Distributed energy efficient data gathering without aggregation via spanning tree optimization

Abstract

A distributed algorithm that solves energy efficient data gathering Weighted Spanning Tree Distributed Optimization (WSTDO) is proposed in this paper. It is based on an optimization performed locally on the data gathering spanning tree. WSTDO algorithm is compared to two centralized spanning tree optimization algorithms MITT and MLTTA. The performance of WSTDO achieves between one half and one third of the MITT performance and proves to be better than MLTTA. The performance depends on the density of the network. It works better for sparse networks. WSTDO has lower overhead than MITT and MLTTA for sparse networks. Though the proposed algorithm has a worse performance than MITT it has other features that over-weights this fact. It is able to perform optimization parallely in disjoint sub-trees and also during data gathering which allows a short data sampling period. It is also prone to link and node failures that can be solved locally. © 2013 Springer-Verlag.