Theoretical Physics

11 December 2012
Time: 15:00 to 16:00
Location: 8.60

Katherine Brown (Louisiana)

Dynamic decoupling in the presence of imperfect waveplates

Dynamic decoupling uses pulses to decoupling a system from its environment. In the case of quantum optics waveplates are used to time reverse the system environment interaction, reducing the dephasing.

In previous work we explored using CPMG (Carr-Purcell-Meiboom-Gill) dynamic decoupling to prevent dephasing when sending polarization qubits through an optical fibre [1]. We showed that it was possible to achieve over 98% fidelity in 10km fibre using 610 perfect waveplates. However, in reality it is impossible to build perfect waveplates, in this presentation we will consider two different forms of imperfect waveplates, and the impact these have on the effectiveness of dynamic decoupling.

Firstly, we will look at finite width pulses, and show that if we shape pulses then it is still possible to achieve high fidelities provided we increase the number of waveplates in our system. In this case we find that CPMG dynamic decoupling scheme remains effective, and beats several alternative schemes.

Secondly, we will explore what happens if our pulses are subject to small random errors with zero mean. In this case we find that standard schemes such as CPMG dynamic decoupling schemes fail, and composite pulse schemes such as KDD (Knill dynamic decoupling) are essential to achieve more than 50% fidelity.

Given these results we will look at how we could realistically use dynamic decoupling to send information for quantum computing given the limitations of the threshold and gates available in quantum error correcting codes.

[1] B.R. Bardhan et al., Phys. Rev. A 85 022340 (2012)



<- Back to: External Seminars