5 ms·
> The density matrix describing C at the beginning and B at the end is the same No, it isn't. C starts in a pure state. B is created and always remains in a
by lisper 8y ago
> The density matrix describing C at the beginning and B at the end is the same
No, it isn't. C starts in a pure state. B is created and always remains in a mixed state.
> (otherwise how could all measurement outcomes be the same?)
Because you cannot distinguish a mixed state from a pure state when measuring only a single particle. You can distinguish them if you make measurements on multiple particles in the same state: pure states produce first-order interference. Mixed states don't. (This is just the old QM trope about "measurement destroys interference", which is not quite true. Entanglement "destroys" interference, except that it doesn't really, it just destroys first-order interference.)
> B is no longer entangled with the A/C system after Alice's measurement.
Yes, it is. The only way to "undo" an entanglement is to bring the originally entangled particles back together at the same location, i.e. it is essentially impossible. If you could cause an entangled particle to become unentangled (i.e. to transition from a mixed state to a pure state) by performing an operation on its partner then you could do FTL signaling by observing the presence or absence of first-order interference.
> how do you explain that the state of B is |up> at the end
It isn't. It's in a mixed state. It is no different than regular EPR with only two particles. If Alice measures A and gets the result "up" then Bob is guaranteed to get "down", but that does NOT mean that B is in the |down> state. It is still in a mixed state. |down> is a pure state.
It is important to remember that mixed states and pure states cannot be distinguished by making measurements on a single particle, so if there is only one particle this is a philosophical distinction, not a scientific one. But if the experiment can be repeated, then the fact that B is mixed can be experimentally demonstrated, with all the usual caveats about quantum weirdness. The point is that QT does not introduce any new weirdness that wasn't already there in EPR.
- kgwgk 8y agoIn case anyone else is reading this, I want to make clear that I'm giving the standard description of quantum teleportation. https://en.wikipedia.org/wiki/Quantum_teleportation#Formal_presentation https://en.wikipedia.org/wiki/Quantum_teleportation#Formal_p... provides a clear description of the initial states of C and A/B, the operations performed by Alice and Bob and the final state of B (which is identical to the initial state of C). Clearly you don't find this description correct but I would be surprised if you can write down an alternative description that makes sense. You're free to have your own theory, but I don't understand why do you feel that you need one in the first place. What is the problem with measurement breaking entanglement? > If you could cause an entangled particle to become unentangled (i.e. to transition from a mixed state to a pure state) by performing an operation on its partner then you could do FTL signaling by observing the presence or absence of first-order interference. Let's say Alice and Bob have entangled spins in the pure state (leaving out the normalization term): |up>_alice * |down>_bob - |down>_alice * |up>_bob According to QM, if Alice measures the spin along the z-axis one of the following alternatives will happen: a) she gets a positive measurement, her spin is in the pure state |up>, Bob's spin is in the pure state |down>, there is no entanglement between the spins or b) she gets a negative measurement, her spin is in the pure state |down>, Bob's spin is in the pure state |up>, there is no entanglement between the spins How do you think this allows for FTL signaling?
- lisper 8y ago> How do you think this allows for FTL signaling? I already told you: with a single particle, it's not possible. But with multiple particles you can tell the difference between a pure state and a mixed state by observing the presence or absence of first-order interference. So to communicate, you send a stream of EPR pairs to Alice and Bob. Alice transmits by either measuring batches of particles or not measuring them, and Bob receives by observing the presence or absence of first-order interference on his end. Fun fact: I was able to get a patent on this idea :-) https://patents.google.com/patent/US7126691B2/en https://patents.google.com/patent/US7126691B2/en
- kgwgk 8y agoPeople get patents for perpetual motion machines as well... Do you really think your idea works? This is what I think: the quantum state of your entangled pair is a pure state and if you look at the quantum state of the photon sent to the "receiver" it is described by the density matrix obtained by tracing out the other photon : 1/2 0 0 1/2 If you measure in the "transmiter" the horizontal polarization, what you get in the "receiver" is a 50/50 mixture of pure states |H> and |V> which is described by the same density matrix. If you measure in the "transmiter" the 45-degrees polarization, what you get in the "receiver" is a 50/50 mixture of pure states |45> and |-45> which is again described by the same density matrix. No matter what yo do or do not measure at the "transmiter" you will get the same measurements at the "receiver" because the density matrix describing the state is always the same (unpolarized light).