Scalable Design of Fault-tolerance for Wireless Sensor Networks

Since wireless sensor networks are inherently fault-prone and since their on-site maintenance is infeasible, scalable self-healing is crucial for enabling the deployment of large-scale sensor network applications. To achieve scalability of self-healing, in this dissertation we focus on addressing (1) the scalability of the cost-overhead of self-healing with respect to the size of the network, and (2) the scalability of the design effort for self-healing with respect to the size of the application software. Our research on fault-containment addresses the first problem: By confining the contamination of faults within a small area, this approach achieves healing within work and time proportional to the size of the perturbation, independent of the size of the network. Our research on specification-based design of self-healing addresses the second problem: Since specifications are more succinct than implementations, this approach yields efficient design of self-healing even for large implementations. These two research directions are complementary, and together enable a scalable design of local self-healing for large-scale sensor network applications.