Faster Recurrent Convolutional Neural Network with Edge Computing Based Malware Detection in Industrial Internet of Things
Ammar Hameed Shnain, Kalyan Gattupalli, C. Nalini, Poovendran Alagarsundaram, Rajeshwari Patil
Abstract
The smart factory environment has become Industrial Internet of Things (1IoT) environment because of integration and openness of the environment. It is always easier to launch epidemic cyber threats in smart factories through malware most of the time, although the differentiation of exact real-time malware in these areas is complex. In order to address this issue, this research work introduces an advanced Faster Recurrent Convolutional Neural Network (Faster R-CNN) integrated with an edge computing based malware identification model. This system successfully discerns the multiple types of malwares by transmitting the huge IIoT traffic information about smart factories to the edge servers. The proposed detection system of malware consists of various layers as the edge device layer, the edge layer, lastly, the cloud layer. For edge-based deep learning it employs four base operations. The model obtained accuracy of 93.77<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, recall of 95.87%, precision of 86.66%, and f1-score of 91.03% when compared to CNN and Long Short-Term Memory (LSTM).