QKD BB84协议：确认量子量是否可能？还是两个将军问题适用？
在量子密钥分布协议BB84中，爱丽丝将X Qubits发送给BOB。没有收据确认。这不能在量子通道上完成。如果我需要确认，则必须通过经典渠道发送。但这意味着我彼此独立发送了两条消息，这使我们陷入了两个将军问题。 我能正确理解吗？还是我缺少可以帮助我与TCP类似的传输的东西？
In the Quantum Key Distribution protocol BB84, Alice sends x qubits to Bob. There is no acknowledgement of receipt. This cannot be done over the quantum channel. If I need an acknowledgement, it must be sent via a classical channel. But that means that I am sending two messages independently of each other, which brings us to the two generals problem.
Do I understand this correctly? Or am I missing something that could help me confirm the transmission similarly to TCP?
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（2）
从Laudenbach Paper
带有高斯调制的连续变量量子密钥分布 - 实际实现理论：因此，仅当您具有身份验证的经典渠道时，BB84才能起作用。我猜这个假设结束了两个将军的问题。我对吗？
From Laudenbach paper
Continuous-Variable Quantum Key Distribution with Gaussian Modulation - The Theory of Practical Implementations:So BB84 works only if you have an authenticate classical channel. I guess this assumption ends the Two Generals' Problem. Am I right?
是的，您是正确的，鲍勃需要告知爱丽丝量子位是哪些Qubits曾经和未收到的。
实际上，由于损失，鲍勃从未收到过爱丽丝发送的大多数量他。例如，典型的纤维每公里损失0.2 dB，这意味着超过50公里的纤维将损失90％的纤维。
鲍勃使用身份验证的公共渠道通知爱丽丝收到哪些量子位，爱丽丝可以从中确定未收到哪些量子位。
有时，这是在经典后处理中的基础筛选步骤的一部分，Bob向爱丽丝宣布了他用于每个量子的测量基础。
在现实生活实现中，这是在基本筛选之前的一个单独的一步，该步骤在FPGA中实现（而基本筛选通常在通用CPU上的软件中实现）。
这留下了一个非平凡的问题：爱丽丝和鲍勃如何识别Qubits？与经典数据包不同，Qubits不能具有例如携带序列编号的标头字段。
相反，爱丽丝和鲍勃具有非常准确且非常精确的同步时钟。爱丽丝和鲍勃使用此时钟来识别量子位。爱丽丝知道她会在TimeStamps T0，T1，T2，T3上发送QUBIT其中t0'-t0是爱丽丝和鲍勃之间的传播延迟。
因此，如果鲍勃（Bob）在时间时收到量子，他知道这是爱丽丝（Alice）在TN发送的Qubit。
而且，如果鲍勃（Bob）在TN'时未能收到Qubit，他就会知道Alice在TN派遣的Qubit丢失了。
Yes, you are correct, Bob needs to inform Alice which qubits were and were not received.
In fact, most qubits that Alice sends are never received by Bob due to loss. For example, typical fiber has 0.2 dB loss per kilometer, which means that over 50 kilometer of fiber 90% of the qubits will be lost.
Bob uses the authenticated public channel to inform Alice which qubits were received, from which Alice can determine which qubits were not received.
Sometimes this is done as part of the basis sifting step in the classical post-processing, where Bob announces to Alice which measurement basis he used for each qubit.
More often, in real-life implementations, this is a separate step before basis sifting, implemented in firmware in FPGAs (whereas basis sifting is typically implemented in software on a general-purpose CPU).
That leaves one non-trivial question: how do Alice and Bob identify the qubits? Unlike classical packets, qubits cannot have a header field that could, for example, carry a sequence number.
Instead, Alice and Bob have very accurate and very precisely synchronized clocks. Alice and Bob use this clock to identify the qubits. Alice knows she sends her qubits at time-stamps t0, t1, t2, t3, ... Bob knows he should expect to receive qubits at time-stamps t0', t1', t2', t3', .... (where t0' - t0 is the propagation delay between Alice and Bob).
So, if Bob receives a qubit at time tn', he knows it is the qubit sent by Alice at tn.
And, if Bob fails to receive a qubit at time tn', he know that the qubit sent by Alice at tn was lost.