A Novel Hybrid Reconfigurable Architecture for Prediction of Side Channel Attacks with Its Countermeasure Mechanism

Abstract

The Internet of Things (IoT) has pushed everyone‘s normal life zone to their comfort zone by making them use embedded IoT devices for controlling and monitoring their daily gadgets. IoT devices find their applications in health care, agriculture, industrial automation, and even vehicles. Since IoT involves numerous devices’ data sharing, which causes network traffic and makes them vulnerable to security breaches, especially Side-Channel Attacks (SCA), it creates demand for an intelligent framework. As of now, many secured cryptoengines are integrated into embedded chips, but still, SCAs play a very vital role in IoT networks. As a result, the paper proposes a novel reconfigurable architecture as Field Programmable Gate Arrays (FPGA) that integrates deep learning models for effective SCA prediction as well as a countermeasure mechanism based on chaotic maps. Finally, the experimentation results prove the excellence of the proposed framework in terms of prediction accuracy (99%), sensitivity (98.9%), and specificity (98.9%) in comparison with LSTM, ELM SVM, RF, NB, and ANN. The results also demonstrated that the proposed framework requires only 50% of the total energy for network operation.