All non-locally Realized Continuous Variable Bipartite Gaussian States are Entangled

Souvik Agasti · 2025

Read All non-locally Realized Continuous Variable Bipartite Gaussian States are Entangled by Souvik Agasti (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-21264.txt.

Introduction

Souvik Agasti examines the link between entanglement and nonlocality for a broad family of quantum states in a 2025 paper. The states in question are continuous-variable bipartite Gaussian states, shared between two parties. Agasti builds the test for nonlocality from the Wigner representation of Bell's function in phase space. The analysis shows entanglement is required for nonlocality but does not by itself guarantee it, while nonlocality always guarantees entanglement.

Catalog record

CREATORSouvik Agasti
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE14 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2507-21264

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

All non-locally Realized Continuous Variable Bipartite Gaussian States are Entangled Souvik Agasti Abstract We investigate the connection between entanglement and non-locality between continuous-variable bipartite Gaussian states. The investigation initiates with formulating non-locality by using the phase-space Wigner representation of Bell’s function. Furthermore, our analysis shows entanglement to be necessary for nonlocality, but not sufficient for it; however, nonlocality is sufficient to ensure entanglement. Preprint: APS/123-QED I Introduction Entanglement is a fundamental aspect of quantum mechanics, had remains to be the key feature in the development of the modern-day advanced quantum technologies, including quantum information processing [ 1 ] , quantum computation [ 2 ] , and communication [ 3 ] . Gaussian states have drawn attention for the generation and application of continuous variable (CV) entangled states [ 4 , 5 ] . A popular example of such Gaussian CV state is the two-mode squeezed vacuum (TMSV) state which show entanglement between two parties: signal(S) and idler(I) modes; had remained in the intensive interest of the community due to its application in sensing and manipulation of quantum states, e.g. quantum memory systems [ 6 ] and even in gravitational wave metrology [ 7 , 8 , 9 ] . Entanglement represents a correlation between parties in hybrid systems, therefore playing the key role in the theoretical establishment of quantum mechanics. The entanglement criteria of a bipartite system are dependent on separability limits. The separability exists if and only if the bipartite quantum state (\rho) can be expressed in the form of \rho=\sum_{j}p^{j}\rho^{j}_{I}\otimes\rho_{S}^{j} where \rho^{j}_{I},\rho_{S}^{j} are the normalized states of of the parties I and S, respectively, and the probability p^{j}\geq 0 satisfies \sum_{j}p^{j}=1 . The separability limit was first proposed by Peres [ 10 ] and afterwards by Horodecki for higher-dimensional cases [ 11 ] . In the Heisenberg picture, the idea has furthermore been established and formulated by Simon [ 12 ] and Duan et. al. [ 13 ] . Many other entanglement measures have also been used in different circumstances, e.g von Neumann entropy of reduced density operator, maximally optimized hybrid squeezed quadrature, Cauchy–Schwarz inequality, Hillery–Zubairy criterion, GHZ (Greenberger–Horne–Zeilinger) equality [ 14 ] . The impact of entanglement reflects a profound feature of quantum measurements on bipartite quantum systems, which is non-local realism, which solved the Einstein-Podolsky-Rosen (EPR) measurement paradox [ 15 , 16 ] . The test of a hidden variable that puts a limitation on local realism has been determined by Bell’s inequality [ 17 , 18 , 19 ] . Quantum nonlocality has been tested by exploring several types of Bell’s inequalities, which are experimentally realizable through homodyne measurements [ 18 , 19 ] . Such a type of inequality was first formulated by Clauser, Horne, Shimony, and Holt (CHSH) [ 20 , 21 ] , and further, it is expressed in phase space using Wigner quasiprobability functions [ 22 , 23 ] . Even though many theoretical conditions have been prescribed to witness entanglement and eventually quantify it, there is no direct analytical expression available for the quantification of non-locality, especially in the Heisenberg picture, which is still a gray area to explore. The maximization of Bell’s function has usually been numerically determined so far, while analyzing homodyne measurements, as the quantification of non-locality [ 17 , 24 ] . Therefore, it becomes essential to provide an analytical expression that formulates Bell’s inequality, especially for Gaussian states, which are often used in quantum optical experiments. Although entanglement and nonlocality are different phenomena, the latter has been vaguely considered as evidence of entanglement [ 14 ] . However, it has been realized that entanglement can be witnessed even though systems are locally realizable [ 25 , 26 , 24 ] . In contrast, strong quantum nonlocality has also been realized without entanglement as well [ 27 ] , and as a consequence, a limited