arXiv:2610.29050
Buffer-Qubit-Based Quantum Circuit Allocation Method Against Crosstalk Attacks
Haruki Nakayama, Satoyuki Tsukano, Satoshi Kono, Keita Matsumoto, Toshio Mori, Takumi Oikawa, Ryo Uchida
Abstract
Quantum Multi-Programming (QMP) enables the simultaneous execution of quantum circuits from multiple users on a single quantum chip in cloud quantum computing services. However, QMP is vulnerable to crosstalk attacks, in which a malicious user deliberately induces crosstalk to disrupt other users' computations. As a countermeasure, prior work has proposed the use of buffer qubits, which remain idle and are allocated between quantum circuits to suppress the effect of crosstalk. Nevertheless, to the best of our knowledge, such countermeasures have not been publicly reported in production quantum cloud software. In this paper, we extend the QMP of OQTOPUS, open-source software that is already in production use for operating real quantum computers, with a buffer-qubit-based circuit allocation method. Experiments on a real device characterize the attack effect in terms of energy relaxation and dephasing, showing that observable degradation appears in dephasing characteristics of qubits adjacent to the attacker circuit. Based on this result, we release the proposed method as part of OQTOPUS and confirm that the buffer qubits are correctly reserved around the allocated circuits.
AI Summary
The paper addresses crosstalk attacks in quantum multiprogramming, where a malicious user’s circuit degrades another user’s simultaneous computation. It implements buffer-qubit allocation in OQTOPUS, an open-source quantum-cloud stack already used in production. The method extends an existing subgraph-isomorphism allocator with constraints excluding reserved qubits. After each circuit is mapped, unused qubits within a specified crosstalk range are reserved before allocating subsequent circuits. To select that range, experiments on a superconducting device compare victim-qubit $T_1$ and $T_2^*$ during repeated attacker CNOTs against time-matched idle controls. Each data point uses seven independent runs of 10,000 shots. Approximate 95% confidence intervals for differences between fitted parameters indicate detectable $T_2^*$ degradation only on adjacent qubits: for example, one qubit’s value falls from $2.28\pm0.07$ to $1.95\pm0.05\,\mu\mathrm{s}$ for $\ket{+}$. No $T_1$ difference is detected, nor any $T_2^*$ difference at the tested distances of two or three edges. The implementation therefore reserves adjacent qubits. An allocation test places 100 randomly generated two- to five-qubit circuits into 12 batches on a 64-qubit square lattice without violating buffer constraints. The illustrated batch uses 30 qubits for circuits and 34 (53.1%) as buffers. This verifies reservation behavior, not end-to-end mitigation of victim-output errors. The stated remaining work includes quantifying throughput overhead across circuit-size distributions, evaluating victim-circuit outputs, and testing attacks with longer-range effects.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- Accepted at the 2nd Workshop on Quantum Computing Security, Privacy and Resilience (Q-SEC), IEEE QCE 2026. The accepted manuscript is available here; the final published version will be available on IEEE Xplore