betaFeedback

WorksPhysics and Astronomy

Quantum tomography on Schwartz space

Abstract

In this work we show that standard results in quantum tomography, which are typically proven in the context of Hilbert spaces, can analogously be established in the theory of Schwartz spaces. In particular, we will see that quantum states are equivalently characterized by their Wigner function, the characteristic of their Wigner function, and the probability distributions for a family of linear combinations of position and momentum. Physically, the latter can be understood as the evolution of its probability distribution in position over time under, for example, free or harmonic evolution. The result is obtained with techniques that are more accessible to a physicist audience, and shows that, at least for tomography results, the full structure of Hilbert spaces is not necessary.

Introduction

In previous work [2], some of us raised awareness of the difficulty of providing a physical justification for Hilbert spaces as a mathematical foundation for quantum mechanics. In particular, the completeness of the space with respect to the topology induced by the inner product necessarily introduces states with pathological physical properties. Efforts to reconstruct the theory of quantum mechanics based on physical requirements might therefore look for alternative targets. Here we follow up on our investigation of Schwartz spaces as a source of promising insights for alternative definitions and see whether we can establish analogue results completely within that theory.

Schwartz functions are smooth functions that decay rapidly at infinity, are dense in L2L^2 and have finite expectation values for all polynomials of position and momentum, which depend continuously on the state. These nice mathematical properties considerably simplify their analysis by avoiding many of the domain issues common for (unbounded) operators on Hilbert spaces and other subtleties like functions only being defined almost everywhere. In this paper, we focus on the use of tomography as a powerful tool to analyze the physical content encoded in Schwartz wavefunctions.

We will see that the state of a particle described by a Schwartz wavefunction can be uniquely identified (up to a global phase) by measuring the probability distribution of a family of linear combinations of position and momentum. This can also be interpreted as the evolution over time of the probability distribution in position, either freely or in the presence of a harmonic potential. While the latter can be achieved in finite time, the former would require infinite time. The probability densities in question are Schwartz functions as well, i.e., rapidly decaying at infinity. This result is analogous to the reconstruction of a probability distribution on classical phase space [6, 11].

While the general claim already exists (in different forms) in the literature (see e.g., Refs. [1, 12, 9, 5, 10]), we provide here a short and straightforward proof, simplified by the properties of Schwartz spaces. In particular, we do not require any reference to the completeness of L2L^2. We hope that this makes the discussion more accessible to a broad physics community.

The paper starts by reviewing some mathematical prerequisites, including Wigner functions. The reconstruction then proceeds in the following steps.

Sequence of reconstructions

Figure 1. Sequence of reconstructions.

  1. The standard definition of a complex (wave-)function ψ\psi with certain regularity/decay properties.

  2. The well-known Wigner function WψW_\psi, i.e., a quasiprobability distribution in phase space.

  3. The characteristic function χψ\chi_\psi, i.e., expectation values of the unitary continuous operator ei(aX+bP)e^{i(aX+bP)} implementing translations in phase space.

  4. Probability distributions ρα\rho_\alpha for a family of linear combinations of the position and momentum operators cos⁡α X+sin⁡α P\cos\alpha\,X+\sin\alpha\,P, which are all continuous operators.

  5. The probability distribution ρt\rho_t of the position operator evolving freely (or harmonically) in time.

Mathematical prerequisites

In this section we provide the necessary prerequisites for what follows. First we introduce the central object of our analysis, namely Schwartz functions, which have a number of nice properties. Then we define Wigner functions as an alternative representation of quantum mechanical states and show the well-known result of how to recover the original wavefunction from its Wigner function, thereby showing that there exists a (constructive) one-to-one correspondence. Since the Wigner function is a special case of a more general transformation between operators and functions on phase space (see § 2.3), it is more convenient to show how to reconstruct the kernel of the wavefunction, which is closer to the density matrix/operator of the state. Lastly we introduce a family of operators that will be used in the reconstruction later.

Schwartz space

Seminorms The Schwartz space S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}) of smooth functions f:Rn→Cf:\mathbb{R}^n\to\mathbb{C} and its topology are defined in terms of the following family of seminorms, indexed by multi-indices α,β∈Nn\alpha,\beta\in\mathbb{N}^n

∥f∥α,β=sup⁡x∈Rn∣xαDβf(x)∣(1)\lVert f\rVert_{\alpha,\beta}=\sup_{x\in\mathbb{R}^n}\left|x^\alpha D^\beta f(x)\right| \tag*{(1)}

where xα=x1α1x2α2⋯xnαnx^\alpha=x_1^{\alpha_1}x_2^{\alpha_2}\cdots x_n^{\alpha_n} are position operators and Dβf=∂1β1∂2β2⋯∂nβnfD^\beta f=\partial_1^{\beta_1}\partial_2^{\beta_2}\cdots\partial_n^{\beta_n}f are (except for a factor of −iℏ-i\hbar) momentum operators.

