Implementation of 400 Gbps quantum noise stream cipher encryption for 1520 km fiber transmission using end-to-end deep learning

Area of Science:

• Quantum communication security
• Optical fiber networks
• Deep learning applications

Background:

• Optical fiber backbone networks require robust security measures.
• Existing quantum noise stream cipher (QNSC) schemes do not meet the high rates of 400G networks.
• Physical layer security is critical in the era of big data.

Purpose of the Study:

• To enhance the security of optical fiber communications using deep learning.
• To develop an end-to-end quantum noise stream cipher (E2E-QNSC) scheme.
• To achieve secure transmission rates compatible with 400G networks.

Main Methods:

• Integration of deep learning into QNSC.
• Development of an end-to-end quantum noise stream cipher (E2E-QNSC) encrypting 16-QAM into E2E-65536QAM/QNSC.
• Experimental validation of the proposed scheme.

Main Results:

• Demonstrated secure optical communication at a single-channel rate of 400 Gbps.
• Achieved a total capacity of 8.4 Tbps and a transmission distance of 1520 km.
• Maintained a detection failure probability (DFP) > 0.9999 even in extreme conditions.

Conclusions:

• The E2E-QNSC scheme significantly enhances optical fiber communication security.
• The proposed method sets a new record for the rate-distance product in QNSC secure transmission.
• Deep learning integration enables QNSC to meet the demands of modern high-speed networks.