Data Processing in Continuous-Variable Quantum Key Distribution Under Composable Finite-Size Security

Continuous-variable quantum key distribution (CV-QKD) uses amplitude and phase modulation of light, in order to establish secure communications between two remote parties. The laws of quantum mechanics ensure the theoretical security of the protocol, in spite of the noise and losses of the communication channel. In practice, however, the resulting secret key rate depends not only on these two factors, but also on a series of data-processing steps, needed for transforming shared correlations into a final secret binary string.

In this work, we investigate the operation of three Gaussian-modulated coherent-state (GMCS) CV-QKD protocols: the homodyne detection, heterodyne detection and the continuous-variable measurement-device-independent (CV-MDI) protocol. We propose a comprehensive strategy covering their entire course, starting from the preparation and transmission of quantum states, until the extraction of a shared secret key. We also provide rigorous security proofs, considering optimal eavesdropper strategies and incorporating the composable framework under finite-size effects, which offers the highest level of security. In addition, we present results, where we explore the performance of different quantities of interest in the high signal-to-noise regime and identify intervals of parameters, where communications are regarded as secure. This is achieved under the assistance of our self-developed open-source Python library, which we use to simulate the stage of quantum communications and, afterwards, to process the resulting data via the stages of parameter estimation, information reconciliation and privacy amplification. Here, short-range communications are of particular interest. To enhance data processing in this high signal-to-noise ratio setting, we have combined an appropriate data preprocessing scheme with the use of high-rate, non-binary low-density parity-check (LDPC) codes. This allows us to examine the performance of short-range CV-QKD in practical implementations and optimize the parameters connected to the aforementioned steps.