Techniques for Improving Numerical Secret Key Rates of Quantum Key Distribution Protocols

Abstract

Quantum Key Distribution (QKD) is a method for producing secure keys between two parties whose security does not rely on computational assumptions potentially breakable by quantum computers. However, physical constraints, such as noise, imperfect devices, and the necessity of finite resources, limit the rate at which experimental implementations of QKD can produce key, and in some cases prevent the generation of secure key altogether. Determination of key generation rate is facilitated by a numerical framework for general QKD protocols, upon which we propose improvements. With protocols used in actual QKD experiments as examples, we present and demonstrate various methods for improving key rate calculations in the regime of a finite number of signals sent. Our methods include a block diagonal optimization for the state shared by the two parties, modifying constraints on acceptance of candidate states, optimizing security parameter distribution, and optimizing the grouping of data into blocks for time-binned data. Through these improvements, we are able to both reduce the computational cost of key rate calculations in our numerical framework and improve key rates in the case of a finite number of sent signals.