Topological vector space The Schwartz space is then

S(Rn,C)={f∈C∞(Rn,C)∣∀α,β∈Nn, ∥f∥α,β<∞}(2)\mathcal{S}(\mathbb{R}^n,\mathbb{C})=\{f\in C^\infty(\mathbb{R}^n,\mathbb{C})\mid\forall\alpha,\beta\in\mathbb{N}^n,\ \lVert f\rVert_{\alpha,\beta}<\infty\} \tag*{(2)}

with the locally convex topology generated by all these semi-norms.

Inner product and Completeness We equip S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}) with the L2L^2 inner product (and consequent norm)

⟨ψ∣ϕ⟩=∫Rnψ∗(x)ϕ(x) dnx(3)\langle\psi\mid\phi\rangle=\int_{\mathbb{R}^n}\psi^*(x)\phi(x)\,\mathrm{d}^n x \tag*{(3)}

which also induces another topology (which defines L2L^2), but is not used here. Note that S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}) is complete w.r.t. to its own topology, but not w.r.t. to that of L2L^2. In fact, the unique completion of S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}) in that topology is L2L^2. This makes L2L^2 a Hilbert space, while S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}) is a pre-Hilbert space.

Physical seminorms Alternatively, one can use the following families of semi-norms closely related to the moments of position and momentum

∥f∥α,βL1=∫Rn∣xαDβf(x)∣ dnx(4)\lVert f\rVert_{\alpha,\beta}^{L^1}=\int_{\mathbb{R}^n}\left|x^\alpha D^\beta f(x)\right|\,\mathrm{d}^n x \tag*{(4)}
∥f∥α,βL2=(∫Rn∣xαDβf(x)∣2 dnx)1/2(5)\lVert f\rVert_{\alpha,\beta}^{L^2}=\left(\int_{\mathbb{R}^n}\left|x^\alpha D^\beta f(x)\right|^2\,\mathrm{d}^n x\right)^{1/2} \tag*{(5)}

which will analogously define the same set and topology [8].

Fourier transform The Schwartz space is closed under the Fourier transform F:S(Rn,C)→S(Rn,C)\mathcal{F}: \mathcal{S}(\mathbb{R}^{n}, \mathbb{C}) \to\mathcal{S}(\mathbb{R}^{n}, \mathbb{C})

(Fψ)(p)=1(2πℏ)n∫Rne−iℏpxψ(x) dx(6)(\mathcal{F}\psi)(p) = \frac{1}{(2\pi\hbar)^{n}} \int_{\mathbb{R}^{n}} e^{-\frac{i}{\hbar}px}\psi(x)\,\mathrm{d}x \tag*{(6)}

and its inverse

(F−1ψ)(x)=∫Rneiℏpxψ(p) dp.(7)(\mathcal{F}^{-1}\psi)(x) = \int_{\mathbb{R}^{n}} e^{\frac{i}{\hbar}px}\psi(p)\,\mathrm{d}p . \tag*{(7)}

The same holds for partial Fourier transforms like

(F1ψ)(p1,x2)=12πℏ∫Re−iℏp1x1ψ(x1,x2) dx1.(8)(\mathcal{F}_{1}\psi)(p_{1},x_{2}) = \frac{1}{2\pi\hbar}\int_{\mathbb{R}} e^{-\frac{i}{\hbar}p_{1}x_{1}}\psi(x_{1},x_{2})\,\mathrm{d}x_{1} . \tag*{(8)}

Operator algebra All operators in this paper are everywhere defined on S(Rn,C)\mathcal{S}(\mathbb{R}^{n}, \mathbb{C}), linear and continuous. These form the algebra L(S)\mathcal{L}(\mathcal{S}) w.r.t. composition. In this context symmetric and self-adjoint are equivalent, namely for all ψ,ϕ∈S(Rn,C)\psi,\phi\in\mathcal{S}(\mathbb{R}^{n}, \mathbb{C})

⟨Oψ∣ϕ⟩=⟨ψ∣Oϕ⟩.(9)\langle O\psi|\phi\rangle= \langle\psi|O\phi\rangle. \tag*{(9)}

Wigner functions

Definition 10 (Wigner function). For a function ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R}, \mathbb{C}), its Wigner function Wψ:R2→RW_{\psi}: \mathbb{R}^{2} \to\mathbb{R} is given by

Wψ(x,p)=12πℏ∫Re−iℏpyψ(x+y2)ψ(x−y2)‾ dy.(10)W_{\psi}(x,p) = \frac{1}{2\pi\hbar}\int_{\mathbb{R}} e^{-\frac{i}{\hbar}py}\psi\left(x+\frac{y}{2}\right)\overline{\psi\left(x-\frac{y}{2}\right)}\,\mathrm{d}y . \tag*{(10)}

The Wigner function is a real Schwartz function in S(R2,C)\mathcal{S}(\mathbb{R}^{2}, \mathbb{C}) (see Prop. 1.92 in [7]).

