Quantum Machine Learning-Driven Cyber Security Architecture for Real-Time Intrusion Detection in Distributed Computing Systems

Abstract

The active development of distributed computing systems and cloud-based networks has greatly exposed the modern network to advanced cyber-attacks thus necessitating a quick and timely intrusion detection system that is intelligent and capable of identifying malicious activities with low latency and high accuracy. The major weaknesses of traditional machine learning-based intrusion detection systems (IDS) are poor scalability, inefficiency in processing high dimensional security data, increased false positive rates, and lower detection performance with changing attack patterns. To overcome these difficulties, this study suggests a Quantum Support Vector Machine (QSVM)-based building on Quantum machine learning (Quantum ML) powered cyber security system with real-time intrusion detection in distributed computing infrastructure. The suggested framework will combine network traffic gathering, preprocessing, quantum feature encoding, and QSVM-based attack identification to improve the effectiveness of cybersecurity intelligence and threat detection. Experimental validation and performance analysis of the system in different cyberattack situations are done using the CICIDS2017 and UNSW-NB15 benchmark intrusion detection datasets. The applicability of the offered architecture is tested with the help of essential classification measures such as Accuracy, Precision, Recall, and F1-score which show that the proposed architecture has a better intrusion detection potential than the traditional machine learning techniques. The experimental findings were that the QSVM-based framework yields a high classification accuracy, detection reliability, false alarms, and real-time threat identification which leads to design of scalable, intelligent and next-generation quantum-enhanced cybersecurity systems in distributed computing settings.