We perform a proof-of-principle demonstration of the measurement-device-independent quantum key distribution (MDI-QKD) protocol using weak coherent states and polarization-encoded qubits over two optical fiber links of 8.5 km each. Each link was independently stabilized against polarization drifts using a full-polarization control system employing two wavelength-multiplexed control channels. A linear-optics-based polarization Bell-state analyzer was built into the intermediate station, Charlie, which is connected to both Alice and Bob via the optical fiber links.Using decoy-states, a lower bound for the secret-key generation rate of 1.04  10 -6 bits/pulse is computed.

A real-time polarization control system employing two non-orthogonal reference signals multiplexed in either time or wavelength with the data signal is presented. It is shown, theoretically and experimentally, that complete control of multiple polarization states can be attained employing polarization controllers in closed-loop configuration. Experimental results on the wavelength multiplexing setup show that negligible added penalties, corresponding to an average added optical Quantum Bit Error Rate of 0.044%, can be achieved with response times smaller than 10 ms, without significant introduction of noise counts in the quantum channel.

No abstract

Please refer to published version for the most recent bibliographic citation information. If a published version is known of, the repository item page linked to above, will contain details on accessing it.

We demonstrate an active polarization drift compensation scheme for optical fibres employed in a quantum key distribution experiment with polarization encoded qubits. The quantum signals are wavelength multiplexed in one fibre along with two classical optical side channels that provide the control information for the polarization compensation scheme. This setup allows us to continuously track any polarization change without the need to interrupt the key exchange. The results obtained show that fast polarization rotations of the order of 40 π rad/s are effectively compensated. We demonstrate that our setup allows continuous quantum key distribution even in a fibre stressed by random polarization fluctuations. Our results pave the way for Bell-state measurements using only linear optics with parties separated by long distance optical fibres.

No abstract

We experimentally demonstrate stable and user-adjustable single-photon interference in a 1 km long fiber-optic Mach-Zehnder interferometer, using an active phase control system with the feedback provided by a classical laser. We are able to continuously tune the single-photon phase difference between the interferometer arms using a phase modulator, which is synchronized with the gate window of the single-photon detectors. The phase control system employs a piezoelectric fiber stretcher to stabilize the phase drift in the interferometer. A single-photon net visibility of 0.97 is obtained, yielding future possibilities for experimental realizations of quantum repeaters in optical fibers and violation of Bell's inequalities using genuine energy-time entanglement.

Shear horizontally (SH) guided waves, generated by periodic permanent magnet arrays, have been used previously in nondestructive evaluation of metal plates and pipes. When an SH guided wave interacts with a defect or a change in sample thickness, the incident SH wave may undergo mode conversion. Analysis of mode conversion is complicated, due to the interference of several propagating modes in the received signal that can often temporally overlap. This paper proposes a mode selection technique to help understand the interaction of SH guided waves with changes in sample thickness. Using an understanding of the propagation characteristics of the guided waves, SH guided waves are sequentially generated and detected on both surfaces of the plate, capturing four distinct waveforms. By superposition of the detected signals, symmetric modes can be clearly separated from antisymmetric modes in the processed received signals. For this method to work well, the transducers used should have very similar responses and be precisely positioned on exactly opposite positions either side of the plate. Finite element simulations are also performed, mirroring the experimental measurements, and the results correlate well with the experimental observations made on an 8-mm-thick plate with a region of simulated wall thinning machined into the sample.