Large-scale sensor networks have become a reality due to recent developments in sensor node
hardware and algorithms. Sensor networks can provide real-time information based on detection
and tracking. This information cannot be reliable if little is known about the sensor coverage of the network, which can be defined as the total sensing range of the network due to contributions from each sensor node. Knowledge about coverage can also be useful in determining if there is any gap in coverage in the region of interest as well as improving the algorithm that determines the placement of nodes. Although coverage estimation is this thesis's central concern, other factors such as energy-efficiency and network lifespan that affect the network performance are investigated. Energy-efficiency and network lifespan depend on the communication model used for obtaining coverage information from each sensor node. This thesis proposes the use of B-splines for describing coverage efficiently. The properties of B-splines also enable communication models such as directed
diffusion and hierarchical clustering to provide better performance as compared to a centralized
scheme. Results obtained from simulation experiments indicate that hierarchical clustering and directed diffusion can be used effectively for coverage measurement. The hierarchical clustering model, however, exhibited some drawbacks such as a dependency on the routing scheme and poor node-failure recovery.
