Energy efficient data gathering for wireless sensor networks

Abstract

The rapid growth of wireless sensor networks (WSNs) is predominately motivated by recent advancements in solid state and Micro Electromechanical Systems (MEMS) technologies. The sensors are equipped with computing, storage, and communication resources. They are deployed in large quantities for various pervasive and ubiquitous applications such as surveillance, industry automation, and environment monitoring. However, due to its having limited computing and battery power, we decided to in- vestigate energy e cient data gathering techniques. In this research we consider a homogeneous WSN, where all nodes have similar characteristics and send their data to the base station (BS). The BS is aware of the location of each node and coordi- nates the schedules for data gathering. The generated data are highly correlated and intermediary nodes are capable of data aggregation. The goal of this thesis is to gen- erate an energy e cient data gathering schedule that distributes the communication load, minimizes the communication cost, and maximizes the network lifetime. A data gathering schedule contains directed aggregation trees with corresponding frequencies (rounds of usage). We propose an energy e cient data gathering (EEDG) algorithm, which is based on maximum ow, to generate an aggregation tree with a xed number of frequencies. The simulation results show that our proposed technique signi cantly extends the network lifetime in comparison with traditional methods, especially for a sparse WSN deployment. The EEDG algorithm is implemented in the TinyOS platform to show the realization and e ectiveness of the proposed algorithm.