Hybrid Quantum Neural Network Approach for Rapid Response to Cyber Attacks

Abstract

Cyber attacks on data servers and critical infrastructure of fifth-generation (5G) wireless communication networks are increasing day by day. Therefore, an intelligent, reliable and cost effictive security of sixth-generation (6G) wireless communication network is inevitable. Quantum technology is one of the enabling technologies of future computing, as well as of the 6G. In the event of a cyberattack, rapid response is critical to minimize the risk of data loss and denial-of-service (DoS) attacks. Quantum computers are a new generation of computers that are expected to be much faster than classical computers. In this study, we have proposed and implemented a hybrid quantum neural network (HQNN) model. The proposed HQNN model was trained and tested on a dataset from the Australian center for cybersecurity. Simulation results show that the proposed HQNN model is faster in training as well as in the testing phase compared to classical neural networks. Moreover, our proposed model helps to overcome the underutilization of centeral processing unit (CPU) resources. CPU utilization of our proposed HQNN model is 95-100%, while that of the classical model is only 35-75% during the training and testing of the dataset.