Title: Energy efficient data aggregation in wireless sensor networks using neural networks

Authors: Fereshteh Khorasani; Hamid Reza Naji

Addresses: Science and Research Branch, IAU, Kerman 93630, Iran ' College of Electrical and Computer Engineering, Graduate University of Advanced Technology, Kerman 93630, Iran

Abstract: Energy efficiency is very important issue in wireless sensor networks (WSNs). In WSN, sensors are distributed in different places, where they can be exposed to contact with the environment. Data aggregation, eliminating of data redundancy and improve the accuracy of the collected data are essential points for these networks. This research has been suggested some algorithms such as MEDA, LMTBPN, RBDA and RGDA. The first algorithm is based on the moment estimation method and the other three algorithms aggregate the data based on backward propagation, radial basis and general regression. These algorithms use a three-layer neural network. Input layer neurons are located in members of each cluster while the hidden layer neurons are located in cluster heads and output layer neurons are located in base station. Simulation results show that the performance of data aggregation is improved and also energy consumption of the network is reduced.

Keywords: wireless sensor networks; WSNs; data aggregation; energy efficient; radial basis neural networks; back propagation neural networks.

DOI: 10.1504/IJSNET.2017.084207

International Journal of Sensor Networks, 2017 Vol.24 No.1, pp.26 - 42

Received: 17 Feb 2015
Accepted: 04 Sep 2015
Published online: 21 May 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article