Dresden 2020 – wissenschaftliches Programm
Die DPG-Frühjahrstagung in Dresden musste abgesagt werden! Lesen Sie mehr ...
Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
HL: Fachverband Halbleiterphysik
HL 70: Spin phenomena in semiconductors
HL 70.6: Vortrag
Donnerstag, 19. März 2020, 16:15–16:30, POT 251
Single-shot spin readout of a spin qubit in silicon measured using a neural network — •Tom Struck1, Javed Lindner1, Arne Hollman1, Floyd Schauer2, Dominique Bougeard2, and Lars Schreiber1 — 1JARA-FIT Institute Quantum Information, RWTH Aachen University, Germany — 2Institut für Experimentelle und Angewandte Physik, Universität Regensburg, Regensburg, Germany
Spin qubits have proven to be a promising candidate for quantum computing, with the field advancing quickly in recent years. One challenge is the fast and reliable qubit state detection. A popular method used for single-shot read-out of an electron spin qubit is spin-selective tunnelling [1,2] combined with charge readout by a single electron transistor (SET). Depending on the qubit's spin state, the SET senses an approx. 60 pA current rise with a statistically distributed timing. Reliable and fast interpretation of the current trace is particularly challenging. Here we compare the traditional current threshold analysis with current trace classification by a neural network. This spin-state classifier is highly noise resilient, allowing a decrease in measurement time by a factor of five. The calculation time per trace is only 30 μs, rendering the implementation of fast-feedback loops possible. [1] J. M. Elzerman et al., Nature 430, 431 (2004) [2] Arne Hollman et al. arXiv:1907.04146v1