and monitor were located at the base station. The wireless camera was moved through Path 1 until the video signal was completely lost. Then the position where the signal loss occurred was recorded. 4.3 WiFi Signal Strength Test The router was setup at the base station. A laptop equipped with signal strength measuring software was connected to the router’s wireless network. The laptop was moved through Path 1 at 0.5m increments and readings of the signal strength were recorded until the signal was lost. The position of the signal loss Figure 2. Actual Facility shown during was recorded. construction 4.4 WiFi Video Test 4.1 Test Setup This test examined video transmission After consultation with PERT members through WiFi transmission. For this test, the we elected to conduct testing in the same manner router and two laptops were used. Laptop 1 and as a deployment would dictate in operation. The the router were at the base station. Laptop 2, confined space was completely sealed with 1.5 equipped with a webcam was moving through cm thick steel doors, except the access point. A the structure along Path 1. Both laptops were master communication point was set up 15 connected on the same network through the router. Laptop 2 was broadcasting the video feed the WiFi signal range can be extended to cover a to Laptop 1. When Laptop 1 could not receive search space through the use of a repeater puck. the video signal from Laptop 2 indicating the It is clear that analog RF is insufficient in this network connection between Laptop 1 and type of hostile communication environment and Laptop 2 was lost, the position of Laptop 2 was only provided up to 1.50 m coverage into the recorded. search space. Digital RF was considerably better with a consistent range of 12.0 m inside the 5 Trials Results tunnels. Tests reviewed how disruptive reinforced concrete can be to analog RF communication. Consistent results showed reception for a clear video signal at 1.0 meter inside the structure and a very weak signal at 1.5 m. The signal was completely lost at 2.0 m. Figure 4. Puck With the use of one puck constructed from readily available materials (Figure 4) we were able to increase the operational range to cover approximately 80 % of the search space. Figure 3. Signal Penetration We see that the miniaturization of repeater pucks and improved power efficiency will be important While the WiFi network also factors for future work. We plan to test this communicated at 2.4ghz, it did considerably system in many different confined spaces to see better than the analog system. We were able to how signal range is affected by different broadcast from inside the structure along Path 1 structures. Of particular interest is the (figure 3) up to the 12.0 m point. At 12.5 meters, development of appropriate algorithms for the the connection to network was dropped and the effective deployment of pucks to maximize the video feed stopped. effective search distance. The thirds set of tests with the puck repeater demonstrated that while the signal was still lost at 12.0 m, communication could be References reestablished with the puck dropped depicted by a red circle in figure 3 and activated. We were 1. Nguyen, H.G., et al. Autonomous able to cover path 2 completely and extend 7 mobile communication relays. in meters into third side of the structure. Proceedings of SPIE - The International Society for Optical Engineering. 2002. Orlando, FL. 6 Conclusion and Future Work 2. Messina, E. and A. Jacoff. Performance standards for urban search and rescue robots. in Proceedings of SPIE - The In this paper we have employed a International Society for Optical purpose built confined space training facility to Engineering. 2006. Kissimmee, FL. emulate similar structures typically found at 3. Hirose, S. and E.F. Fukushima, Snakes urban disaster sites. We have demonstrated that and strings: New robotic components for rescue operations. International Journal of Robotics Research, 2004. 23(4-5): p. 341-349. 4. Nguyen, H.G., et al. Maintaining communication link for a robot operating in a hazardous environment. in Conference on Robotics and Remote Systems- Proceedings. 2004. Gainesville, FL. 5. Nguyen, H.G., N. Farrington, and N. Pezeshkian. Maintaining communication link for tactical ground robots. in AUVSI's Unmanned Systems North America 2004 - Proceedings. 2004. Anaheim, CA. 6. Gerkey, B.P., et al. Most valuable player: A robot device server for distributed control. in IEEE International Conference on Intelligent Robots and Systems. 2001. Maui, HI. 7. Thompson, E.A., et al., Robot teleoperation featuring commercially available wireless network cards. Journal of Network and Computer Applications, 2006. 29(1): p. 11-24. 8. Hyncica, O., F. Zezulka, and P. Fiedler, Wireless standards for mobile platform. WSEAS Transactions and Communications, 2005. 4(5): p. 199- 210. 9. Laird, R.T., Issues in Vehicle Teleoperation for Tunnel and Sewer Reconnaissance. iEEE, 2000.