Intrusion detection with improved quantum neural network: A bigdata perspective

An Intrusion Detection System (IDS) is a pivotal component of cybersecurity infrastructure which is designed to protect networks, systems, and data from unauthorized access, misuse, or malicious activities. Its primary function is to monitor network or system activities in real-time that analyze incoming traffic and identify any anomalous behavior or patterns that deviate from established norms or signatures of known attacks. Both conventional ML and DL-based IDS may be subject to adversarial attacks, where malicious actors deliberately operate input data to evade detection. Consequently, a proposed solution involves the development of an ID model based on Improved Quantum Neural Network and LinkNet (IQNN-LinkNet) architecture aimed at addressing the aforementioned challenges. This paper adopts a methodical process encompassing pre-processing, handling the bigdata, and intrusion detection. The input data is first subjected to pre-processing via the Improved min-max normalization technique. Subsequently, the bigdata is handled via MRF which also incorporates feature extraction procedures. These extracted features are then utilized as input for a hybrid detection model that integrates IQNN and LinkNet classifiers. Extensive analyses are used to validate the effectiveness of the suggested IQNN-LinkNet model through simulation and experimental evaluations. Eventually, this paper presents a robust and confirmed model for intrusion detection which highlights the potential of the IQNN-LinkNet model particularly in bigdata applications.