模擬量子密鑰分發在竊聽下的表現

Abstract

不同機制被提出來提升量子密鑰分發系統的表現，同時伴隨著不同對竊聽者資訊量假設的安全證明。在我們的量子密鑰分發裡，我們建立差分相移和相位編碼BB84的模組以及分光攻擊和攔截回送攻擊的竊聽模組去看竊聽下系統表現和三方資訊量關係。相位編碼BB84雖然碼率只有差分相移量子密鑰分發的一半，但其更高的安全閥值使其更適合長距離傳輸，同時也更能對抗分光攻擊和攔截回送攻擊。此外我們發現在無錯誤因素的差分相移量子密鑰分發系統在攔截回送攻擊下最高的洩漏資訊是0.15比特。此外，我們建立誘騙態模組並去使用這兩種竊聽攻擊去分析多光子訊號的洩漏資訊量。在攔截回送攻擊下取得的資訊量跟使用誘騙態的量子密鑰分發系統的訊號態平均光子數較無關聯。相反地，可藉由分光攻擊從高平均光子數的訊號態取得更多資訊。竊聽者可在平均光子數越大情況下取得超過0.15比特。整體來說，此研究著重透過量子密鑰分發模擬器去看不同機制下受到不同竊聽攻擊的系統表現和三方資訊量，並進而探討系統安全性。

Various of protocols are proposed for better quantum key distribution (QKD) performance accompanied various assumption on the eavesdropper's information in the corresponding security proof. In our QKD simulator, differential-phase-shift QKD (DPS-QKD) and phase-encoding BB84 (PEBB84) were built while photon-number-splitting (PNS) and intercept-and-resend (IAR) attacks are also built to see the system performance and tripartite relation under eavesdropping. Compared with DPS-QKD, PEBB84 can transmit longer distance with its higher security threshold while its key rate is only half of the DPS-QKD. PEBB84 is also resistant to PNS attack and IAR attack. In addition, we found the maximum leakage of information, 0.15 bit, within security threshold by applying IAR attack in an error-free DPS-QKD. Furthermore, we built decoy method module and apply these two hacking strategy to analyze the leaking information for the multi-photon signals. The information acquired from the IAR attack are less correlated to the signal mean photon number (MPN) in DPS-QKD using decoy method. In contrast, the higher signal MPN leaks more information by using PNS attack. By using PNS attack, eavesdropper can acquire more information than 0.15 bit as the MPN gets higher. Overall, this study focused on the system performance and tripartite information for different protocols under eavesdropping by using the QKD simulator and discuss the system security in advance.