| Summary: | 碩士 === 國立交通大學 === 應用數學系所 === 104 === In wireless sensor networks (WSNs), hole detection and hole healing are two important problems in maintaining a stable coverage. Coverage holes may arise in a Region of Interests (RoI) due to the running out of batteries of sensors or due to the damage of sensors. Hence, the development of means to detect and to heal the coverage holes is of great significance. For hole detection, Kang et al. [3] provided a
decentralized, coordinate-free, node-based coverage hole detection algorithm which maps out the coverage holes by connecting the identified boundary critical points. In this thesis, we first adopt the approach of [3] to map out the boundary critical points. Then, we try to reduce the number of edges of coverage holes; the purpose is to reduce the total number of healing sensors. Our approach to reduce the number of edges of coverage holes is: we replace each boundary critical point with the center of the corresponding sensing circle, and remove redundant centers. For hole
healing, Shiu et al. [9] used a divide-and-conquer deployment algorithm based on partitioning the hole into triangles and then, healing each triangle. In this thesis, we propose a dynamic programming approach to minimize the total length of diagonals used in partitioning the hole into triangles. The most important contribution of this thesis is that we find that T can be minimized if D is minimized; where T is the total number of healing sensors and D is the total length of diagonals used to partition coverage holes into triangles. Compared to [3], our hole-healing algorithm guarantees that there is no remaining hole, i.e., the RoI is fully covered. Compared to [9], our hole healing algorithm usually has a smaller T .
|
|---|
