Quantum correlation provides a promising measure beyond entanglement. Here, we propose a necessary and sufficient condition for nonzero quantum correlation in continuous variable systems, which is simple and easy to perform in terms of a marker Q r . In order to get this condition, we introduce continuous-variable local orthogonal bases of the operator space, which are generalized from the orthogonal basis sets in local operator space for discrete variables. Based on this, we obtain the marker Q r for all bipartite continuous variable states, and provide several examples including two-mode Gaussian and non-Gaussian states. Our result may provide a candidate for quantum correlation measures, and can be measured by designed quantum circuits.
Quantum discord provides a promising measure for the quantum correlation beyond entanglement, and may be responsible for the quantum speedup in certain quantum computing tasks. Here, we propose a necessary and sufficient condition for nonzero quantum discord in continuous variable systems, which is simple and easy to perform in terms of a marker
Q. In order to obtain this condition, we introduce the continuous-variable local orthogonal basis of the operator space generalized from the orthogonal basis sets in local operator space for discrete variables. Based on this, we derive an explicit expression of the marker
Q for all the two-mode states, and present examples of all Gaussian states and several non-Gaussian states. Our result provides a candidate for quantum correlation measures, and can be easily measured by a designed quantum circuit.
We propose a measure of nonclassical correlation [Formula: see text] in terms of local Gaussian unitary operations based on square of the fidelity [Formula: see text] for bipartite continuous-variable systems. This quantity is easier to be calculated or estimated and is a remedy for the local ancilla problem associated with the geometric measurement-induced nonlocality. A simple computation formula of [Formula: see text] for any [Formula: see text]-mode Gaussian states is presented and an estimation of [Formula: see text] for any [Formula: see text]-mode Gaussian states is given. For any [Formula: see text]-mode Gaussian states, [Formula: see text] does not increase after performing a local Gaussian channel on the unmeasured subsystem. Comparing [Formula: see text] in scale with other quantum correlations such as Gaussian geometric discord for two-mode symmetric squeezed thermal states reveals that [Formula: see text] is much better in detecting quantum correlations of Gaussian states.
The characterisation of quantum correlations, also known as entanglement, is central to quantum information theory. Operationally, entanglement gives rise to the notion of quantum non-locality, which challenges the ‘reality’ of quantum mechanics (adopting the terminology of Einstein, Podolski and Rosen) and arguably represents its most distinctive feature, in the sense that it calls for a departure from certain foundational tenets of classical physics. It should thus not come as a surprise that entanglement turns out to be a key resource for the implementation of a number of quantum communication protocols and also necessary to achieve computational speed-ups with quantum hardware, in the sense that accessing regions of the Hilbert space with entangled states allows one to shorten the computational depth of certain algorithms. Not surprisingly, generating and maintaining entanglement is generally difficult in practice, as it requires strong interactions between selected components of a system that must at the same time be isolated from the environment. Thanks to the techniques developed in quantum optical set-ups to achieve squeezed light, entanglement is, however, comparatively accessible with optical quantum continuous variables. As we shall see, in fact, continuous variable entanglement is closely related to the notion of squeezing. It is therefore of the utmost importance to be able to qualify and quantify the entanglement of quantum continuous variables, which is the subject matter of the present chapter. The first part of the chapter is devoted to techniques and criteria based on second moments to detect the entanglement of Gaussian states and also to the description of possible ways to quantify such quantum correlations, bearing in mind that the determination of entanglement monotones of clear operational significance is still an open problem, even in the restricted arena of Gaussian states. Although the criteria based on second moments are sufficient, but not necessary, for the detection of non-Gaussian entanglement too, we will push our analysis further by introducing more general criteria based on higher order statistical moments, which will be put to use to detect, on paper, the quantum correlations of an entangled cat-like state. We will also address the issues of entanglement distillation and quantum non-locality tests, in whose contexts we shall encounter other non-Gaussian operations and states, such as photon-subtracted states.
This thesis explores the role of non-classical correlations in bipartite continuous variable quantum systems, and the approach taken is three-fold. We show that given two initially entangled atomic ensembles, it is possible to probabilistically increase the entanglement between them using a beamsplitter-like interaction formed from two quantum non-demolition (QND) interactions with auxiliary polarised light modes. We then develop an elegant method to calculate density matrix elements of non-Gaussian bipartite quantum states and use this to show that the entanglement in a two mode squeezed vacuum can be distilled using QND interactions and non-Gaussian elements. \n Secondly, we introduce a potential new measure of quantum entanglement in bipartite Gaussian states. This measure has an operational meaning in quantum cryptography and provides an upper bound on the amount of a secret key that can be distilled from a Gaussian probability distribution shared by two conspirators, Alice and Bob, given the presence of an eavesdropper, Eve. \n Finally, we go beyond the realm of quantum entanglement to explore other non-classical correlations in continuous variable systems. We provide solutions for a number of these measures on two mode Gaussian states and introduce the Gaussian Ameliorated Measurement Induced Disturbance (GAMID). The interplay between these different measures and quantum entanglement is examined. We then attempt to take small steps into the non-Gaussian regime by computing these non-classicality measures on the three-parameter continuous variable Werner states.
In this paper, a computable multipartite multimode Gaussian quantum correlation measure
M(k) is proposed for any
k-partite continuous-variable (CV) systems with
k≥2.
M(k) depends only on the covariance matrix of CV states, is invariant under any permutation of subsystems, is a quantification without ancilla problem, nonincreasing under
k-partite local Gaussian channels (particularly, invariant under
k-partite local Gaussian unitary operations), vanishes on
k-partite product states. For a
k-partite Gaussian state
ρ,
M(k)(ρ)=0 if and only if
ρ is a
k-partite product state. Thus, for the bipartite case,
M=M(2) is an accessible replacement of the Gaussian quantum discord and Gaussian geometric discord. Moreover,
M(k) satisfies the unification condition, hierarchy condition that a multipartite quantum correlation measure should obey.
M(k) is not bipartite like monogamous, but,
M(k) is complete monogamous and tight complete monogamous.
Entanglement detection and estimation are fundamental problems in quantum information science. Compared with discrete-variable states, for which lots of efficient entanglement detection criteria and lower bounds of entanglement measures have been proposed, the continuous-variable entanglement is much less understood. Here we shall present a family of entanglement witnesses based on continuous-variable local orthogonal observables (CVLOOs) to detect and estimate entanglement of Gaussian and non-Gaussian states, especially for bound entangled states. By choosing an optimal set of CVLOOs, our entanglement witness is equivalent to the realignment criterion and can be used to detect bound entanglement of a class of 2+2 mode Gaussian states. Via our entanglement witness, lower bounds of two typical entanglement measures for arbitrary two-mode continuous-variable states are provided.