Search

We design a deep learning assisted end to end free-space optical ON-OFF Keying communications system, and demonstrate its efficacy in significantly lowering the bit error rate, while requiring no knowledge of the channel state information.

We present a mathematical method of correcting non-physical results from polarimetry measurements based on convex optimization. We show the experimental nonphysical measurements and the corresponding corrected results.

We use a deep learning neural network to successfully indicate when an eavesdropper is present in a free-space optics ON-OFF keying communication scheme, in which the bit stream has passed through turbulence.

Reconstruction of two-qubit systems typically performed using 36 coincidence measurements. We trained an ensemble of CNN & XGBoost ML models, employed imputation to demonstrate high-fidelity state estimation when multiple measurements are missing.