Title: Intrusion detection and attack mitigation in WSN using deep Kronecker fuzzy Zeiler and Fergus network

Authors: Bhushan Chaudhari; Sachin Kamble; Madhuri Patil

Addresses: Department of Information Technology, SVKM's Institute of Technology, Dhule, Maharashtra 424001, India ' Department of Information Technology, SVKM's Institute of Technology, Dhule, Maharashtra 424001, India ' Department of Information Technology, SVKM's Institute of Technology, Dhule, Maharashtra 424001, India

Abstract: Wireless sensor network (WSN) is a group of wireless sensors employed to monitor various environmental conditions in the collaborative mode without relying on a few underlying model support. In this approach, a deep Kronecker fuzzy Zeiler and Fergus network (DKFZFNet) is developed for intrusion detection (ID) and attack mitigation in WSN. Quantile normalisation (QN) is employed for data pre-processing. The feature fusion (FF) is performed based on the Shepard convolutional neural network (ShCNN) with angular separation distance (ASD). Then, the bootstrapping method is employed to carry out data augmentation (DA). Next, ID is carried out utilising DKFZFNet, which is an incorporation of deep Kronecker network (DKN), fuzzy concept and Zeiler and Fergus network (ZFNet). Lastly, attack mitigation is conducted to remove malicious nodes and to transfer other data. Furthermore, DKFZFNet obtained 91.7%, 92% and 91.8% of precision, recall, and F1-score.

Keywords: quantile normalisation; QN; Shepard convolutional neural network; ShCNN; intrusion detection; ID; Zeiler and Fergus Network; ZFNet; deep Kronecker network; DKN.

DOI: 10.1504/IJAMECHS.2025.144589

International Journal of Advanced Mechatronic Systems, 2025 Vol.12 No.1, pp.12 - 29

Received: 12 Jun 2024
Accepted: 28 Aug 2024
Published online: 23 Feb 2025 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article