--- title: Rotation-Invariant Symmetries of Gaussian States url: https://www.emergentmind.com/topics/rotation-invariant-symmetries-of-gaussian-states type: topic --- # Rotation-Invariant Symmetries of Gaussian States Rotation-invariant symmetries of Gaussian states refer to the invariance of these quantum states—specifically in continuous-variable (CV) systems—under rotations within phase space or, more generally, under orthogonal or symplectic transformations that preserve canonical commutation relations. These symmetries are foundational in characterizing the structure of Gaussian states, defining their symmetry groups, classifying their transformations, and providing operational tools in quantum information theory. They also underlie efficient state discrimination, optimal measurement design, entanglement theory, and tomography. This topic integrates deep results from representation theory, convex geometry, symplectic analysis, functional analysis, and experimental methodology. ## 1. Symmetry Groups of Gaussian States In the Hilbert space $L^2(\mathbb{R}^n)$, the set $S_n$ of all Gaussian states is exactly invariant under a group $\mathcal{G}_n$ of unitary operators defined as those $U$ for which $U \rho U^* \in S_n$ for any Gaussian state $\rho$ [1101.5041]. This group can be fully characterized as \[ U = \lambda\, W(\boldsymbol{\alpha})\, \Gamma(L) \] where: - $\lambda$ is a complex phase ($|\lambda| = 1$), - $W(\boldsymbol{\alpha})$ is a Weyl displacement (phase space translation) operator, - $L \in Sp(2n, \mathbb{R})$ is a real symplectic matrix, - $\Gamma(L)$ is the (Bogoliubov) automorphism implementing $L$ at the operator level. The symplectic group $Sp(2n, \mathbb{R})$ is defined as follows: \[ Sp(2n, \mathbb{R}) = \{ L \in GL(2n, \mathbb{R}) : L^T J L = J \} \] with \[ J = \begin{pmatrix} 0 & -I \\ I & 0 \end{pmatrix} \] where $I$ is the $n \times n$ identity. The transformations encoded in $Sp(2n, \mathbb{R})$ encompass all passive (rotational) and active (squeezing) canonical transforms that preserve commutation relations. Every symmetry operation—rotation, squeezing, and translation—corresponds to an element of this group, with rotations as a prominent subset. Under such symplectic transformations, covariance matrices transform as $S \mapsto L^T S L$. ## 2. Covariance Matrices, Convex Geometry, and Purification The convex set $K_n$ of admissible $2n \times 2n$ real symmetric covariance matrices $S$ for Gaussian states is constrained by the uncertainty relation: \[ 2S - iJ \geq 0 \] $K_n$ is closed and convex, with its extreme points exactly those matrices $S = \frac{1}{2} L^{T} L$, $L \in Sp(2n, \mathbb{R})$ [1101.5041]. Pure Gaussian states correspond to these extreme covariance matrices, whereas any mixed Gaussian state has a covariance matrix expressible as \[ S = \frac{1}{2}(L^T L + M^T M) \] for some $L, M \in Sp(2n, \mathbb{R})$. This structure immediately links covariance geometry with symplectic/rotational invariance. A crucial structural property is that any Gaussian state, mixed or pure, can be purified into a (higher-dimensional) pure Gaussian state via extension to $L^2(\mathbb{R}^{2n})$ (i.e., $S_n$ to $S_{2n}$), in a way compatible with symplectic symmetry [1101.5041]. ## 3. Unitary Implementations: Role of Weyl and Bogoliubov Operators General rotation-invariant symmetry operations on Gaussian states factor into Weyl operators (implementing phase-space translations) and unitaries implementing Bogoliubov automorphisms (transformations of the canonical operators generated by linear symplectic maps). The Weyl operators $W(a)$ act as \[ W(a) W(b) = e^{-i \mathrm{Im} \langle a, b \rangle} W(a + b) \] preserving the spread/covariance structure of the state but shifting mean values. Bogoliubov transformations correspond to $I(L)$, with $L \in Sp(2n, \mathbb{R})$, transforming \[ I(L) W(a) I(L)^* = W(La) \] rotating and squeezing the state in phase space. Both pieces together guarantee that all unitary symmetry operations preserving the Gaussian family arise from this construction, and thus all rotation invariance is encapsulated. ## 4. Rotation-Invariant Symmetries in State and Measurement Design Rotation-invariant symmetries are central to designing measurement and communication protocols: - **Geometrically Uniform Symmetry (GUS):** A set of states $\{|\gamma_j\rangle\}$ where each is generated by applying a unitary symmetry (typically rotation) to a reference, i.e., $|\gamma_j\rangle = Q^j |\gamma_0\rangle$ with $Q^K = I$ for some $K$ [1410.5282]. For $N$-mode Gaussian states, rotations leave the squeezed-displaced state class invariant, i.e., \[ R(\varphi) D(\alpha) S(z) |0\rangle = D(e^{i\varphi}\alpha) S(e^{i\varphi} z e^{i\varphi^T}) |0\rangle \] - **Optimal Quantum Measurements:** For geometrically uniform constellations, the Gram matrix of inner products is circulant, allowing diagonalization via discrete Fourier transform, which simplifies square-root measurement (SRM) constructions and yields closed-form optimal detection rules for quantum communications (e.g., in pulse position modulation). ## 5. Invariant Measures and Typicality Several natural invariant measures exist on the manifold of Gaussian states—most notably, those inherited from the Haar (unitary) measure on the Gaussian symmetry group. For multimode states, the invariant measure can be decomposed over local and nonlocal degrees of freedom using the symplectic eigenvalues of the covariance matrix, with the nonlocal (entanglement-relevant) part given by \[ \prod_{j