Quantum Keystroke Logging in Quantum Cryptography

• Quantum keystroke logging is a side-channel technique that exploits physical imperfections in quantum devices to covertly record user inputs.
• It leverages methods such as detector blinding, injection locking, and state preparation logging to extract sensitive quantum state information.
• Countermeasures include enhanced monitoring, robust source isolation, and security protocols incorporating realistic device behaviors.

Quantum keystroke logging refers to techniques by which an adversary or malicious entity surreptitiously records user inputs—typically logical operations, password entries, or authentication actions—on devices and protocols employing quantum information. Quantum keystroke logging exploits vulnerabilities arising from physical imperfections in quantum hardware, preparation procedures in quantum communication systems, and quantum mechanical side channels, yielding information about user operations without raising detectable alarms. This paradigm is relevant in contexts ranging from quantum key distribution (QKD) systems to quantum password checking and GKP-based encoded communication.

1. Principles and Paradigms of Quantum Keystroke Logging

Quantum keystroke logging fundamentally differs from its classical counterpart. In classical systems, keystrokes are intercepted via software or hardware routines that monitor input streams or hardware-level events. Quantum keystroke logging subsumes several mechanisms whereby adversaries exploit quantum device imperfections or inherent quantum properties to infer user input/actions on quantum states.

One principal method is side-channel exploitation: non-idealities in the physical implementation of quantum protocols enable the extraction of logged information (for example, QKD detector blinding (Gerhardt et al., 2010)) without disturbing protocol parameters monitored by legitimate parties. Another paradigm leverages the provider’s privileged position in state preparation (e.g., GKP state provider (Chang, 17 Sep 2025)), enabling quantum phase estimation techniques to deduce applied user logical operations by measuring induced geometric phases on ancillary systems.

A summary table of quantum keystroke logging modes in contemporary protocols:

Mode	Principle	Main Exploit
Detector side-channel	Exploit physical non-idealities (blinding, thresholds)	Record measurement events undetectably
Source injection-locking	Induce side channel via injected photons	Log polarization-dependent keys
State preparation logging	Interrogate geometric phase/fingerprint	Extract logical qubit manipulations

2. Detector-Side Keystroke Logging in Quantum Key Distribution

The "faked-state attack" (Gerhardt et al., 2010) provides a prototypical example: Avalanche photodiodes (APDs) in modern QKD systems are susceptible to blinding by strong illumination. When Eve floods the detector array with continuous-wave light above a threshold, the APDs revert to classical linear response, losing their quantum sensitivity. Eve then injects bright trigger pulses engineered to cause detector "clicks" only on intended elements, precisely controlling Bob's measurement outcomes via timing and polarization.

Mathematically, Eve's control is described by:

$P_{\text{click}}(I) = \Theta(I - I_{\text{th}})$

where $I_{\text{th}}$ is the threshold current and $\Theta$ is the Heaviside step function. In the blinded regime, Eve's pulses guarantee $P_{\text{click}} = 1$ on targeted detectors.

Critically, this attack leaves the quantum bit error ratio (QBER) and other monitored parameters unchanged. Eve achieves perfect correlation with Bob's raw key, creating a scenario indistinguishable from normal operation to Alice and Bob. This is analogous to covert keystroke logging in classical systems: the entire sequence of key events ("clicks") is perfectly and invisibly recorded by the adversary.

3. Source-Side Logging via Injection Locking

A distinct attack demonstrated in measurement-device-independent QKD protocols is injection locking (Pang et al., 2019). Here, Eve exploits controllable spectral dynamics of Alice’s semiconductor laser. By injecting near off-resonance photons (polarization-matched to Alice's encoding) at finely tuned powers, Eve induces the slave laser to lock its frequency to the "master" (Eve's own source) only when her polarization matches Alice's state.

The photon dynamics, modeled by the modified Lang-Kobayashi equations,

$$\frac{dE(t)}{dt} = \frac{1}{2}(1 + i\alpha) G_n \Delta N(t) E(t) + \kappa E_x e^{i\nu t}$$

allow spectral filtering (e.g., via Fabry–Pérot cavities) to distinguish the occurrence of locking events. These events correspond to successful decoding of key bits—the source effectively "logs" the keystroke in the emission spectrum. The reported success rate is 60% of raw keys, with an error rate of 6.1% at optimal injection power.

This mode of quantum keystroke logging arises not from detector vulnerabilities, but from side channels in active quantum sources. A plausible implication is that securing QKD against keystroke logging requires robust monitoring and physical isolation at both source and detector levels.

