关于交换 ASAP 量子中继链中的噪声:精确解析、分布及紧致近似
On noise in swap ASAP repeater chains: exact analytics, distributions and tight approximations
摘要 Abstract
损耗是量子网络中纠缠分发的主要瓶颈之一,可通过量子中继器的实现加以克服。最基本的量子中继链形式是交换 ASAP 中继链。在这样的中继链中,基本链路以概率方式生成,并在两个相邻链路生成后立即确定性地进行交换。由于每个纠缠态在等待交换的过程中会经历去相干,因此链两端节点之间的纠缠态保真度变为随机变量。完全表征随中继链增长的(平均)保真度仍然是一个开放问题。在这里,我们对间距相等的中继器进行了解析研究,在最多 25 段的情况下找到了保真度所有矩的精确解析公式。我们通过提供一个生成函数的一般解法来获得这些公式;该生成函数的 Maclaurin 级数的第 n 项给出了 n 段保真度的矩。我们将这种方法推广到全局截断策略——一种以更长的纠缠传输时间为代价提高保真度的方法——从而消除了蒙特卡洛模拟的需求,实现了截断参数的快速优化。此外,我们还找到了平均保真度的简单指数逼近公式,并且对于最多 10 段的情况,得到了交付保真度的完整分布。我们利用这一点解析计算了密钥率,包括有无分桶方法的情况。
Losses are one of the main bottlenecks for the distribution of entanglement in quantum networks, which can be overcome by the implementation of quantum repeaters. The most basic form of a quantum repeater chain is the swap ASAP repeater chain. In such a repeater chain, elementary links are probabilistically generated and deterministically swapped as soon as two adjacent links have been generated. As each entangled state is waiting to be swapped, decoherence is experienced, turning the fidelity of the entangled state between the end nodes of the chain into a random variable. Fully characterizing the (average) fidelity as the repeater chain grows is still an open problem. Here, we analytically investigate the case of equally-spaced repeaters, where we find exact analytic formulae for all moments of the fidelity up to 25 segments. We obtain these formulae by providing a general solution in terms of a generating function; a function whose n'th term in its Maclaurin series yields the moments of the fidelity for n segments. We generalize this approach as well to a global cut-off policy -- a method for increasing fidelity at the cost of longer entanglement delivery times -- allowing for fast optimization of the cut-off parameter by eliminating the need for Monte Carlo simulation. We furthermore find simple approximations of the average fidelity that are exponentially tight, and, for up to 10 segments, the full distribution of the delivered fidelity. We use this to analytically calculate the secret-key rate, both with and without binning methods.