Seminarium Optyczne

In my talk I will summarize recent results obtained with my group, in which we combinequantum description of continuously monitored atomic sensors with efficient techniques ofBayesian statistical inference, in order to track fluctuating signals in atomic magnetometry. Iwill start by discussing a linear-classical model in which it is sufficient to use KalmanFiltering (KF) methods to optimally estimate signals encoded in the light pumping theatomic ensemble. However, I will demonstrate how this scenario may be directly generalizedwith use of the Extended Kalman Filter (EKF) to track magnetic fields, i.e. the Larmorfrequency. Importantly, I will afterwards move onto the nonlinear-quantum setting in which,thanks to the phenomena of measurement back-action and continuous spin-squeezing, theclassical limits imposed on precision may be spectacularly breached. On one hand, I willshow how to determine then the correct ultimate noise-induced limits by employing tools ofquantum information theory. On the other, I will demonstrate how to achieve such quantum-enhanced resolutions by resorting to the effective quantum stochastic description of thedynamics involving measurement-based feedback and the EKF