Scholarlink

<p>Quantum Key Distribution (QKD) ensures secure communication through quantum mechanics, yet optimizing its efficiency and security on noisy intermediate-scale quantum (NISQ) devices remains challenging. This study introduces a novel VQA-based QKD framework that leverages Variational Quantum Algorithms (VQAs) to enhance key generation by integrating hybrid quantum-classical optimization with machine learning techniques. Implemented on IBM Quantum’s Falcon processor (6 qubits) and Qiskit Aer simulator (up to 8 qubits), our approach achieves a computational time of 120&#xa0;ms for a 256-bit key, a secret key rate of 0.95 bits per sifted bit (f = 1.1), and key retention of 95% under 10% Quantum Bit Error Rate (QBER), with 92% retention against 5% photon-number-splitting attacks. Compared to BB84 (170&#xa0;ms, 0.85 bits/sifted bit, 85% retention), our method offers superior key rates and resilience, though it trades off speed for enhanced security and scalability (up to 8 qubits). We demonstrate its effectiveness through optimization benchmarks (Rosenbrock, TSP) and QKD-specific metrics, addressing scalability for larger key lengths and real-world noise.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Quantum Key Distribution Using Variational Quantum Algorithms

  • 期刊论文

  • 发表日期：

    2026年6月19日

查看全文
  • Salim Salmi,
  • Abdelkrim Khartoch,
  • Lahcen Oughdir,
  • Anass El Affar

摘要

Quantum Key Distribution (QKD) ensures secure communication through quantum mechanics, yet optimizing its efficiency and security on noisy intermediate-scale quantum (NISQ) devices remains challenging. This study introduces a novel VQA-based QKD framework that leverages Variational Quantum Algorithms (VQAs) to enhance key generation by integrating hybrid quantum-classical optimization with machine learning techniques. Implemented on IBM Quantum’s Falcon processor (6 qubits) and Qiskit Aer simulator (up to 8 qubits), our approach achieves a computational time of 120 ms for a 256-bit key, a secret key rate of 0.95 bits per sifted bit (f = 1.1), and key retention of 95% under 10% Quantum Bit Error Rate (QBER), with 92% retention against 5% photon-number-splitting attacks. Compared to BB84 (170 ms, 0.85 bits/sifted bit, 85% retention), our method offers superior key rates and resilience, though it trades off speed for enhanced security and scalability (up to 8 qubits). We demonstrate its effectiveness through optimization benchmarks (Rosenbrock, TSP) and QKD-specific metrics, addressing scalability for larger key lengths and real-world noise.