AbstractsComputer Science

Wireless sensor networks have shown great potential as the technology that will change the way we interact with the physical world around us and have forced researchers to reconsider the way they think about distributed systems. However, these networks have to deal with a great deal of uncertainty arising out of the unique differences in their computational model such as unreliable communication, severely resource constrained devices and vulnerability to different types of faults. To meet these challenges, we must first understand the different reliability issues related to wireless sensor networks and then design appropriate mechanisms to deal with them. We believe network management to be a key enabler for such networks to deal with these challenges. In this dissertation, we first present a comprehensive study of different types of node and network faults that occur in wireless sensor networks and propose a fault model for these networks. Based on this fault model, we identify key elements of a network management architecture for wireless sensor networks. We then present MASE, a Management Architecture for SEnsor networks, that addresses management issues at all levels in a sensor network: at individual nodes, in the network, and also at the base station. We emphasize self-stabilizing designs for MASE components to deal with anticipated and unanticipated faults. We present key network management services such as the Stabilizing Reconfiguration service, the Chowkidar health monitoring service and the Reporter termination detection service that we have designed and implemented as part of MASE. We also present our network-based experiment orchestration framework which closes the loop in sensor network management by automating common execution and experimentation patterns. The different architectural components presented in this dissertation have been validated not only through experiments, but also in field deployments for managing large scale sensor network systems such as “A Line In The Sand” and “ExScal”. Implementations for existing services and tools developed as part of the MASE architecture, for mote, Stargate and server platforms, are also publicly available in the form of a MASE toolkit.