Distributed Denial of Service (DDoS) attacks pose a significant threat to the security of networking systems, as they can cause widespread disruption and even bring down entire distributed systems platforms. In this paper, we propose an approach called the DEQSVC that leverages quantum machine learning techniques to detect DDoS attacks with high accuracy. The DEQSVC integrates the most efficient dimensionality reduction techniques, a robust feature map method, and an efficient kernel estimation technique to improve data encoding, learning process, and detection accuracy. To evaluate the performance of the proposed DEQSVC, we conducted simulations using the Qiskit platform and executed the approach on an IBM quantum computer. Our results demonstrate that the DEQSVC outperforms several benchmark algorithms commonly used in intrusion detection systems. Specifically, the DEQSVC achieves a detection accuracy of 99.49, indicating its effectiveness as a highly accurate and efficient method for detecting DDoS attacks.
The notion of Cyber Physical Systems is based on using recent computing, communication, and control methods to design and operate intelligent and autonomous systems that can provide using innovative technologies. The existence of several critical applications within the scope of cyber physical systems results in many security and privacy concerns. On the other hand, the distributive nature of these CPS increases security risks. In addition, certain CPS, such as medical ones, generate and process sensitive data regularly, hence, this data must be protected at all levels of generation, processing, and transmission. In this paper, we present a taxonomy based analysis for the state of the art work on security issues in CPS. We identify four types of analysis for security issues in CPS: Modeling, Detection, Prevention, and Response. In addition, we identified six applications of CPS where security is relevant: eHealth and medical, smart grid and power related, vehicular technologies, industrial control and manufacturing, autonomous systems and UAVs, and finally IoT related issues. Then we mapped existing works in the literature into these categories.
5 August 2024 in IEEE Transactions on Artificial Intelligence
This study proposes a Quantum Reinforcement Learning (QRL) approach for robotic applications, which incorporates a Grover-based Autonomous Quantum Agent (GAQA) and a quantum environment represented as a Quantum TicTacToe (QTTT) game. The QTTT environment is a quantum circuit of qubits in their superposition states, manipulated by the agent through quantum gates to establish a goal state. By utilizing amplitude estimation and Grover search techniques, the proposed ReLAQA enhances the probability amplitudes of the actions taken, which results in reducing the number of observed states required to reach a solution. Empirical results substantiate the quantum advantages of the proposed GAQA in Reinforcement Learning (RL) tasks by observing fewer states of 6300, outperforming classical agents. Therefore, signifying its potential to enhance complex problem-solving in robotics.
1 October 2025 in IEEE Transactions on Artificial Intelligence
In this work, we present the design and plan of Quantum machine learning (QML) course in a computer science (CS) University program at senior undergraduate level / first year graduate level. Based on our survey, there is a lack of detailed design and assessment plan for the delivery of QML course. In this paper we have presented the QML course design with week by week details of QML concepts and hands on activities that are covered in the course. We also present how this QML course can be assessed from CS program learning outcomes perspective.