Kalman filter-based phase estimation of continuous-variable quantum key distribution without sending local oscillator

Phase estimation is one of the crucial procedures in continuous-variable quantum key distribution without sending local oscillator (LLO-CVQKD), aiming at compensating the phase drift in post-processing. To decrease the phase noise caused by the phase drift and measurement error, we propose Kalman filter-based phase noise estimation of LLO-CVQKD, which involves the measurements and filtering through exploiting the small amplitude reference pulses, achieving the accurate phase noise estimation. Simulation results indicate that Kalman filter-based phase noise estimation is capable of reducing measurement error and phase noise, while increasing the secret key rate by 7 times.