class of states nonlocally realizable can witness quantum entanglement [ 28 ] . In fact, all classically correlated and non-product states admit violation of Bell’s inequality [ 29 , 30 , 31 ] . However, Barrett has proven that entangled mixed states do not always violate a Bell inequality [ 32 ] . Therefore, it remains an open quension and becomes necessary to see how the non-local behavior of Gaussian CV bipartite states is being modified with entanglement, and are there any relation can be established between the two phenomena. In this article, from the correlation between two parties for CV Gaussian states, using Wigner phase-space formalism, we propose a formulation of maximized Bell’s function as a measure of non-locality. Furthermore, we prove that entanglement could be necessary, but may not be sufficient, for non-locality; however, the nonlocality is sufficient for entanglement for bipartite Gaussian states. II Correlation Matrix in Standard Form The correlation between two parties in Gaussian states can be completely determined by a 4\times 4 real symmetric correlation matrix: V^{\prime}=\begin{bmatrix}V^{\prime}_{I}&V^{\prime}_{IS}\\ {V^{\prime}_{IS}}^{T}&V^{\prime}_{S}\end{bmatrix}, (1) where V^{\prime}_{I}(V_{S}) and V^{\prime}_{IS} are the 2\times 2 real matrices, represent for the systems I(S) and their cross-correlation, respectively. While studying their separability property, it remained convenient to transform the correlation matrix to some standard form using a local linear unitary Bogoliubov operator (LLUBO) U_{l}=U_{I}\otimes U_{S} [ 13 , 12 ] , and we will use the same standard form to formulate non-locality as well. The LLUBO operation works with a combination of squeezing transformation and rotation [ 33 ] , which transforms the correlation matrix of Eq. ( 1 ) to a standard form as V=\begin{bmatrix}n&&c_{1}&\\ &n&&c_{2}\\ c_{1}&&m&\\ &c_{2}&&m\end{bmatrix}\,,\,\,(n,m\geq\frac{1}{2}) (2) The operation goes with diagonalizing V^{\prime}_{I} and V^{\prime}_{S} by the first LLUBO operation using local orthogonal U_{I,S} , and then squeezing locally to obtain V_{I}=nI_{2} and V_{S}=mI_{2} where I_{2}=\text{diag}[1,1] . After these steps of operation, the transformed version of V^{\prime}_{IS} is furthermore diagonalized by another LLUBO with a suitable choice of orthogonal U_{I,S} , which does not influence V_{I} and V_{S} . The newly generated parameters in Eq. ( 2 ) are related to the inverients as \det(V^{\prime}_{I})=n^{2},\det(V^{\prime}_{S})=m^{2},\det(V^{\prime}_{IS})=c_{1}c_{2} and \det(V^{\prime})=(nm-c_{1}^{2})(nm-c_{2}^{2}) . For a bipartite Gaussian state, the mixedness of the state is quantified by \text{Tr}(\rho^{2})=\frac{1}{4\sqrt{\det(V)}} , which leads to 4(nm-c_{1}^{2})(nm-c_{2}^{2})\geq\frac{1}{4} (3) The inequality furthermore gives: nm-|c_{1}c_{2}|\geq\frac{1}{4} . III Quantum non-Locality The non-locality is justified by the violation of Bell’s inequality, which can be quantified by the maximum value of the Bell’s function (|B_{max}|) , expressed in terms of Wigner functions as [ 17 , 23 , 22 ] B=\pi^{2}\left[W(u_{00})+W(u_{I0})+W(u_{0S})-W(u_{IS})\right] (4) where W({u_{IS}})=\frac{1}{4\pi^{2}\sqrt{\det(V)}}\exp\left[-\frac{1}{2}{u_{IS}}^{T}{V}^{-1}{u_{IS}}\right] (5) represents the Wigner function for the vector {u_{IS}}=[Q_{I},P_{I},Q_{S},P_{S}]^{T} in phase space ([Q_{0},P_{0}]=[0,0]) . The sufficient condition for non-local realization of a quantum mechanical bipartite state is the violation of |B_{max}|\leq 2 , and a larger |B_{max}| indicates stronger non-locality. The Bell function defined in Eq. ( 4 ) will be maximized when the following function at the exponent will be maximized \displaystyle-\frac{1}{2}{u_{IS}}^{T}{V}^{-1}{u_{IS}}=-\frac{1}{2}\big[m\alpha_{I}^{2}+n\alpha_{S}^{2} (6) \displaystyle-2c\alpha_{I}\alpha_{S}(\cos\phi\cos\theta_{I}\cos\theta_{S}+\sin\phi\sin\theta_{I}\sin\theta_{S})\big] where \alpha_{I}^{2}=(\tilde{Q_{I}}^{2}+\tilde{P_{I}}^{2}),\alpha_{S}^{2}=(\tilde{Q_{S}}^{2}+\tilde{P_{S}}^{2}),\tilde{Q_{I}}=\alpha_{I}\cos\theta_{I},\tilde{P_{I}}=\alpha_{I}\sin\theta_{I},\tilde{Q_{S}}=\alpha_{S}\cos\theta_{S},\tilde{P_{S}}=\alpha_{S}\sin\theta_{S} , where \tilde{Q_{I}}=\frac{Q_{I}}{\sqrt{mn-c_{1}^{2}}},\tilde{P_{I}}=\frac{P_{I}}{\sqrt{mn-c_{2}^{2}}},\tilde{Q_{S}}=\frac{Q_{S}}{\sqrt{mn-c_{1}^{2}}},\tilde{P_{S}}=\frac{P_{S}}{\sqrt{mn-c_{2}^{2}}} , and c^{2}=c_{1}^{2}+c_{2}^{2},c_{1}=c\cos\phi,c_{2}=c\sin\phi , and therefore, \alpha_{I},\alpha_{S},c\geq 0 and 1\geq(\cos\phi\cos\theta_{I}\cos\theta_{S}+\sin\phi\sin\theta_{I}\sin\theta_{S})\geq-1 . As \phi is a fixed angle, for a given correlation matrix, we have the liberty to vary \theta_{I} and \theta_{S} to maximize the Bell function given in Eq. ( 4 ). We therefore find out the intermediatory maximized Bell function as \displaystyle B_{m} \displaystyle=\frac{1}{4\sqrt{\det(V)}}\big[1+\exp\left(-\frac{1}{2}m\alpha_{I}^{2}\right) (7) \displaystyle+\exp\left(-\frac{1}{2}n\alpha_{S}^{2}\right)-\exp\left(-\frac{1}{2}\left[m\alpha_{I}^{2}+n\alpha_{S}^{2}+2\tilde{c}\alpha_{I}\alpha_{S}\right]\right)\big], where \tilde{c}=\max[|c_{1}|,|c_{2}|] . It can be

Read the complete text