Detection of Malicious Nodes in Wireless Sensor Networks based on Features Using Neural Network Computing Approach
B. Rajasekaran1, C. Arun2
1B. Rajasekaran, Research Scholar, Department of ECE, St. Peter’s Institute of Higher Education & Research, Chennai (Tamil Nadu), India.
2Dr. C. Arun, Professor, Department of ECE, R.M.K College of Engineering & Technology, Chennai (Tamil Nadu), India.
Manuscript received on 14 December 2018 | Revised Manuscript received on 26 December 2018 | Manuscript Published on 24 January 2019 | PP: 188-192 | Volume-7 Issue-4S2 December 2018 | Retrieval Number: Es2058017519/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The detection of malicious or hidden nodes in Wireless Sensor Network (WSN) is important for improving the performance of the WSN. In this paper, distance metric and probabilistic features are extracted from each individual node in WSN with respect to its surrounding nodes. These individual extracted features are given to the input of the classification algorithm. This paper uses feed forward back propagation neural networks for training and testing the individual nodes using the extracted node features. The concert of this hidden node identification in WSN using metric and probabilistic features based classification algorithm analyzed energy consumption, throughput and delay.
Keywords: Malicious, Nodes, Metric, Features, Neural Networks.
Scope of the Article: Neural Information Processing