This thesis work focuses on the design and validation of a novel control scheme for open quantum systems subject to continuous homodyne measurements. This control law is based only on the measured output signal, without any additional state estimation and filtering, thus offering a massive improvement in the computational effort needed to compute the value of the control in real time. After introducing the elements of probability theory, stochastic calculus and quantum mechanics we need in the rest of the thesis, we discuss how the continuously-monitored system behaves in free evolution. Starting from this, we design a new control law based on the output signal and we prove that, at least in the ideal setting, this control law is able to render a desired target pure state a globally asymptotically stable state for the closed-loop dynamics. This is first done for qubit systems and then generalized to multilevel systems. We also provide a possible practical implementation of the control law which uses the noisy signals that are actually available. Once this control law is in place, we study potential applications to improve the convergence speed when the convergence is already ensured by other feedback techniques, such as the Wiseman-Milburn (derivative) feedback law, highlighting the benefits of the new added component. We validate all the proposed control schemes, by means of numerical simulations with different systems of interest. Lastly, we describe possible future research directions that emerged as promising avenues to further improve feedback control performance.
Questo lavoro di tesi si concentra sul progetto e validazione di un nuovo controllore per sistemi quantistici aperti soggetti a misure omodine continue. Questa legge di controllo è basata unicamente sulle misure effettuate, senza ulteriore filtraggio, rappresentando un netto miglioramento del tempo necessario al controllore per calcolare il nuovo segnale di controllo. Dopo aver introdotto gli elementi necessari di teoria della probabilità, equazioni differenziali stocastiche e meccanica quantistica, studiamo il comportamento asintotico del sistema, misurato in maniera continua e in evoluzione libera. A partire da questa, deriviamo una nuova legge di controllo e dimostriamo che, almeno nel caso ideale, questa è sufficiente a rendere lo stato puro desiredato globalmente asintoticamente stabile in catena chiusa. Dimostriamo questo prima per un qubit e poi generalizziamo a sistemi multilivello, fornendo anche una possibile implementazione pratica della legge di controllo che opera con i segnali rumorosi effettivamente a nostra disposizione. Vengono inoltre studiate possibili applicazioni a sistemi per cui abbiamo già una legge di controllo che ci garantisce la stabilità globale e asintotica di un certo stato puro, analizzando come l'aggiunta del nuovo termine possa migliorare la velocità di convergenza in particolari casi. La validazione di questa legge di controllo viene effettuata per mezzo di simulazioni numeriche, confermandone la bontà. Infine, discutiamo anche di possibili ulteriori aree di ricerca che sono emerse come promettenti direzioni per migliorare la qualità del controllore.
Models and methods for quantum output feedback
FRANCESCHETTI, LORENZO
2024/2025
Abstract
This thesis work focuses on the design and validation of a novel control scheme for open quantum systems subject to continuous homodyne measurements. This control law is based only on the measured output signal, without any additional state estimation and filtering, thus offering a massive improvement in the computational effort needed to compute the value of the control in real time. After introducing the elements of probability theory, stochastic calculus and quantum mechanics we need in the rest of the thesis, we discuss how the continuously-monitored system behaves in free evolution. Starting from this, we design a new control law based on the output signal and we prove that, at least in the ideal setting, this control law is able to render a desired target pure state a globally asymptotically stable state for the closed-loop dynamics. This is first done for qubit systems and then generalized to multilevel systems. We also provide a possible practical implementation of the control law which uses the noisy signals that are actually available. Once this control law is in place, we study potential applications to improve the convergence speed when the convergence is already ensured by other feedback techniques, such as the Wiseman-Milburn (derivative) feedback law, highlighting the benefits of the new added component. We validate all the proposed control schemes, by means of numerical simulations with different systems of interest. Lastly, we describe possible future research directions that emerged as promising avenues to further improve feedback control performance.| File | Dimensione | Formato | |
|---|---|---|---|
|
Franceschetti_Lorenzo.pdf
accesso aperto
Dimensione
1.93 MB
Formato
Adobe PDF
|
1.93 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/93671