Definition 12 (Kernel). The kernel Kψ:R2→CK_{\psi}: \mathbb{R}^{2} \to\mathbb{C} of a function ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R}, \mathbb{C}) is given by

Kψ(x,y)=ψ(x)ψ(y)‾.(11)K_{\psi}(x,y) = \psi(x)\overline{\psi(y)} . \tag*{(11)}

Proposition 14. The kernel Kψ:R2→CK_{\psi}: \mathbb{R}^{2} \to\mathbb{C} of a Schwartz function ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R}, \mathbb{C}) is again a Schwartz function, i.e., Kψ∈S(R2,C)K_{\psi} \in\mathcal{S}(\mathbb{R}^{2}, \mathbb{C}).

Proof. Since ψ∈C∞(R,C)\psi\in C^{\infty}(\mathbb{R}, \mathbb{C}) also ψ‾∈C∞(R,C)\overline{\psi} \in C^{\infty}(\mathbb{R}, \mathbb{C}) and their product. The norm factors

∥Kψ∥α,β=sup⁡x∈R2∣xαDβψ(x1)ψ(x2)‾∣(12)\|K_{\psi}\|_{\alpha,\beta} = \sup_{x\in\mathbb{R}^{2}}\left|x^{\alpha}D^{\beta}\psi(x_{1})\overline{\psi(x_{2})}\right| \tag*{(12)}
=sup⁡x∈R2∣x1α1Dβ1ψ(x1)x2α2Dβ2ψ(x2)‾∣(13)= \sup_{x\in\mathbb{R}^{2}}\left|x_{1}^{\alpha_{1}}D^{\beta_{1}}\psi(x_{1})x_{2}^{\alpha_{2}}D^{\beta_{2}}\overline{\psi(x_{2})}\right| \tag*{(13)}
=sup⁡x∈R2∣x1α1Dβ1ψ(x1)∣∣x2α2Dβ2ψ(x2)‾∣(14)= \sup_{x\in\mathbb{R}^{2}}\left|x_{1}^{\alpha_{1}}D^{\beta_{1}}\psi(x_{1})\right|\left|x_{2}^{\alpha_{2}}D^{\beta_{2}}\overline{\psi(x_{2})}\right| \tag*{(14)}
≤(sup⁡x1∈R∣x1α1Dβ1ψ(x1)∣)(sup⁡x2∈R∣x2α2Dβ2ψ(x2)‾∣)(15)\leq\left(\sup_{x_{1}\in\mathbb{R}}\left|x_{1}^{\alpha_{1}}D^{\beta_{1}}\psi(x_{1})\right|\right)\left(\sup_{x_{2}\in\mathbb{R}}\left|x_{2}^{\alpha_{2}}D^{\beta_{2}}\overline{\psi(x_{2})}\right|\right) \tag*{(15)}
≤∥ψ∥α1,β1∥ψ‾∥α2,β2<∞(16)\leq\|\psi\|_{\alpha_{1},\beta_{1}}\|\overline{\psi}\|_{\alpha_{2},\beta_{2}} < \infty \tag*{(16)}

:::

Theorem 20 (Kernel reconstruction). Given the Wigner function Wψ:R2→RW_{\psi}: \mathbb{R}^{2} \to\mathbb{R} of some function ψ∈S(R,C)∗wecanreconstructitskernel∗\psi\in\mathcal{S}(\mathbb{R}, \mathbb{C}) *we can reconstruct its kernel*K_{\psi}$ by

Kψ(x,y)=∫Reiℏp(x−y)Wψ(x+y2,p) dp.(17)K_{\psi}(x,y) = \int_{\mathbb{R}} e^{\frac{i}{\hbar}p(x-y)}W_{\psi}\left(\frac{x+y}{2},p\right)\,\mathrm{d}p . \tag*{(17)}

