Bayesian Kriging Regressive Similarity Index Deep Highway Transfer Learning Network for Efficient Cyber Attack Detection in Wireless Networks

Abstract

Wireless Network has integrated sensing technology and wireless communication
for gathering the sensed data. Security attacks have major concern in wireless sensor
networks (WSNs). Many researchers carried out on efficient cyber attack detection in
WSN. But, the attack detection accuracy was not improved and time complexity was not
minimized by existing methods. To address these issues, Bayesian Kriging Regressive
Similarity Index Deep Highway Transfer Learning Network (BKRSIDHTLN) Method is
introduced to perform efficient cyber attack detection with high accuracy and minimum
time. Deep transfer learning involves adapting a pre-trained highway network for attack
detection. Deep Highway Network collects the number of data samples and their features
as input. After that, the feature selection is performed using Bayesian Kriging Regression
to select the more relevant features.