Artificial Neural Networks for developing localization framework in Wireless Sensor Networks

Abstract

Intelligent sensing, communication, and computing is leading the development of Wireless Sensor Networks (WSNs) and its applications. Artificial Neural Network (ANNs) is being extensively used to develop localization framework in WSN. In this paper, we present a comparative analysis of some conjugate gradient based Feedforward Artificial Neural Network (FFANNs) for developing localization framework in wireless sensor networks (WSNs). Localization is the process by which the sensor nodes in the network can identify their own location in the overall network. Localisation in WSN plays a crucial role in implementing myriad of applications such as healthcare management, disaster management, environment management, and agriculture management. Artificial neural network based localization framework are gaining importance due to faster speed of convergence and low cost of computation. We present analysis of conjugate gradient backpropagation with Powell-Beale restarts, conjugate gradient backpropagation with Fletcher-Reeves update, conjugate gradient backpropagation with Polak-Ribiere updates, one-step secant, and scaled conjugate gradient backpropagation training algorithms in this paper. Comprehensive evaluation of these training algorithms is done in perfect simulation scenario. The proposed method effectively demonstrates that conjugate gradient FFANNs based sensor motes can be designed for developing cost-effective localization framework.