Proof. Plugging def. 10 into eq. (21) yields1

Kψ(x,y)=12πℏ∫Reiℏp(x−y)(∫Re−iℏpzψ(x+y+z2)ψ(x+y−z2)‾ dz)dp(18)K_{\psi}(x,y) = \frac{1}{2\pi\hbar}\int_{\mathbb{R}}e^{\frac{i}{\hbar}p(x-y)}\left(\int_{\mathbb{R}}e^{-\frac{i}{\hbar}pz}\psi\left(\frac{x+y+z}{2}\right)\overline{\psi\left(\frac{x+y-z}{2}\right)}\,\mathrm{d}z\right)\mathrm{d}p \tag*{(18)}

where the inner integral over zz is the Fourier transform of a(z)=ψ(x+y+z2)ψ(x+y−z2)‾a(z)=\psi\left(\frac{x+y+z}{2}\right)\overline{\psi\left(\frac{x+y-z}{2}\right)} leaving

Kψ(x,y)=∫Reiℏp(x−y)a^(p) dp=a(x−y)(19)K_\psi(x,y)=\int_{\mathbb{R}}e^{\frac{i}{\hbar}p(x-y)}\widehat{a}(p)\,\mathrm{d}p=a(x-y) \tag*{(19)}

which then simplifies, after inserting the definition of aa again, to

Kψ(x,y)=a(x−y)=ψ(x+y+x−y2)ψ(x+y−(x−y)2)‾=ψ(x)ψ(y)‾(20)K_\psi(x,y)=a(x-y)=\psi\left(\frac{x+y+x-y}{2}\right)\overline{\psi\left(\frac{x+y-(x-y)}{2}\right)}=\psi(x)\overline{\psi(y)} \tag*{(20)}

as expected. □

Alternatively we can prove this using a generalized notion of orthogonality for the de Broglie family eiℏpxe^{\frac{i}{\hbar}px}. See appendix § 5.§ 2 for more.

Theorem 25 (Wavefunction reconstruction). Given the kernel Kψ:R2→CK_\psi:\mathbb{R}^2\to\mathbb{C} of some (normalized) function ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R},\mathbb{C}), we can find a function φ∈S(R,C)\varphi\in\mathcal{S}(\mathbb{R},\mathbb{C}) with the same kernel. For that, find a point y0y_0 where KψK_\psi does not vanish and fix the phase convention

φ(y0)=Kψ(y0,y0)>0.(21)\varphi(y_0)=\sqrt{K_\psi(y_0,y_0)}>0. \tag*{(21)}

Then set

φ(x)=Kψ(x,y0)Kψ(y0,y0).(22)\varphi(x)=\frac{K_\psi(x,y_0)}{\sqrt{K_\psi(y_0,y_0)}}. \tag*{(22)}

The (re)constructed function φ\varphi only differs from ψ\psi by a phase, i.e., φ=eiθψ\varphi=e^{i\theta}\psi.

Proof. Notice that diagonal entries in the kernel are necessarily real and the square root is therefore well-defined, real and positive since we picked a point y0y_0 where the kernel does not vanish, which must exist for the kernel of anything but the zero function.

φ(x)=Kψ(x,y0)Kψ(y0,y0)=ψ(x)ψ(y0)‾ψ(y0)ψ(y0)‾=ψ(y0)‾∣ψ(y0)∣ψ(x)=eiθψ(x)(23)\varphi(x)=\frac{K_\psi(x,y_0)}{\sqrt{K_\psi(y_0,y_0)}}=\frac{\psi(x)\overline{\psi(y_0)}}{\sqrt{\psi(y_0)\overline{\psi(y_0)}}}=\frac{\overline{\psi(y_0)}}{|\psi(y_0)|}\psi(x)=e^{i\theta}\psi(x) \tag*{(23)}

Theorem 29 (Reconstruction). From the previous two theorems it immediately follows that we can reconstruct any (normalized) function ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R},\mathbb{C}) from its Wigner function WψW_\psi, up to a global phase eiθe^{i\theta}.

Weyl–Wigner Correspondence

Definition 30. For a function w:R2→C∈OM(R2)w:\mathbb{R}^2\to\mathbb{C}\in O_M(\mathbb{R}^2) its Weyl quantization Ow:S(R,C)→S(R,C)O_w:\mathcal{S}(\mathbb{R},\mathbb{C})\to\mathcal{S}(\mathbb{R},\mathbb{C}) is given by

(Owφ)(x)=12πℏ∫Reiℏpx(∫Re−iℏpyw(x+y2,p)φ(y) dy)dp.(24)(O_w\varphi)(x)=\frac{1}{2\pi\hbar}\int_{\mathbb{R}}e^{\frac{i}{\hbar}px}\left(\int_{\mathbb{R}}e^{-\frac{i}{\hbar}py}w\left(\frac{x+y}{2},p\right)\varphi(y)\,\mathrm{d}y\right)\mathrm{d}p. \tag*{(24)}

Compare eq. (2.3) in [7].

