Optimal Universal Disentangling Machine

Optimal Universal Disentangling Machine for Two Qubit Quantum States

from:
http://adsabs.harvard.edu/abs/1999quant.ph..5036G

to download the paper:
http://arxiv.org/abs/quant-ph/9905036

Sibasish Ghosh, Somshubhro Bandyopadhyay, Anirban Roy, Debasis Sarkar, Guruprasad Kar
(Submitted on 11 May 1999 (v1), last revised 14 Oct 1999 (this version, v4))
We derive the optimal curve satisfied by the reduction factors, in the case of universal disentangling machine which uses only local operations. Impossibility of constructing a better disentangling machine, by using non-local operations, is discussed.


+

Disentangling Quantum States while Preserving All Local Properties

http://prl.aps.org/abstract/PRL/v83/i7/p1451_1

Electrical Engineering, UCLA, Los Angeles, California 90095-1594
Received 12 January 1999; published in the issue dated 16 August 1999

We consider here a disentanglement process which transforms a state ρ of two subsystems into an unentangled (i.e., separable) state, while not affecting the reduced density matrix of either subsystem. Recently, Terno [Phys. Rev. A 59, 3320 (1999)] showed that an arbitrary state cannot be disentangled, by a physically allowable process, into a tensor product of its reduced density matrices. In this Letter we show that there are sets of states which can be disentangled, but only into separable states other than the product of the reduced density matrices, and other sets of states which cannot be disentangled at all. Thus, we prove that a universal disentangling machine cannot exist.

hyper-entanglement

Controlling multiple qubits with hyper-entanglement



Controlling multiple qubits with hyper-entanglement

Controlling multiple qubits with hyper-entanglement
By Casey Johnston | Published about a year ago
Scientists are quickly putting the single-qubit system out of fashion with new setups that can simultaneously manipulate and read multiple qubits. An international collaboration recently completed an experiment involving the control of up to ten qubits at once, using hyper-entanglement and simple "cat states." While the system doesn't always read out perfectly, the approach could be further refined to produce better results.

Because qubit behavior is based in probability, it is difficult to exert a lot of control over a qubit. This problem gets a bit more significant as each additional qubit is added to the system, which has limited the number we can entangle at once. To hold down uncertainty and increase control as they add more qubits, scientists are now experimenting with hyper-entanglement, or entangling qubits on multiple levels at once. To put that another way, instead of entangling 10 different quantum objects, the authors entangled two separate properties of five items.

In this new experiment, scientists hyper-entangled sets of six, eight, and ten qubits in "cat states," or an equal superposition of two states (named after Schrodinger's cat, which occupied a superposition of the states "dead" or "alive"). The photons were entangled in two degrees of freedom: their polarization and their spatial modes. To get output from the photons once they were entangled, scientists used a special kind of interferometer that could gather information about one of the degrees of freedom without disturbing the other.

When the photons were measured, the photons produced the desired state around 60 percent of the time, with anything greater than 50% considered to be good enough to indicate that the system works at all. The eight-qubit system gave the best results, at 77.6 percent. The greatest limit of the system, according to the authors, was the photon detection efficiency, which will need to be significantly improved before implementation would be practical.

(Incidentally, the references to cat states start in the title—"Experimental demonstration of a hyper-entangled ten-qubit Schrödinger cat state"—and continue from there, with references to "ideal cat states" and "the hyper-entangled 2n-qubit cat state." "Cat" even appears as a term in some equations.

Nature Physics, 2010. DOI: 10.1038/NPHYS1603 (About DOIs).