Segmenting a sensor field: Algorithms and applications in network design

• Authors: Xianjin Zhu;Rik Sarkar;Jie Gao
• Venue: ACM Transactions on Sensor Networks (TOSN)
• Year: 2009

The diversity of the deployment settings of sensor networks is naturally inherited from the diversity of geographical features of the embedded environment, and greatly influences network design. Many sensor network protocols in the literature implicitly assume that sensor nodes are deployed inside a simple geometric region, without considering possible obstacles and holes in the deployment environment. When the real deployment setting deviates from that, we often observe degraded performance. Thus, it is highly desirable to have a generic approach to handle sensor fields with complex shapes. In this article, we propose a segmentation algorithm that partitions an irregular sensor field into nicely shaped pieces such that algorithms and protocols that assume a nice sensor field can be applied inside each piece. Across the segments, problem dependent structures specify how the segments and data collected in these segments are integrated. Our segmentation algorithm does not require any extra knowledge (e.g., sensor locations) and only uses network connectivity information. This unified spatial-partitioning approach makes the protocol design become flexible and independent of deployment specifics. Existing protocols are still reusable with segmentation, and the development of new topology-adaptive protocols becomes much easier. We verified the correctness of the algorithm on various topologies and evaluated the performance improvements by integrating shape segmentation with several fundamental problems in network design.