2026-09-19T23:14:37Zhttps://dspace.mit.edu/server/oai/requestoai:dspace.mit.edu:1721.1/1400242022-02-08T03:33:38Zcom_1721.1_7582com_1721.1_7581col_1721.1_131022
Oliver, William D.
Lienhard, Benjamin
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science
2022-02-07T15:19:36Z
2022-02-07T15:19:36Z
2021-09
2021-09-21T19:31:00.835Z
https://hdl.handle.net/1721.1/140024
Quantum computers hold the promise to solve specific problems significantly faster than classical computers. However, to realize a practical quantum computer, the quantum processor’s constituent components, their control, and their readout must be very well-calibrated. Over the last few decades, infrastructure and protocols have been developed to operate small-scale quantum processors efficiently. However, the operation of medium- to large-scale quantum processors presents new engineering challenges. Among those challenges are efficient and high-fidelity multi-qubit control and readout. In particular, qubit-state readout is a significant error source in contemporary superconducting quantum processors. The fidelity of dispersive qubit-state readout depends on the readout pulse shape and frequency as well as the resulting qubit-state discriminator. For a single qubit, fast and high-fidelity readout is achieved with minor changes to the rising and falling edge of a rectangular microwave pulse and a linear matched filter discriminator. However, in resource-efficient, frequency-multiplexed readout of multiple qubits, optimizing the readout pulse shape and discriminator becomes a computationally intensive task.
In this thesis, control and readout hardware and software tools for multiple superconducting qubits are developed. First, I discuss the principles to engineer microwave packages for multiple qubits. I designed and engineered a novel multiqubit package to enable efficient qubit control and readout and minimize errors due to interactions between the quantum processor and its immediate environment. Second, I demonstrate deep machine learning techniques to improve frequency-multiplexed superconducting qubit readout pulse shapes and discrimination for a five-qubit system. Compared with currently employed readout methods, these novel techniques reduce the required measurement time, the readout resonator reset, and the discrimination error rate by about 20% each. The developed readout techniques are a significant step towards efficient implementations of near-term quantum algorithms based on iterative optimization and quantum error correction protocols necessary for future universal quantum processors.
Ph.D.
Massachusetts Institute of Technology
In Copyright - Educational Use Permitted
Copyright MIT
http://rightsstatements.org/page/InC-EDU/1.0/
Machine Learning Assisted Superconducting Qubit Readout
Thesis
application/pdf
Doctoral
Doctor of Philosophy
Publication
unknown
unknown
<Publication xmlns="https://www.openaire.eu/cerif-profile/1.1/" id="8c174bc1-7caa-4779-bf35-0733c3af6606">
<Type xmlns="https://www.openaire.eu/cerif-profile/vocab/COAR_Publication_Types">http://purl.org/coar/resource_type/c_1843</Type>
<Title>Machine Learning Assisted Superconducting Qubit Readout</Title>
<PublishedIn>
<Publication>
</Publication>
</PublishedIn>
<PublicationDate>2021-09</PublicationDate>
<Authors>
<Author>
<DisplayName>Lienhard, Benjamin</DisplayName>
<Affiliation>
<OrgUnit>
</OrgUnit>
</Affiliation>
</Author>
</Authors>
<Editors>
</Editors>
<Publishers>
<Publisher>
<DisplayName>Massachusetts Institute of Technology</DisplayName>
<OrgUnit />
</Publisher>
</Publishers>
<License>http://rightsstatements.org/page/InC-EDU/1.0/</License>
<Abstract>Quantum computers hold the promise to solve specific problems significantly faster than classical computers. However, to realize a practical quantum computer, the quantum processor’s constituent components, their control, and their readout must be very well-calibrated. Over the last few decades, infrastructure and protocols have been developed to operate small-scale quantum processors efficiently. However, the operation of medium- to large-scale quantum processors presents new engineering challenges. Among those challenges are efficient and high-fidelity multi-qubit control and readout. In particular, qubit-state readout is a significant error source in contemporary superconducting quantum processors. The fidelity of dispersive qubit-state readout depends on the readout pulse shape and frequency as well as the resulting qubit-state discriminator. For a single qubit, fast and high-fidelity readout is achieved with minor changes to the rising and falling edge of a rectangular microwave pulse and a linear matched filter discriminator. However, in resource-efficient, frequency-multiplexed readout of multiple qubits, optimizing the readout pulse shape and discriminator becomes a computationally intensive task.
In this thesis, control and readout hardware and software tools for multiple superconducting qubits are developed. First, I discuss the principles to engineer microwave packages for multiple qubits. I designed and engineered a novel multiqubit package to enable efficient qubit control and readout and minimize errors due to interactions between the quantum processor and its immediate environment. Second, I demonstrate deep machine learning techniques to improve frequency-multiplexed superconducting qubit readout pulse shapes and discrimination for a five-qubit system. Compared with currently employed readout methods, these novel techniques reduce the required measurement time, the readout resonator reset, and the discrimination error rate by about 20% each. The developed readout techniques are a significant step towards efficient implementations of near-term quantum algorithms based on iterative optimization and quantum error correction protocols necessary for future universal quantum processors.</Abstract>
<Access xmlns="http://purl.org/coar/access_right"
>
</Access>
</Publication>