An Approach to Construct Weighted Minimum Spanning Tree in Wireless Sensor Networks

Abstract

Topology control is critical to extend the lifetime of energy constrained Wireless Sensor Networks (WSNs). Topology control mechanism can be divided into two processes: topology construction and topology maintenance. During topology construction one creates a reduced topology to ensure network connectivity and coverage. In topology maintenance, one recreates or changes the reduced topology when the network is no longer optimal. In this research the authors concentrate on Minimum Spanning Tree (MST) which is a commonly seen problem during the design of a topology construction protocol for WSNs. As the amount of running time and messages successfully delivered are important metrics to measure the efficacy of distributed algorithms, much research to create simple, local and energy efficient algorithms for WSNs thereby creating sub optimal MSTs has been studied. In this research, two popular approaches are discussed to create a Spanning Tree in the WSNs- Random Nearest Neighbor Tree (Random NNT) and Euclidian Minimum Spanning Tree (Euclidian MST). Next, the authors propose a method which has the goals to balance the network load evenly among all of the nodes and increase the number of successful message deliveries to the sink. Finally a comparison between the three algorithms is conducted in the Matlab environment. Simulation results demonstrate significant improvement for both load balancing and number of message deliveries after implementation of the proposed algorithm.KeywordsTopology construction protocolMinimum Spanning TreeNearest Neighbor TreeLoad balancingSimple Weighted Spanning Tree