| 研究概要 |
In this project, we designed a quantum repeater consisting of three qubits. A nitrogen vacancy center is used as the transmitter qubit, a superconducting flux qubit to execute logic operations, and a bismuth vacancy center in silicon as a long-term memory storage qubit. We have studied the need to balance the number of NV centers. More NV centers allows a stronger interaction with the flux qubit, and therefore a faster gate rate. However, large numbers of centers result in large stray thermal excitations (the Boltzmann distribution), damaging the fidelity of the resulting Bell pairs. For a repeater link, gate speed is less important and fidelity is more important, resulting in a different tradeoff compared to using a similar structure for a quantum computer.
In the course of examining this hardware architecture, we simulated a purification scheme known as "double selection", in which three Bell pairs are purified down to one pair. Double selection, when it succeeds, eliminates one round trip from the necessary classical communications and therefore improves the net performance.
Our simulations demonstrate that the utility of double selection is rather constrained. It works well when Bell pairs are of good fidelity but local gate errors are high. A significant drawback is the need for having three Bell pairs available at the same time. Repeater links succeed in entanglement creation with only low probability, this limits the rate and which we can attempt double selection. On the whole, double selection is found to be useful only in very narrow circumstances.
|
