Quantum Technology: Computational Models for Quantum Device Design (12w5061)
Lloyd Hollenberg (University of Melbourne)
Frank Langbein (Cardiff University)
Sophie (Sonia) Schirmer (Swansea University)
(1) What are the most promising applications in nearer future? E.g., semiconductor or superconductor devices for quantum metrology, information processing, or perhaps modelling and understanding quantum effects in biological systems, or medical applications. This will further lead into a discussion of what design tools are needed for the most promising applications. An important issue to consider is also the type of models needed for the different operations required for an application such as the relation between a geometric model of the device and the model required for dynamic control.
(2) What techniques and models are needed to design, simulate and control the operation of these quantum devices? Dynamic control simulations require control system models, conventionally based on partial differential equations derived from fundamental principles, such as the control dependent Schroedinger or quantum Liouville equation. Efficient simulation algorithms for physically accurate computational models are crucial for this task. Statistical models simply describing the behaviour of an actual device may sometimes be sufficient, or at least be used to augment the differential equation approach to consider specific material properties and the complete behaviour of real devices.
(3) What protocols for experimental system identification, parameter estimation and model verification are available? How efficient and reliable are these? What are their experimental requirements and how realistic are these? Do they provide the models we need? E.g. common techniques such as spectroscopy and quantum process tomography provide information about the system but do not directly provide the type of dynamic control systems models we require. How can techniques be combined, modified or extended to enable construction of the models we need?
(4) What are efficient ways to solve the inverse of the modelling and simulation problem, i.e., design and control optimization from optimal device geometries to dynamic voltage profiles applied to control electrodes or optimally shaped control pulses to be applied? What should be the main objectives of such optimizations? What are the practical constraints? How well can current algorithms cope with large-scale problems?
(5) What computational tools are available and what new tools are required or desirable, especially with a view towards integration of device or system design, dynamic simulations of quantum evolution, experimental data analysis and design and control optimization. What is the efficiency of simulation algorithms on different hardware platforms? What role could effective visualization of raw simulation data and automated, though perhaps human-guided, data analysis play? How could it make the process of model analysis and verification more efficient?
Overall by answering these questions we envisage to devise a roadmap towards computer-aided design systems for quantum devices, listing the required functionalities and various approaches to provide these, ensuring interoperability between the approaches as far as feasible, and dependencies between results and approaches. We expect that the workshop will improve the mutual understanding of the various problems and issues involved and will initiate collaborations between the participants to work on the individual issues towards realising the roadmap.
We expect to spend about one day for each of the questions, with introductory talks about the issues involved by the respective experts in the morning followed by discussions of specific issues and networking sessions in the afternoon with a final wrap up session per day.
The questions cover a diverse range of topics from computational modelling of physical systems, to high performance and parallel computing, to device design and nano-fabrication, to control and optimisation, to machine learning and pattern analysis, to quantum theory and experimental physics. This is reflected in the list of participants which come from mathematics, computer science, engineering and physics with relevant background in some of these areas. Even if the workshop is strongly focused on computational models for designing and controlling quantum devices, this wide range of expertise is necessary to cover the issues and combine the expertise of similar problems in different areas. Recent experience of the organisers in identification, simulation and optimisation of quantum devices has shown that there are many still unexplored links between the physics and control of quantum devices and engineering and computer science topics. This workshop aims to bring these together under the common theme of computational models. Participants range from postdocs, to young and senior staff, and we also intend to invite PhD students working on directly relevant topics at the time of the workshop (as identified by the more senior staff on the participant list).
In the past Banff IRS has contributed to advance research in this cutting edge area of nanoscience at the quantum edge with recent workshops such as "Physics-Based Mathematical Models of Low Dimensional Semiconductor Nanostructures (LDSN): Analysis and Computation" in 2007, and an upcoming workshop on Quantum Control in 2011. This workshop focuses on the models and computational tools to enable the application of these results in the currently emerging area of quantum technology and engineering.
Models with efficient design and control methods are crucial to enable engineering complex quantum devices, whose ultimate importance lies in the applications. Quantum communication and encryption is maybe the most advanced area on the application side, with important effects already on secure communication. However, similar high-impact technologies are on the horizon in electronics, metrology, imaging, biology and medicine.