Abstract
Quantum communications are based on the law of physics for information security and the implications for this form of future information security enabled by quantum science has to be studied. Physics-based vulnerabilities may exist due to the inherent physics properties and behavior of quantum technologies such as Quantum Key Distribution (QKD), thus resulting in new threats that may emerge with attackers exploiting the physics-based vulnerabilities. There were many studies and experiments done to demonstrate the threat of physics-based attacks on quantum links. However, there is a lack of a framework that provides a common language to communicate about the threats and type of adversaries being dealt with for physics-based attacks. This paper is a review of physics-based attacks that were being investigated and attempt to initialize a framework based on the attack objectives and methodologies, referencing the concept from the well-established MITRE ATT&CK, therefore pioneering the classification of Indicator of Compromises (IoCs) for physics-based attacks. This paper will then pave the way for future work in the development of a forensic tool for the different classification of IoCs, with the methods of evidence collections and possible points of extractions for analysis being further investigated.
Introduction: The Gap in Quantum Security
Quantum Key Distribution (QKD) is often heralded as the pinnacle of secure communication, theoretically "unhackable" because any attempt at eavesdropping disturbs the quantum state, alerting the legitimate users. However, a critical gap exists between the mathematical perfection of quantum protocols and the physical reality of their implementation. Physical hardware—such as lasers, fiber optics, and avalanche photodiodes—possesses inherent vulnerabilities that do not exist in the digital realm.
Currently, the quantum research community lacks a standardized language to describe these threats. While traditional cybersecurity relies on frameworks like MITRE ATT&CK to categorize adversary behavior, quantum security has remained largely fragmented. This paper proposes a unified framework to classify physics-based attacks, bridging the gap between quantum physics and operational cybersecurity to enable the development of future forensic tools.
Adapting the MITRE ATT&CK Framework
To provide a structured approach to quantum threats, this research adapts the industry-standard MITRE ATT&CK methodology. The framework maps the relationships between various components of a physical attack, including the adversary's tools, the specific methodologies used, and the ultimate objectives.
By standardizing these relationships, the framework distinguishes between Tactics (the attacker's high-level goal) and Techniques (the specific physical method employed). This categorization allows security professionals to move beyond simply demonstrating that an attack is possible to understanding the operational requirements and mitigations for each threat.
Categorizing Physics-Based Attacks
The framework groups attacks into three primary entry points based on the physical component targeted:
- The Environment: Attacks targeting the transmission medium, such as fiber tapping or intercepting free-space quantum links.
- The Source of Photons: Exploiting vulnerabilities in the light source, such as photon-number splitting or phase re-mapping.
- The Detectors of Photons: Manipulating the receiving hardware, most notably through "detector blinding" attacks.
In a detector blinding attack, an adversary uses a bright continuous-wave laser to force the single-photon detectors into a classical regime. This allows the attacker to intercept and resend keys without triggering the usual quantum disturbances that would reveal their presence.
Identifying Indicators of Compromise (IoCs)
A core contribution of this work is the identification of measurable physical parameters—Indicators of Compromise (IoCs)—that can be used to detect active attacks. Unlike digital IoCs (like IP addresses or file hashes), quantum IoCs are derived from real-time sensor data and hardware performance metrics:
- Quantum Bit Error Rate (QBER): While some error is expected, the paper suggests that a QBER sustained between 8% and 12% is a strong indicator of an anomaly or active eavesdropping.
- Received Power ($P_{recv}$): Monitoring the power levels on the link is vital. Received power $\ge 10^{-15}$ W suggests high noise or a potential attack, while power $\ge 10^{-3}$ W can indicate an attempt to thermally blind the detectors.
- Detector Recovery Metrics: Changes in Afterpulse Probability or Deadtime (the time a detector needs to recover after a click) can reveal sophisticated "after-gate" or blinding attacks.
- Photocurrent Monitoring: A monitor readout exceeding a threshold of 8100 (for specific sampled values) serves as a definitive indicator of an active physical intervention.
Conclusion and Future Work
This framework represents an initial but essential step toward operational security in the quantum era. By moving the focus from theoretical vulnerabilities to measurable indicators, the research paves the way for automated forensic tools. Future work will focus on developing these tools to extract evidence from quantum links in real-time, allowing for incident response and post-breach investigations in environments where security is guaranteed by the laws of physics, but challenged by the limitations of hardware.
Summary updated