7

The representation of bits in different technological areas:

  1. Normal digital bits are mere abstractions of the underlying electric current through wires. Different standards, like CMOS or TTL, assign different thresholds to such signals: "if the voltage goes above this level, then the bit is 1; if it goes below this level, then the bit is 0; discard in any other case".

  2. In genetics, we usually consider a signal as a 1 if it is "enough" to trigger the target response; 0 otherwise. In this scenario, the thresholding is qualitative.

From the point of view of quantum information, qubits also abstractions, but in practice measurements will need standards to be comparable.

Question: From the point of view of quantum engineering, is there any standard technique/method to identify their value e.g. based on detection thresholds or fidelity verification like Bell inequalities violation? Are there units for that hidden signals?

The best possible answer would probably contain specific details for different architectures (e.g. superconductors vs photons) or contexts (e.g. quantum computing vs quantum communications).

agaitaarino
  • 3,827
  • 2
  • 12
  • 42
AG-M
  • 381
  • 1
  • 3

1 Answers1

3

It seems to me that, from the point of view of quantum engineering, we are a few years away from being at the point of fixing standards.

Standards are a good way to ensure the reproducibility of the behaviour of a piece of information technology, and the interoperability of the functionality of multiple pieces of information technology. It is clear that at some point we will require such standards. The question is: how would one begin to formulate those standards?

  • Before we ensure the reproducibility of a piece of quantum technology, we should build a piece of quantum technology whose behaviour we want to reproduce, rather than immediately set out to improve upon. With perhaps one exception, I expect that almost everyone in the quantum technologies game are more interested in bettering their own designs — possibly involving significant revisions to any design parameters which could play the role of standards — rather than setting down parameters which they expect that everyone will be happy to use.

    Conceivably D-WAVE is at this stage — obviously they would also like to improve upon their existing technology (as for instance do conventional chip manufacturers), but my understanding is that they are in the business of making $\,N>1\,$ computers of a given model whose behaviour is intended to achieve a certain, well, 'standard'. Whatever the computational power of their machines, it seems that they are in the business of engineering complex systems with multiple components, and doing so in a reproducible manner. So it seems very likely that they have standards for their qubits: but it is not clear that anyone apart from them will be interested in conforming to those standards (rather than building more versatile quantum computers for instance), or to what extent D-WAVE's standards are publicly available.

    Another incipient exception is in the field of optics, where the inclination is very strong to use the existing materials technology of fibre optic cables: while there is no formal standard likely exists, a practical standard of using wavelengths of light which have very low attenuation in modern-day optical fibre is one that could be comfortably predicted to continue for the foreseeable future.

  • Given this situation for individual approaches to quantum technologies, the question of interoperation is even more premature. No-one knows what their equipment is going to be interoperating with — again, with the probable exception of fibre-optic cable, and perhaps the mains frequency of your electrical grid if this is somehow important to take into account.

But ask the question again in five or ten years (more likely ten), and you may get a more interesting answer.

Niel de Beaudrap
  • 12,052
  • 1
  • 31
  • 69