Practical issues in quantum-key-distribution postprocessing

Quantum key distribution (QKD) is a secure key generation method between two distant parties by wisely exploiting properties of quantum mechanics. In QKD, experimental measurement outcomes on quantum states are transformed by the two parties to a secret key. This transformation is composed of many logical steps (as guided by security proofs), which together will ultimately determine the length of the final secret key and its security. We detail the procedure for performing such classical postprocessing taking into account practical concerns (including the finite-size effect and authentication and encryption for classical communications). This procedure is directly applicable to realistic QKD experiments and thus serves as a recipe that specifies what postprocessing operations are needed and what the security level is for certain lengths of the keys. Our result is applicable to the BB84 protocol with a single or entangled photon source.

Figure 1

Flow chart of our data postprocessing procedure [58].

Figure 2

Lower bound for the key rate as a function of the raw key length; parameters used: $e_{\mathit{bx}}=e_{\mathit{bz}}=4\%$ and the error correction efficiency is 100%. The three curves correspond to three different values of failure probability $\varepsilon$.

Figure 3

Minimum raw key length to yield a positive key length as a function of $\varepsilon$; parameters used: $e_{\mathit{bx}}=e_{\mathit{bz}}=4\%$ and the error correction efficiency is 100%.

Figure 4

Effect of the bias ratio on the final key length; parameters used: $e_{\mathit{bx}}=e_{\mathit{bz}}=4\%$, target failure probability $\varepsilon ={10}^{-7}$, the raw key length is ${10}^6$, and the error correction efficiency is 100%.

Figure 5

Plot of the optimal bias ratio vs. the raw key length; parameters used: $e_{\mathit{bx}}=e_{\mathit{bz}}=4\%$, target failure probability $\varepsilon ={10}^{-7}$, and the error correction efficiency is 100%.