A hardware-model-based simulator for Quantum Key Distribution systems employing photon polarization encoding

The forthcoming release of quantum computers, as also the advances of classical computers, may threaten most of the cryptosystems used today, in particular the asymmetric ones. Although new and more secure cryptosystems can be developed, the best long-term solution is quantum cryptography. Nowadays, the first Quantum Key Distribution (QKD) systems are already on the market. QKD is employed in symmetric encryption schemes, like “one-time pad”, and it ideally permits to obtain unconditionally secure communications, exploiting properties of quantum mechanics such as the no-cloning theorem. The private key is exchanged as a stream of single-photon pulses encoding the quantum information (qubits). In this thesis a simulation framework for QKD systems based on polarization encoding is presented. The objective of this simulator is to help in designing new QKD systems, estimating how the fundamental performance parameters, such as Quantum Bit Error Rate (QBER) and Secure Key Rate, vary with the communication length or with the employed hardware. The simulator environment consists of a MATLAB simulator for the optical system and of a Verilog-A simulator for Single-Photon Avalanche Diode (SPAD), the photon detectors most used in these systems. The optical simulator is based on the use of coherent states, semi-classical states that well fit with the description of the coherent light emitted by the attenuated lasers, the most employed type of light source. The propagation of the state is obtained thanks to the relations that link the creation (or similarly annihilation) operators at the input and output of an optical device through its Jones matrix. The Verilog-A code has been developed with a focus on InGaAs/InP SPADs, the best suited to work at near infrared wavelengths, but it can also simulate other types of SPADs. Integrating the model of the diode in the quenching circuit one plans to use, the total Dark Count Rate (DCR) as a function of the working conditions can be obtained, as also other useful parameters. The model of dark counts and afterpulsing is one of the crucial parts of the code; a peculiar strategy for afterpulsing is presented, based on the use of the afterpulsing probability. With the data obtained by these two simulations, the QBER and the Secure Key Rate of the system can be easily obtained in a post-processing phase. To validate this methodology, a real BB84 system based on polarization encoding is simulated. If the parameters of the devices employed in the setup are used, QBER is slightly underestimated with respect to the experimental one, and consequently the Secure Key Rate is lightly overestimated. The simulation results are more than satisfactory, considering that this is only a starting point for the development of a simulator for QKD systems. It is expected that an improvement of both the optical simulator and the Verilog-A code for SPADs and their integration in a unified software infrastructure would permit to obtain a complete and reliable design tool for QKD systems engineering.