We, together with Cedric Ulmer, got a paper related to “Secure Sensor Networks for Critical Infrastructure Protection” accepted to the SensorComm’10 International Conference on Sensor Technologies and Applications.
This paper is a first research result of a joint French German research project RESCUEIT.
Abstract: Wireless Sensor Networks raise the interest of different business domains, including public security. The ability of Wireless Sensor Networks to monitor and control physical environments such as football stadiums makes them very attractive. The integration of Wireless Sensor Networks into Command and Control (C2) systems aims at avoiding catastrophes such as the Heysel Stadium disaster. In this paper, the architecture and implementation of a prototype for stadium surveillance is presented. We demonstrate the notification of alerts processed on sensor nodes and routed toward a Command and Control system. The prototype aims at increasing situational awareness of decision makers and at providing real time information related to the occurrence of incidents within the stadium. Decision makers can react accordingly by assigning available first responders to the incidents. Such integration comes along with security requirements; the use of encryption-based access control and trustworthiness evaluation for the generated alerts is therefore proposed to ensure their confidentiality and reliability. In this paper, we demonstrate
the integration of a wireless sensor into a real Command and Control system, and the security requirements raised by such integration.
Wireless Sensor Networks (WSNs) are considered as the next hype of the 21st century . The lack of WSN integration into legacy systems is mainly due to technical (e.g., data routing, processing, heterogeneity) and security related (information confidentiality, integrity) issues introduced by WSNs . Even though many different business domains (e.g., healthcare, military, traffic control) have a strong interest in WSNs, these issues are still an obstacle to their integration. And Command and Control systems (C2) systems are no exception, as we described in .Within the public security life-cycle, the presented approach can be positioned in the Prevent and Respond phases. In the Prevent phase, the main priority is to prevent any dangerous situation as early as possible. WSNs would support decision makers in early detection of a possibly catastrophic event. WSNs have the ability to control and monitor large physical environment, even in potential harmful places for humans. In the Respond phase, the use of WSNs information in C2 systems has two major advantages: (i) it raises situational awareness of decision makers with the incident detection in real-time, (ii) and it reduces the response time between incident detection and proper assignment of first responder forces (e.g., ambulance, firefighters, police) to a critical situation.In order to be profitable for public safety systems, that WSN integration must be as seamless as possible, and face the technical and security issues in a transparent way for the legacy systems. Several architectures have been developed in order to address that integration (see Section V). Nevertheless, these approaches either strongly focus on specific WSNs, or address specific technical or security issues. In addition, most of the existing architectures only consider the data acquisition from WSNs, and barely take the delivery of information to business applications into account. In this paper, we discuss an architecture, which covers the whole chain from the data acquisition in WSNs, to its delivery to C2 systems. Furthermore, we developed a prototype based on a real WSN and a SAP C2 system as proof of concept. We based our prototype on the following real cases. In the past, several catastrophic accidents happened during football games such as during the Heysel Stadium disaster . In our scenario, we place several sensors, sensing acceleration, noise and temperature, in critical areas of a stadium (e.g., on fence, in walls). These sensors are in charge of detecting any dangerous crowd activities (e.g., jumping, or intensively moving or hitting fences and walls) or of detecting dangereous situation (e.g., sudden temperature increase). In case of any abnormal measurement, an alert is sent to a Command and Control System. Based on the received alert, a decision-maker can choose to assign available first responder forces (e.g., ambulance, police) to the incident scene. Our prototype demonstrates alert detection in a WSN and alert notification to a Command and Control Systems. In addition, we show the assignment of first responder forces to incidents in real SAP systems.Such scenarios come along with specific security requirements.
We identified confidentiality and trustworthiness of alerts as the most important for our scenario. Alert confidentiality aims at ensuring that only authorized users have access to the alerts. Alert trustworthiness evaluates the confidence that decision maker can have in the alert. We propose to apply two existing security and trust mechanisms in order to fulfill the identified security requirements.
 C.-Y. Chong and S. P. Kumar, “Sensor networks: evolution, opportunities, and challenges,” Proceedings of the IEEE, vol. 91, no. 8, pp. 1247–1256, 2003. Y. Zhou, Y. Fang, and Y. Zhang, “Securing wireless sensor networks: a survey,” Proceedings of the IEEE Communications Surveys Tutorials, 2008. L. Gomez, A. Laube, and A. Sorniotti, “Secure sensor network for public safety command and control system,”Proceedings of the IEEE International Conference on Technologies for Homeland Security, 2009.
 Wikipedia, “Heysel stadium disaster.” [Online]. Available: http://en.wikipedia.org/wiki/Heysel Stadium Disaster
Laurent Gomez is a Senior Researcher in Security & Trust at SAP Research. He joined SAP Research in Sophia Antipolis in October 2001. He is participating to the EU WASP, Ginseng, and RESCUEIT Projects for secure and trusted integration of wireless sensor networks into business applications.