Synchronization techniques for polarization-encoded quantum key distribution systems

• Main Author: Felisberto, Alexandra da Costa
• Publication Date: 2024
• Format: Master thesis
• Language: eng
• Source: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Download full: http://hdl.handle.net/10773/44651
• topic: Quantum communications; Quantum key distribution; Symbol synchronization; Symbol synchronization recovery; Efficiency; Dark counts

In today’s world, a vast amount of sensitive information - from personal communications to financial transactions and government data - is constantly moving across platforms and networks. As a result, ensuring the security and privacy of this information has become a priority for individuals, organizations and governments alike. Quantum key distribution (QKD) offers a promising solution for secure communications, allowing the generation of identical keys between two entities separated in space. Although QKD systems have evolved considerably in recent years, there is still a need for optimization in various subsystems to make these technologies more efficient. One such area in need of improvement is the synchronization process, which is critical to the proper functioning of QKD systems. Taking this into account, this thesis investigates an effective synchronization technique with low computational complexity for QKD systems, aiming to strike a balance between efficiency and simplicity/cost, using the Qubit4Sync method as a model. The technique studied eliminates the need to prepare reference signals, as is the case with current systems operating in gated mode, and recovers the period using only the qubits similar to those exchanged during the QKD protocol. In this context, simulations were carried out using the Qubit4Sync method and the setup was adapted to work without the reference signal, with the detectors operating in free-running mode. The data from the setup used was processed and the simulation confirmed that once the dark counts are greater than the quibts detected, it becomes more difficult to recover the correct period. In the laboratory, the method was tested with various conditions of efficiency and dead-time of the single photon detectors. Due to the high optical noise of the channel, the recovery of the period was unsuccessful, as it was always equal to the value of the dead-time. After comparing the simulation with the practical results, it turned out that the noise in practice was higher than predicted. When investigating the setup, it was found that the extinction rate imposed by the Mach-Zehnder Modulator (MZM) was lower than expected. This may have contributed to the transmitter generating more optical noise than initially anticipated. Consequently, this factor may explain the practical results, since the detection rate of the detector is saturated at its dead-time value.