What are destructive measurements or incomplete measurements, and what is the (conceptual) difference between them and a usual measurement?

I read somewhere that destructive measurements consume their qubit.

reference: measurement calculus
p.6 we simplified equations 2 and 3 to equations 4 and 5.

  • Destructive measurements are processes that completely destroy the system they are measuring, and they are primarily used when detecting light. As an example, to detect the polarization of a photon you can pass it through a polarizing beam splitter and put detectors on either output port: you get complete information of a projective measurement on the photon, but the photon also ceases to exist.

    This describes most measurements performed on light, but not all measurements do this, and the alternative, called quantum non-demolition experiments (example, doi), earned Serge Haroche the 2012 Nobel prize in physics.

  • An incomplete measurement is a completely different beast: it's a projective measurement that doesn't fully collapse the wavefunction, i.e. it completely collapses some superpositions but it leaves others untouched. As a simple example, suppose that you had the wavefunction $$ |\psi⟩ = a |1⟩ + b|2⟩ + c|3⟩ + d|4⟩;$$ an example of an incomplete measurement is one that determines whether the state is in either the pair $|1⟩$ and $|2⟩$ or in $|3⟩$ and $|4⟩$, leaving behind the state $$ |\psi_{12}⟩ = \frac{a |1⟩ + b|2⟩}{\sqrt{|a|^2+|b|^2}} \quad\text{or}\quad |\psi_{34}⟩ = \frac{c|3⟩ + d|4⟩}{\sqrt{|c|^2+|d|^2}},$$ with probabilities $|a|^2+|b|^2$ and $|c|^2+|d|^2$, respectively. This can be achieved by e.g. making the system interact with some auxiliary two-level system which is later projectively measured, but the details there are not particularly important.

  • $\begingroup$ thanks for your answer.I read a paper that it ignores the pauli operators that act on the qubit after the measurement in the case of destructive measurements. I can't match your definition with this result! $\endgroup$ – sahab110 Jan 3 '18 at 16:53
  • $\begingroup$ If you want people to respond to "this result" and you don't specify exactly what the result is and the precise reference where you found it, then I imagine you think other people are telepathic and can read your mind? I never understand why people omit references and context, but for general reference: it is never a good idea. $\endgroup$ – Emilio Pisanty Jan 3 '18 at 17:00
  • $\begingroup$ I'm sorry. You are right. I add them to the question. $\endgroup$ – sahab110 Jan 3 '18 at 19:22
  • 1
    $\begingroup$ @sahab110 That is a much more specific question which should be asked separately (and providing full context and a much clearer description of what it is you find confusing). The fact that someone answers one of your question doesn't turn them into your personal tutor - I won't answer further queries on this thread. $\endgroup$ – Emilio Pisanty Jan 3 '18 at 19:25
  • $\begingroup$ my question is :"what is the destructive measurements?" $\endgroup$ – sahab110 Jan 3 '18 at 19:39

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.