Tue, 2010-03-02 14:55 - Daniele Binosi

Information processing nowadays is commonly implemented using quantities such as charges, voltages, or currents in electronic devices which operate on the basis of classical physics. Instead, Quantum Computing (QC) and more generally, quantum information processing (QIP) employ the laws of quantum mechanics for information processing. For such devices, corresponding building blocks are quantum bits (qubits) and quantum registers, and the basic gate operations are given by logical and coherent operations on individual qubits (single qubit operations) and controlled coherent interactions between two qubits (two-qubit operations) such that the state of the target qubit is changed conditional to the state of the controlling qubit. In principle, a large scale quantum computer can be built using these primitives which must be realized by a controllable quantum system, provided the physical system meets the following requirements (DiVincenzo criteria):

- System is comprised of well characterized qubits and allows for scalability;
- Ability to initialize the state of the qubits;
- System provides long coherence times, much longer than a gate operation time;
- A universal set of gates is experimentally feasible;
- Qubit specific measurement capability;
- Ability to interconvert stationary and flying qubits;
- Faithful transmission of flying qubits between specified locations;

At present, there are a number of technologies under investigation for their suitability to implement a quantum computer. No single technology meets currently all of these requirements in a completely satisfactory way. Therefore, the ongoing research on quantum information processing is highly interdisciplinary, diverse and requires a coordinated effort to create synergies while the common goal is the implementation of a working quantum processor. While at present several approaches have demonstrated basic gate operations and are even able to prove that quantum computing has become reality with few qubits, large scale quantum computation is still a vision which requires ongoing research for many years to come.

The long-term goal in quantum computation is, of course, a large-scale quantum computer which will be able to efficiently solve some of the most difficult problems in computational science, such as integer factorization, quantum simulation and modeling, intractable on any present or conceivable future classical computer.

Therefore, the general problems to be solved for QC and QIP are in particular

- Identification of the best suitable physical system which allows for scalability, coherence and fast implementation of QIP;
- Engineering and control of quantum mechanical systems far beyond anything achieved so far, in particular concerning reliability, fault tolerance and using error correction;
- Development of a computer architecture taking into account quantum mechanical features;
- Development of interfacing and networking techniques for quantum computers;
- Investigation and development of quantum algorithms and protocols;
- Transfer of academic knowledge about the control and measurement of quantum systems to industry and thus, acquisition of industrial support and interest for developing and providing quantum systems.

Quoting a recent article by David DiVicenzo in "Science" (7 october 2011), "during the past decade, a wide array of physical systems - atoms, semiconductors, and superconductors - have been used in experiments to create the basic components of quantum-information processing. Precision control over elementary quantum two-state systems (qubits) is now well advanced, and it is now possible to ask how a complete, functioning quantum computer with many qubits would really work. Although the physical qubits used are still extremely different, it is now time to attack device-independent questions of system functionality."