Theorem 32 (Weyl–Wigner identity). The expectation values are preserved when transforming operators to functions on phase space and wave functions to Wigner functions, i.e.,

⟨ψ∣Owψ⟩=∬R2w(x,p)Wψ(x,p) dx dp.(25)\langle\psi|O_w\psi\rangle=\iint_{\mathbb{R}^2}w(x,p)W_\psi(x,p)\,\mathrm{d}x\,\mathrm{d}p. \tag*{(25)}

See Prop. 2.5 in [7].

Weyl-Heisenberg Operators

We consider the family of unitary operators W(a,b):S(R,C)→S(R,C)W(a,b): S(\mathbb{R},\mathbb{C}) \to S(\mathbb{R},\mathbb{C}) defined for any a,b∈Ra,b \in\mathbb{R} as

W(a,b)=exp⁡(iℏ(aX+bP))(26)W(a,b)=\exp\left(\frac{i}{\hbar}(aX+bP)\right) \tag*{(26)}

where the position operator X:S(R,C)→S(R,C)X:S(\mathbb{R},\mathbb{C}) \to S(\mathbb{R},\mathbb{C}) and the momentum/derivative operator P:S(R,C)→S(R,C)P:S(\mathbb{R},\mathbb{C}) \to S(\mathbb{R},\mathbb{C}) are defined respectively as

(Xψ)(x)=xψ(x)(27)(X\psi)(x)=x\psi(x) \tag*{(27)}
(Pψ)(x)=−iℏdψdx(x).(28)(P\psi)(x)=-i\hbar\frac{d\psi}{dx}(x). \tag*{(28)}

Both of these operators are defined everywhere, continuous in the Schwartz topology, and symmetric, which is the same as self-adjoint in this case on S(R,C)S(\mathbb{R},\mathbb{C}). The exponential function can be defined in terms of its power series, which would not be possible on L2L^2. Alternatively, they can be defined via translations and modulations.

Translation For b∈Rb \in\mathbb{R}, define Tb:S(R,C)→S(R,C)T_b:S(\mathbb{R},\mathbb{C}) \to S(\mathbb{R},\mathbb{C}) as

(Tbφ)(x)=φ(x+b).(29)(T_b\varphi)(x)=\varphi(x+b). \tag*{(29)}

Modulation For a∈Ra \in\mathbb{R}, define Ma:S(R,C)→S(R,C)M_a:S(\mathbb{R},\mathbb{C}) \to S(\mathbb{R},\mathbb{C}) by

(Maφ)(x)=eiℏaxφ(x).(30)(M_a\varphi)(x)=e^{\frac{i}{\hbar}ax}\varphi(x). \tag*{(30)}

Definition 39. For (a,b)∈R2(a,b) \in\mathbb{R}^2, define the Weyl–Heisenberg operator on S(R,C)S(\mathbb{R},\mathbb{C}) by

W(a,b)=ei2ℏabMaTb.(31)W(a,b)=e^{\frac{i}{2\hbar}ab}M_aT_b. \tag*{(31)}

Equivalently, for φ∈S(R,C)\varphi\in S(\mathbb{R},\mathbb{C}),

(W(a,b)φ)(x)=eiℏa(x+b2)φ(x+b).(32)(W(a,b)\varphi)(x)=e^{\frac{i}{\hbar}a\left(x+\frac{b}{2}\right)}\varphi(x+b). \tag*{(32)}

Reconstructions

We now proceed with the next steps in our reconstruction.

Characteristic function

Definition 42. The characteristic function χψ:R2→C\chi_\psi:\mathbb{R}^2 \to\mathbb{C} of a function ψ∈S(R,C)\psi\in S(\mathbb{R},\mathbb{C}) is

χψ(a,b)=⟨ψ∣W(a,b)ψ⟩.(33)\chi_\psi(a,b)=\langle\psi\vert W(a,b)\psi\rangle. \tag*{(33)}

Proposition 44 (Fourier relation). For ψ∈S(R,C)\psi\in S(\mathbb{R},\mathbb{C}),

χψ(a,b)=∬R2eiℏ(ax+bp)Wψ(x,p) dx dp,(34)\chi_\psi(a,b)=\iint_{\mathbb{R}^2}e^{\frac{i}{\hbar}(ax+bp)}W_\psi(x,p)\,dx\,dp, \tag*{(34)}

and conversely,

Wψ(x,p)=1(2πℏ)2∬R2e−iℏ(ax+bp)χψ(a,b) da db.(35)W_\psi(x,p)=\frac{1}{(2\pi\hbar)^2}\iint_{\mathbb{R}^2}e^{-\frac{i}{\hbar}(ax+bp)}\chi_\psi(a,b)\,da\,db. \tag*{(35)}

