Author Description

We consider the problem of correlated data gathering in sensor networks. Existing solutions to this problem are impractical because they have not considered all of the following factors: (i) distributed implementation, (ii) capacity and interference associated with the shared-medium, and (iii) realistic data correlation model. In this thesis, we propose a new distributed framework to achieve minimum energy data gathering, while considering all three mentioned factors. The framework is first constructed with the assumption that the network has a single sink and perfect data correlation. Our work is then extended to where the network has multiple sinks with arbitrary data correlation. For both cases, the problem is first modeled as an optimization formulation. The formulation is then relaxed with Lagrangian dualization and solved with the subgradient algorithm. To evaluate its effectiveness, we have conducted extensive simulations. The results indicate that the algorithm supports asynchronous networks, sink mobility, and duty schedules.