Efficient data collection through multi-path routing structures in wireless sensor networks
Luu, Van Hai
Date of Issue2012
School of Computer Engineering
Parallel and Distributed Computing Centre
In this thesis, we investigate data collection in wireless sensor networks through multi-path routing structures. We study methods to improve the robustness, energy efficiency and time efficiency of sensor data collection in error-prone communication environments. First, we investigate the construction of a class of multi-path routing structures called rings overlay for enhancing the robustness of sensor data collection. Rings overlay exploits the broadcast nature of wireless communication for transporting sensor data through multiple interleaving propagation paths to the base station. We propose a new distributed approach for organizing sensor nodes into different rings around the base station to form a rings overlay.The objective of our proposed approach is to assign sensor nodes to appropriate rings to let them benefit from multi-path routing as much as possible. The proposed ring assignment approach is fully distributed and does not require sensor nodes to have global knowledge about the entire network.We also design and analyze an enhanced scheme for relaying data to the base station from the sensor nodes next to the base station. The goal is to improve the resilience of these nodes to communication failures in data collection without requiring them to transmit their data multiple times. Experimental results show that compared with a baseline greedy construction approach and the original relay scheme, the proposed techniques of overlay construction and relay enhancement significantly improve the robustness and accuracy of sensor data collection through the rings overlay.
DRNTU::Engineering::Computer science and engineering::Computer systems organization::Computer-communication networks