Proof. The first relationship follows directly from Theorem 32 since W(a,b)W(a,b) is the Weyl quantization for fixed aa and bb of

w(x,p)=eiℏ(ax+bp)(36)w(x,p)=e^{\frac{i}{\hbar}(ax+bp)} \tag*{(36)}

To see this explicitly:

(W(a,b)φ)(x)=12πℏ∫Reiℏpx(∫Re−iℏpyeiℏ(ax+y2+bp)φ(y) dy)dp(37)(W(a,b)\varphi)(x)=\frac{1}{2\pi\hbar}\int_{\mathbb{R}}e^{\frac{i}{\hbar}px}\left(\int_{\mathbb{R}}e^{-\frac{i}{\hbar}py}e^{\frac{i}{\hbar}\left(\frac{a x+y}{2}+bp\right)}\varphi(y)\,\mathrm{d}y\right)\mathrm{d}p \tag*{(37)}
=eiℏa2x∫Reiℏp(x+b)(12πℏ∫Re−iℏ(p−a2)yφ(y) dy)dp(38)=e^{\frac{i}{\hbar}\frac{a}{2}x}\int_{\mathbb{R}}e^{\frac{i}{\hbar}p(x+b)}\left(\frac{1}{2\pi\hbar}\int_{\mathbb{R}}e^{-\frac{i}{\hbar}(p-\frac{a}{2})y}\varphi(y)\,\mathrm{d}y\right)\mathrm{d}p \tag*{(38)}
=eiℏa2x∫Reiℏp(x+b)φ~(p−a2) dp(39)=e^{\frac{i}{\hbar}\frac{a}{2}x}\int_{\mathbb{R}}e^{\frac{i}{\hbar}p(x+b)}\widetilde{\varphi}\left(p-\frac{a}{2}\right)\,\mathrm{d}p \tag*{(39)}
=eiℏa2x∫Reiℏ(p′+a2)(x+b)φ~(p′) dp′(40)=e^{\frac{i}{\hbar}\frac{a}{2}x}\int_{\mathbb{R}}e^{\frac{i}{\hbar}(p'+\frac{a}{2})(x+b)}\widetilde{\varphi}(p')\,\mathrm{d}p' \tag*{(40)}
=eiℏ(a2x+a2(x+b))∫Reiℏp′(x+b)φ~(p′) dp′(41)=e^{\frac{i}{\hbar}(\frac{a}{2}x+\frac{a}{2}(x+b))}\int_{\mathbb{R}}e^{\frac{i}{\hbar}p'(x+b)}\widetilde{\varphi}(p')\,\mathrm{d}p' \tag*{(41)}
=eiℏa(x+b2)φ(x+b)(42)=e^{\frac{i}{\hbar}a(x+\frac{b}{2})}\varphi(x+b) \tag*{(42)}

The inverse relationship follows immediately, once we recognize the first one to be the Fourier transform.

Corollary 54. Since the Fourier transform preserves the Schwartz space we conclude that the characteristic function χψ\chi_\psi of a Schwartz function ψ\psi is again a Schwartz function.

Theorem 55. The expectation values of W(a,b)W(a,b)

⟨ψ∣W(a,b)ψ⟩(43)\langle\psi\mid W(a,b)\psi\rangle \tag*{(43)}

uniquely define the state ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R},\mathbb{C}) up to a global phase.

Proof. The expectation values, i.e., values of the characteristic function, are the Fourier transform of the Wigner function. Since the Fourier transform is invertible and the Wigner function uniquely determines the state, we can also recover the state from these expectation values.

Probability distributions

We will define the probability distribution of linear combinations of XX and PP like2

X(α)=cos⁡αX+sin⁡αP.(44)\mathcal{X}(\alpha)=\cos\alpha X+\sin\alpha P. \tag*{(44)}

Proposition 58. This parametrized curve can be generated by the operator A:S(R,C)→S(R,C)A:\mathcal{S}(\mathbb{R},\mathbb{C})\to\mathcal{S}(\mathbb{R},\mathbb{C}), defined as3

A=12(X2+P2),Uα=e−iℏαA(45)A=\frac{1}{2}\left(X^2+P^2\right),\qquad U_\alpha=e^{-\frac{i}{\hbar}\alpha A} \tag*{(45)}
X(α)=UαXUα†=cos⁡αX+sin⁡αP.(46)\mathcal{X}(\alpha)=U_\alpha XU_\alpha^\dagger=\cos\alpha X+\sin\alpha P. \tag*{(46)}

Note that there are no domain problems for AA since XX and PP are already defined everywhere and map back to S(R,C)\mathcal{S}(\mathbb{R},\mathbb{C}).

