Hybrid Sensor Networks for Emergency Critical Scenarios

Short Description single page overview of the project, its objectives, and expected outcomes comprehensible to the non-expert Situation awareness is at the basis of security in many scenarios going from private and public building to outdoor areas such as forests, harbors, or nation boundaries, for both civil and military applications. Monitoring systems are widely recognized as an invaluable tool whenever surveillance, intrusion or hazard detection are needed to ensure prevention, prediction and protection against critical events such as natural disasters or homeland security threats. Such monitoring systems provide collaborating sensing devices to gather measures concerning the status of the monitored field of interest, to report the collected information to data collection and processing points, and to ultimately trigger the necessary intervention when needed. While sensor networks are common place in safely accessible environments, such as production plants, bridges, or buildings, where sensing devices are deployed and maintained by skilled personnel, we envision their application even in critical scenarios such as hostile environments, battlefields, or areas subject to natural disasters such as earthquakes, volcano eruptions, floods or to large scale accidents such as nuclear plants explosions or chemical plumes. Several attempts in this direction have been made with wireless sensor networks, be them terrestrial, underground, aerial or underwater. Perhaps the most important reason preventing the use of such networks in the above mentioned operative settings is the difficulty in quickly deploying a network which guarantees complete coverage of the area of the critical event of interest, with the necessary connectivity and time responsiveness. A prior deployment of the network is not necessarily a solution, as critical events are not always predictable. Furthermore, an already deployed wireless sensor network has limited lifetime because sensor nodes are powered by short-lived batteries whose replacement or recharge is expensive, if even possible. Existing solutions lack of practical applicability in the above described operative settings due to several unrealistic assumptions, which we remove thanks to an accurate preliminary experimental study of existing technologies and their limitations in the project target scenarios. Therefore we address the case of on-demand, event-driven deployments, where complete and prompt coverage of the event area is achieved by a network composed of heterogeneous devices, hereby called HSNs (Hybrid Sensor Network). We aim at exploiting the sensing capabilities of cheap terrestrial sensors, and to increase the network coverage capability by means of UAV (Unmanned Aerial Vehicles). UAVs can be used for enhancing network coverage by both performing additional sensing and dropping additional small sized static sensors, and for increasing network connectivity by acting as data mules to drive the information collected by the static sensors towards a possibly remote collection and processing center. Furthermore we consider UAVs as a way to offload the static network, when sensors have limited memory and quickly saturate their capacities due to upcoming events which require a sudden increase in the data collection rate. Finally we want to explore UAVs endowed with cutting edge wireless recharging technologies to ensure the realization of a long lasting network. In summary these are the objectives of the proposed research: • Design new algorithms and protocols for event driven deployment of hybrid networks, including interaction schemes between terrestrial and aerial sensors for increasing coverage and connectivity. • Design new routing and communication protocols for hybrid networks in which some nodes are static and others may move in an autonomous and active manner. Design protocols that overcome limits of existing technology in terms of configurability, power consumption, scalability, fault tolerance and security. • Define new mission assignment algorithms to let devices autonomously coordinate with each other to perform several sensing tasks with different requirements, priorities and sampling rates. • Explore the possibility to use autonomous mobile vehicles such as small robots, or UAVs, which can be dispatched to recharge static sensors using radio waves and design corresponding algorithms and path planning strategies. • Design new algorithms and protocols to exploit UAVs to prevent sensing node storage overflow which may occur in case of poor network connectivity. Bartolini/Khalifeh project plan • Define distributed algorithms and protocols for context assessment and situation awareness, which will enable nodes to perform a collaborative work of determining the status of the environment and consequently trigger an intervention if necessary. • Develop machine learning mechanisms for determining and possibly predicting environmental changes. These mechanisms are necessary to ensure the necessary responsiveness of a real time, mission critical monitoring network. We underline that the current project will bring the end users into the research cycle from the very beginning. We will gather requirements from their experience and receive continuous feedback through all the research phases, and we will finally assist them on the exploitation of the developed monitoring systems. We will develop a new generation of large-scale, heterogeneous monitoring systems, which can be deployed on demand, even in the most critical scenarios, guaranteeing a very flexible service which can adapt to geographical conditions, dynamically changing operative settings, and upcoming critical events. We address realistic problems that have plagued real deployments of hybrid sensor networks existing so far, in security critical scenarios. With the proposed solutions, we believe that practical implementations may be pursued to make such autonomous hybrid sensor networks a reality.