Criteria for a Scalable Physical Technology to Implement a Quantum Computer
In 2000, David DiVincenzo of IBM published a set of criteria or requirements for assessing the viability of any physical implementation for a quantum information processing system. They can be described as follows:
- A scalable physical system with well-characterized qubits. The stable quantum state could entail the spin (up or down) of an electron or the polarization (vertical or horizontal) of a photon. It requires accurate knowledge of physical parameters, internal energy, and coupling between qubits.
- The ability to initialize the state of the qubits to a simple trusted state, such as . Technology-dependent approaches would need to be developed to initialize the quantum registers, including the cooling or measurement operations; an important question involves how long this would take. Quantum error correction requires ancillary qubits in known states.
- Long decoherence times. Decoherence (the collapse of the probability wave) must take longer than 105 times the quantum computer clock time. The dynamics of the qubit interacting with its environment will need to be better understood and controlled. Faulty control mechanisms lead to faulty gates. Even worse, noise is essentially analog acting on 2n complex numbers in superposition. Also, as a result of entanglement with the environment, coherent superposition becomes incoherent. The duration of decoherence may be the biggest obstacle to quantum computing.
- A universal set of quantum gates. Quantum algorithms describe a sequence of unitary transformations. It is difficult in some cases to create these operations for two and three qubits, and difficult to control the on/off interactions for these gates as a result of imperfect implementation. Not all gates would be available in each technology.
- A qubit-specific measurement capability. A technology-dependent readout mechanism is required to read specific qubits without perturbing other qubits. Current techniques are much less than 100 percent efficient.
- The ability to convert stationary and flying qubits. Flying qubits (e.g., photons) can be used to store and transport information; doing this at will has yet to be achieved.
- The ability to faithfully transmit flying qubits between specified locations. In a real-world implementation, transmission losses could affect computation.