Proof. X(α)X(\alpha) satisfies the following 2nd order ODE:

d2Xdα2(α)=[[X(α),iℏA],iℏA]=[P(α),iℏA]=−X(α)(47)\frac{\mathrm{d}^2X}{\mathrm{d}\alpha^2}(\alpha)=\left[\left[X(\alpha),\frac{i}{\hbar}A\right],\frac{i}{\hbar}A\right]=\left[P(\alpha),\frac{i}{\hbar}A\right]=-X(\alpha) \tag*{(47)}

which is solved by the Ansatz: X(α)=a(α)X+b(α)PX(\alpha)=a(\alpha)X+b(\alpha)P. Then aa and bb satisfy

d2adα2=−a(α)d2bdα2=−b(α)(48)\frac{\mathrm{d}^2a}{\mathrm{d}\alpha^2}=-a(\alpha)\qquad\frac{\mathrm{d}^2b}{\mathrm{d}\alpha^2}=-b(\alpha) \tag*{(48)}
dadα∣α=0=0dbdα∣α=0=1(49)\left.\frac{\mathrm{d}a}{\mathrm{d}\alpha}\right|_{\alpha=0}=0\qquad\left.\frac{\mathrm{d}b}{\mathrm{d}\alpha}\right|_{\alpha=0}=1 \tag*{(49)}
a(0)=1b(0)=0(50)a(0)=1\qquad b(0)=0 \tag*{(50)}

and have well-known solutions:

a(α)=cos⁡(α)b(α)=sin⁡(α)(51)a(\alpha)=\cos(\alpha)\quad b(\alpha)=\sin(\alpha) \tag*{(51)}
⟹X(α)=cos⁡(α)X+sin⁡(α)P(52)\Longrightarrow\mathcal{X}(\alpha)=\cos(\alpha)X+\sin(\alpha)P \tag*{(52)}

See appendix § 5.§ 4 for more detailed calculations.

Similarly, we can transform the initial probability distribution in X=X(0)X=\mathcal{X}(0) for a given state ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R},\mathbb{C}) as

ρ0(q)=∣ψ(q)∣2 ⟶ ρα(q)=∣(Uαψ)(q)∣2(53)\rho_0(q)=\lvert\psi(q)\rvert^2\ \longrightarrow\ \rho_\alpha(q)=\lvert(U_\alpha\psi)(q)\rvert^2 \tag*{(53)}

Theorem 68. The Fourier transform ρ^α\hat{\rho}_\alpha of the probability distribution ρα\rho_\alpha of X(α)\mathcal{X}(\alpha) (as defined above) is related to the characteristic function χψ\chi_\psi as follows

ρ^α(s)≔∫Reiℏsqρα(q) dq=χψ(scos⁡α,ssin⁡α).(54)\hat{\rho}_\alpha(s)\coloneqq\int_{\mathbb{R}}e^{\frac{i}{\hbar}sq}\rho_\alpha(q)\,\mathrm{d}q=\chi_\psi(s\cos\alpha,s\sin\alpha). \tag*{(54)}

Proof. Let us call ψα=Uαψ\psi_\alpha=U_\alpha\psi.

ρ^α(s)=∫Rψα(q)‾eiℏsqψα(q) dq(55)\hat{\rho}_\alpha(s)=\int_{\mathbb{R}}\overline{\psi_\alpha(q)}e^{\frac{i}{\hbar}sq}\psi_\alpha(q)\,\mathrm{d}q \tag*{(55)}
=⟨Uαψ∣eiℏsXUαψ⟩(56)=\langle U_\alpha\psi\lvert e^{\frac{i}{\hbar}sX}U_\alpha\psi\rangle \tag*{(56)}
=⟨ψ∣Uα†eiℏsXUαψ⟩(57)=\langle\psi\lvert U_\alpha^\dagger e^{\frac{i}{\hbar}sX}U_\alpha\psi\rangle \tag*{(57)}
=⟨ψ∣eiℏsUα†XUαψ⟩(58)=\langle\psi\lvert e^{\frac{i}{\hbar}sU_\alpha^\dagger XU_\alpha}\psi\rangle \tag*{(58)}
=⟨ψ∣eiℏsX(α)ψ⟩(59)=\langle\psi\lvert e^{\frac{i}{\hbar}s\mathcal{X}(\alpha)}\psi\rangle \tag*{(59)}
=⟨ψ∣eiℏs(cos⁡αX+sin⁡αP)ψ⟩(60)=\langle\psi\lvert e^{\frac{i}{\hbar}s(\cos\alpha X+\sin\alpha P)}\psi\rangle \tag*{(60)}
=⟨ψ∣W(scos⁡α,ssin⁡α)ψ⟩(61)=\langle\psi\lvert W(s\cos\alpha,s\sin\alpha)\psi\rangle \tag*{(61)}
=χψ(scos⁡α,ssin⁡α)(62)=\chi_\psi(s\cos\alpha,s\sin\alpha) \tag*{(62)}

Note that ρ^α\hat{\rho}_\alpha is again a Schwartz function for every α\alpha.