4. State Preparation Logging in GKP-Based Quantum Communication

The most direct realization of quantum keystroke logging is proposed in the context of GKP codeword-based communication (Chang, 17 Sep 2025). A communication provider who controls encoded state preparation can extract user-applied logical operations (displacements) via an adapted phase estimation protocol, even though codewords are not eigenstates of the relevant logical operators and operations are executed in "one-shot" fashion.

The operational principle is that geometric phases acquired by closed trajectories in phase space—embodied by sequences like

$$\hat{D}(-\beta)\hat{D}(\alpha)\hat{D}(\beta)= e^{-2\mathrm{i}\theta}\hat{D}(\alpha), \quad \theta=\Im(\alpha^* \beta)$$

induce an effective Pauli-$Z$ rotation on an ancilla qubit coupled to the system. By measuring the ancilla, the provider retrieves $\theta$, revealing the displacement (user input), without disturbing the codeword.

The quantum Fourier transform (QFT) in conventional phase estimation is simplified: an auxiliary oscillator and cross-Kerr nonlinearities enact a "bosonic QFT" mapping, extracting phase information into a tractable probability distribution. This avoids the need for repeated applications, enabling a scalable and hardware-efficient quantum keystroke logger.

5. Quantum Password Checking and Logging Resistance

Quantum password checking protocols (Garcia-Escartin et al., 2015) encode passwords in symmetric quantum states of dimension too small to permit full extraction of classical encoded bits. The core of the protocol is the preparation of a state

$$|\psi_j\rangle = \frac{1}{\sqrt{D}} \sum_{l=0}^{D-1} e^{i 2\pi j l / D} |l\rangle$$

with $j$ determined by $H(p \| r_i)$, where $p$ is the password and $r_i$ a random round-dependent string.

Authentication proceeds via SWAP tests; the security is guaranteed by quantum information principles—specifically, Holevo’s and Nayak’s bounds:

• The adversary’s probability of extracting $k$ bits from the state is exponentially small when $D \ll 2^n$ ($n$ is the number of bits encoded).
• Repeated randomization by $r_i$ enforces unique states per round, so interception (quantum keystroke logging) of state transmissions yields negligible information on the password.

A plausible implication is that cryptographic protocols leveraging low-dimensional quantum systems and comparison-based authentication can be made robust against quantum keystroke logging—provided that passwords are sufficiently strong and system dimensions are chosen to minimize information leakage even across multiple recorded rounds.

6. Security Implications and Countermeasures

Quantum keystroke logging constitutes a significant class of side-channel vulnerabilities in quantum cryptography and authentication:

• Protocols proven secure under idealized quantum assumptions are often susceptible due to neglected device imperfections and side channels (e.g., APD blinding (Gerhardt et al., 2010), source injection locking (Pang et al., 2019)).
• Theoretical advances in communication encoding (e.g., GKP protocols (Chang, 17 Sep 2025)) demonstrate structural vulnerabilities when trusted preparation is not enforced.

Necessary countermeasures include:

• Real-time physical monitoring of detector states (photocurrent, temperature, bias), watchdog detectors, and randomized internal verification,
• Enhanced source isolation and active monitoring, design of lasers resistant to external influence,
• Security proofs and protocol designs incorporating realistic device models and verified implementation behavior,
• Integration of quantum bounds (Holevo, Nayak) into protocol analyses for schemes relying on low-dimensional quantum state encoding.

A plausible implication is that the evolution of quantum protocol engineering must increasingly prioritize mitigation of quantum keystroke logging in both device and protocol design—extending beyond theoretical cryptographic security to comprehensive physical security integration.

7. Future Directions and Research Opportunities

Research in quantum keystroke logging exposes unresolved and emerging challenges:

• Expanding protocol security proofs to realistically include physical side channels (e.g., device-independent, measurement-device-independent QKD),
• Investigating novel hardware countermeasures—from highly isolated devices to dynamically randomized operational states,
• Developing protocols that, even given provider control, preclude phase-based or geometric side channel exploitation,
• Quantitatively benchmarking keystroke logging resistance in complex multi-user quantum systems,
• Extending Holevo-type bounds to composite scenarios involving multiple adversarial recordings or imperfect state preparations.

Future work will likely focus on providing sufficient and necessary conditions for logging resistance, engineering quantum devices with certified information leakage profiles, and integrating cross-disciplinary approaches to physical, algorithmic, and protocol-level security against quantum keystroke logging attacks.