Hybrid Neuro-Fuzzy Clustering and Modified ECC Based Dynamic Key Management in WSN

Abstract

Wireless Sensor Network becomes a puzzling field of research once it involves security issues. Important design goal of WSN is to keep up the energy required for communication to reduce the packet loss or packet drop and quick energy depletion and reduction in node performance due to severe threats across the network which may lead to delay in packet delivery. The resource constrained device nodes consume additional energy and is conjointly liable to threats. Moreover, to provide better lifetime and security services, sensor nodes need strong cryptographic measures. It has been proved that the asymmetrical cryptosystems are more efficient in security achievement as compared to symmetrical ones. Also, clustering is an important technique used in WSN that results in maximization of network lifetime thereby reducing the energy consumption. Hence, a hybrid combination of Neuro-Fuzzy based clustering technique and modified Elliptic Curve Cryptography (ECC) based key management is proposed to identify the cluster heads and ensure secure transmission of data between the cluster head and sink nodes. Since ECC provides higher level of security by using a key which is smaller in size, the proposed scheme also implements mutual authentication between base station and sink node using hashing technique.