Real-Time Quantum Computing Anomaly Detection Model on Vulnerabilities in Government Systems

The growth of vulnerabilities in network traffic data has been increasing yearly, coupled with cyberattacks on systems. The attack surface increases when zero-day exploits are not detected at quicker speeds. Vulnerabilities can be a multitude of things, such as backdoors, unpatched software, and shell code. Attackers target systems with less practical protections, and companies cannot detect movement within systems. This praxis provides a quantum computing anomaly detection model that uses deep-learning techniques to detect anomalous behavior and classify vulnerabilities in real-time. This detection model uses quantum conditions to allow for faster detection. This praxis uses the UNSW-NB15 dataset, which has 2 million records of data traffic that simulate modern normal activities and synthetic contemporary attack behaviors. Some of the findings within this research are that the record total duration and source-to-destination packet counts are features that, when placed under deep-learning techniques under quantum conditions, can accurately and efficiently find vulnerabilities in the dataset. The praxis researches a convolutional neural network algorithm under quantum computing conditions, outperforms logistic regression, k-means clustering, and Gaussian mixture naïve Bayes under quantum conditions. In real-time, the model can accurately and efficiently detect vulnerabilities in data streams.

Author

• Singh, Gurvinder

Language

• English

Keyword

• classical models
• security
• convolutional neural network
• quantum computing
• cybersecurity
• anomaly detection

Date created

• 2025

Type of Work

• Dissertation

License

• All rights reserved

Rights statement

• In Copyright

GW Unit

• Cybersecurity Analytics

Degree

• D.Engr.

Advisor

• Etemadi, Amir
• Stepanov, Artem

Committee Member(s)

• Wehr, Evan

Persistent URL

•  https://scholarspace.library.gwu.edu/etd/5d86p123m