Authors: Shoudong Zou, Ioanis Nikolaidis, Janelle Harms
Addresses: Department of Computing Science, University of Alberta, Edmonton, Alberta T6G 2E8, Canada. ' Department of Computing Science, University of Alberta, Edmonton, Alberta T6G 2E8, Canada. ' Department of Computing Science, University of Alberta, Edmonton, Alberta T6G 2E8, Canada
Abstract: We study the approaches to extending the lifetime of Wireless Sensor Networks (WSNs) based on in-network data aggregation at the First Hop (FH) away from each sensor data source, followed by flow-based routing of the resulting traffic. We introduce the concept of flow loss multiplier to express the impact of data aggregation on the conveyed data. A Mixed Integer Linear Programming (MILP) model is formulated for the problem of determining the optimal FH aggregation nodes that maximise network lifetime. Heuristics are proposed to obtain significant aggregation and to prolong the system lifetime. To facilitate performance evaluation, we adopt a flow loss multiplier that depends on the spatial relationship among sensed areas. Simulation results show that FH aggregation provides significant increase in network lifetime for both flow-based and tree-based delivery schemes, and that flow-based mechanisms provide better network lifetime than tree-based mechanisms but at the cost of added complexity and overhead.
Keywords: wireless sensor networks; WSN; data aggregation; routing; load balancing; first hop selection; network lifetime; flow loss multiplier; simulation.
International Journal of Sensor Networks, 2008 Vol.4 No.1/2, pp.55 - 67
Published online: 04 Jul 2008 *Full-text access for editors Access for subscribers Purchase this article Comment on this article