Localization Approach for Tracking the Mobile Nodes Using FA Based ANN in Subterranean Wireless Sensor Networks

Localization is an essential approach in the branch of wireless sensor networks that have been introduced crucial research interest in academic circles and research association. Main aim is to create the localization scheme to enhance the localization accuracy. With the aim is to support long battery life for network devices with low rate, low power consumption and minimum resource requirements. The ZigBee network formation is carried out in the proposed model. The position of the mobile node is evaluated depend upon received signal strength indicator by means of firefly algorithm based artificial neural network (FA-ANN) technique. RSSI data for mobile points are calculated in advance and they maintained in fingerprint database. The finding phase size and principal component analysis is calculated for reducing the size of RSSI fingerprints. The affinity propagation clustering technique is affiliated to decrease the higher position error and improve the effectiveness of the location prediction. The proposed trained FA neural network is based on the clustered RSSI value for accurate localization. Finally, trained FA based neural network is utilized to find the accurate position of the mobile node with minimal consumption of mobile node energy. Thus the hybrid approach, the localization error is reduced and node prediction is achieved in a faster rate. The implementation output of the presented system shows that can be provide localization accuracy of 95% and significantly improves the prediction speed in terms of minimum location time.