The vulnerability of quantum key distribution (QKD) systems to low-intensity attacks that masquerade as natural optical channel noise is analyzed. An approach is proposed for detecting hidden anomalies based on multiparametric telemetry analysis using the Isolation Forest machine learning algorithm. A pipeline for generating computed temporal features is developed, which enriches the system state vector with dynamic signal characteristics. Using a software testbed, simulation modeling of the BB84 protocol is performed to evaluate the effectiveness of attack detection within the equipment’s permissible operating range. The superiority of the proposed method over the classical threshold-based approach is experimentally demonstrated.
The vulnerability of quantum key distribution (QKD) systems to low-intensity attacks that masquerade as natural optical channel noise is analyzed. An approach is proposed for detecting hidden anomalies based on multiparametric telemetry analysis using the Isolation Forest machine learning algorithm. A pipeline for generating computed temporal features is developed, which enriches the system state vector with dynamic signal characteristics. Using a software testbed, simulation modeling of the BB84 protocol is performed to evaluate the effectiveness of attack detection within the equipment’s permissible operating range. The superiority of the proposed method over the classical threshold-based approach is experimentally demonstrated.
In this paper, the results of research of spectral features of several passive fiber optic components are presented. There are spectra of isolators, circulators, spectral filters, and WDM, that are used in quantum key distribution (QKD) systems, and contribute to the protection of these systems against active optical probing attacks. We also provide an estimate of the degree of isolation required to protect the QKD system from Trojan-horse attacks. Based on the isolation requirements and measurement data we analyzed the efficiency of the QKD system protection through the use passive isolating components. As an example, we have shown the achievement of the required isolation level (150 dB) in the considered spectral range in the system presented in the article.
The rapid development of quantum computing threatens the traditional cryptographic schemes and suggests that the best alternative solution for ensuring the security of key exchange is Quantum Key Distribution (QKD).Despite the information-theoretic security guarantee, the practical implementations of QKD present vulnerabilities that were derived from hardware flaws and the exploitation of the quantum channel.This thesis is intended to address security vulnerabilities by analyzing and proposing an innovative architecture for the identification and mitigation of quantum attacks, via classification techniques of Quantum Machine Learning and the use of the metrics of Quantitative Risk Analysis.The study begins with the concepts of quantum information, the functionality of basic QKD protocols (BB84, B92, E91) and the mathematical modeling of Implementation-level attacks and side-channel attacks in quantum systems.Subsequently, the operational processes of Quantum Support Vector Machine (QSVM) and the Variational Quantum Classifier (VQC) are analyzed during the training phase.Based on the theoretical background of this study, QML-IDPS architecture is proposed.Threat detection and classification are performed using the QSVM and VQC classifiers, while risk metrics are utilized and fed into a classical Intrusion Prevention System (IPS) to support the final decision-making process for the prevention of quantum threats.The final key validated by the threat assessment mechanisms, is then used for data encryption with a classical cryptographic algorithm, such as AES-256.The proper functionality of QML-IDPS architecture is verified with the mathematical Intercept-Resend Use Case Scenario, which basically proves the successful classification and prevention of system anomalies.In conclusion, it is confirmed that the proposed QML-IDPS architecture demonstrates that Quantum Machine Learning combined with Quantitative Risk Analysis can enhance the security and reliability of QKD infrastructures.
Quantum computing advancements pose a growing threat to conventional cryptographic systems based on com- putational hardness assumptions such as RSA and Elliptic Curve Cryptography. Quantum Key Distribution (QKD) has emerged as a theoretically secure alternative by leveraging the fundamental principles of quantum mechanics to establish shared secret keys between communicating parties. However, practical QKD deployments remain vulnerable to channel noise, hardware imperfections, side-channel attacks, and sophisticated intercept- resend strategies that exploit implementation weaknesses. Tra- ditional QKD security analysis relies on static Quantum Bit Error Rate (QBER) thresholds, which are insufficient for dis- tinguishing malicious eavesdropping from natural environmental disturbances. Recent advancements in artificial intelligence and machine learning have introduced intelligent anomaly detec- tion techniques capable of analyzing complex quantum channel behaviour in real time. This paper presents a comprehensive survey of AI-driven threat detection and response systems for quantum cryptographic networks. Various techniques including Random Forest classifiers, deep learning models, Support Vector Machines, and reinforcement learning-based adaptive response mechanisms are discussed along with their strengths and limita- tions. The paper also highlights major research challenges such as false-positive detection, computational overhead, hardware dependency, and limited integration of automated mitigation strategies. Furthermore, a conceptual multimodal framework combining IBM Qiskit-based BB84 simulation, machine learning- based threat classification, and automated Privacy Amplification is discussed as a promising solution for securing next-generation quantum communication infrastructures.
Quantum Key Distribution (QKD) is a potential defense against future quantum attacks that can break traditional encryption algorithms, such as RSA and ECC. The current work is based on the development of the BB84 protocol with the features of utilizing quantum mechanics and eavesdropping detection through signal disturbance. A simulation of the BB84 protocol was built using polarization encoding over a 10 km single-mode optical fiber using OptiSystem software. The system consists of standard telecom equipment, such as a continuous wave laser at a frequency of 193.1 THz, polarization controllers, and single-photon detectors. Both secure and insecure cases were simulated, where an attacker (Eve) had used an intercept-resend attack. The outcome showed significant Bit Error Rate (BER) and Quantum Bit Error Rate (QBER) increases during attacks, thus proving the protocol's vulnerability to interference. In addition, a special intrusion detection system based on BER was used to identify such anomalies in real-time. The simulation illustrates the feasibility of using QKD in today's optical networks to secure communications.
The rapid advancement of quantum computing poses a significant threat to conventional cryptographic systems that rely on computational complexity for security [3]. Quan-tum Key Distribution (QKD) offers a secure communication mechanism based on the principles of quantum mechanics, enabling the detection of unauthorized interception attempts [1]. However, practical QKD deployments remain vulnerable to channel noise, hardware imperfections, and sophisticated eavesdropping attacks [5], [6]. This paper presents an AI-driven threat detection and response framework that combines BB84 Quantum Key Distribution, IBM Qiskit simulation, and machine learning-based anomaly detection [9]. A high-fidelity quantum simulator is developed using IBM Qiskit and AerSimulator to model realistic quantum communication channels with depo-larizing noise. A Random Forest classifier trained on 50,000 simulated transmission sequences identifies malicious activities by analyzing Quantum Bit Error Rate (QBER), noise levels, sifted key lengths, and engineered heuristic features [7]. Experimental analysis demonstrates high detection accuracy while maintaining real-time operational capabilities. The proposed system further incorporates automated mitigation techniques including Privacy Amplification and E91 protocol migration to enhance quantum network resilience [2]. The results demonstrate that integrating artificial intelligence with quantum cryptography significantly improves threat detection, response efficiency, and overall communication security.