A. Munteanu, J. Beaver, A. Labrinidis, and P.K. Chrysanthis (USA)
load balancing, in-network aggregation, reliability
Advances in sensor technology provide the opportunity for a wide range of applications not examined before. These advances come with the realization of many limitations, like energy constraints, communication limitations, or sen sor node failures. To combat these limitations, several so lutions have been proposed, most of which organize the sensor nodes into a tree-like configuration that enables in network aggregation. These solutions can lead to a heavy energy burden being put on nodes higher up in the tree, causing node failure and "stranded" nodes, unable to com municate their results. One solution to this problem is to use multiple query routing trees, which leads to a more bal anced network with less burden on the nodes higher in the tree. In this paper, we present a system framework for us ing multiple query routing trees, along with an analytical examination which enables us to determine the appropri ate number of trees to be used and the proper placement of those trees. We also provide an evaluation tool for different network configurations.
Important Links:
Go Back
