Active noise suppression of semiconductor spin qubits

Semiconductor quantum dots are one of the promising solid-state platforms for quantum computing. Data are encoded in the spin states of individual electrons in quantum dots and their quantum mechanical coherence must be preserved in the noisy environment of solid-state devices. To improve the coherence without relying on material engineering, here we discuss an approach using a feedback control algorithm implemented on an FPGA. We use an electron spin as a quantum sensor for magnetic noise and estimate the fluctuation of the spin resonance frequency with the Bayes' theorem. The microwave signal for spin control is then adjusted to compensate for the noise effect, thereby suppressing the noise-induced dephasing effect. We observed that the spin coherence time in a GaAs quantum dot was improved from 30 ns to 760 ns, enabling high-fidelity quantum control. Feedback control techniques based on a similar experimental setup will pave the way for practical quantum computation. (author)