Theorem 78. Given the probability distributions ρα\rho_\alpha with α∈[0,π)\alpha\in[0,\pi) of the observables X(α)=cos⁡αX+sin⁡αP\mathcal{X}(\alpha)=\cos\alpha X+\sin\alpha P for a state ψ∈S(R,C)\psi\in\mathcal{S}(\mathbb{R},\mathbb{C}), as defined above, we can reconstruct the original state.

Proof. As shown in theorem 68, we can reconstruct the characteristic function χψ\chi_\psi from ρ^α\hat{\rho}_\alpha, which we obtain from ρα\rho_\alpha using the inverse Fourier transform. Then theorem 55 tells us that we can reconstruct the original state from χψ\chi_\psi.

Physical interpretation

As has been noted above the generator of the family is the Hamiltonian of the harmonic oscillator. Physically, this means that the family can be interpreted as the position operator evolving under the harmonic potential in the Heisenberg picture, and the parameter α\alpha represents time. However, this is not the only Hamiltonian one can use for tomography.

For free evolution, we know classically x(t)=x(0)+pmt=x(0)+tmpx(t)=x(0)+\frac{p}{m}t=x(0)+\frac{t}{m}p. Setting tan⁡(α)=tm\tan(\alpha)=\frac{t}{m} we can see that this curve of operators corresponds to observing the position XX over time tt during free evolution.

X~(t)=1cos⁡(α(t))X(α(t))=X+tan⁡(α)P=X+tmP(63)\widetilde{\mathcal{X}}(t)=\frac{1}{\cos(\alpha(t))}\mathcal{X}(\alpha(t))=X+\tan(\alpha)P=X+\frac{t}{m}P \tag*{(63)}

Note that while the point α=π/2\alpha=\pi/2 is never reached we can get arbitrarily close and since expectation values are continuous the single point does not contain any further data.

To see this explicitly, define the Hamiltonian for free evolution.

H=12mP2(64)H=\frac{1}{2m}P^2 \tag*{(64)}

It generates time evolution via

Ut=e−iℏHt(65)U_t = e^{-\frac{i}{\hbar}Ht} \tag*{(65)}

and therefore

ψt=Utψ(66)\psi_t = U_t\psi \tag*{(66)}

The probability density for the distribution of position XX in the state ϕt\phi_t at time tt is then

ρt(x)=∣ψt(x)∣2=∣(Utψ)(x)∣2=∣(e−iℏHtψ)(x)∣2=∣(e−i2mℏP2tψ)(x)∣2.(67)\rho_t(x) = |\psi_t(x)|^2 = |(U_t\psi)(x)|^2 = \left|\left(e^{-\frac{i}{\hbar}Ht}\psi\right)(x)\right|^2 = \left|\left(e^{-\frac{i}{2m\hbar}P^2t}\psi\right)(x)\right|^2. \tag*{(67)}

In the Heisenberg picture this corresponds to the distribution of the observable UtXUt†U_t X U_t^\dagger. See appendix § 5.5 for a derivation.

Conclusion

We presented a short proof of several tomographic characterizations of Schwartz wavefunctions. The proof requires only basic tools from analysis, which we hope will make it accessible to a broad audience. The characterization in terms of the probability distribution ρα\rho_\alpha at specific angles α\alpha in phase space has a precise analogue in classical mechanics. Under harmonic evolution, probability densities on phase space rotate around the origin (for some choice of units). The marginals for all rotation angles α\alpha uniquely determine the full phase-space measure by application of the Cramér–Wold theorem [6, 11].

This result sits within our larger goal of understanding what vector space representation of quantum states can be both mathematically precise and physically meaningful. The result shows that, at least for quantum tomography, the Schwartz spaces provide analogous results that are independent of the Hilbert space theory. While Schwartz spaces are limited to quantum systems describing only positional degrees of freedom of particles, and therefore cannot be the ultimate solution, studying Schwartz analogues may pave the way to find better alternatives to the standard Hilbert space formalism.

Appendix

Further mathematical prerequisites

We provide some further properties of Schwartz spaces that allow us to use different techniques that are much closer to common practice in physics to prove some of the results in this paper.

Multipliers Schwartz spaces are closed under multiplication by any function in OM(Rn)O_M(\mathbb{R}^n), the space of slowly increasing smooth function, i.e., all f∈C∞(Rn,C)f \in C^\infty(\mathbb{R}^n,\mathbb{C}), s.t.

∀α∈Nn,∃k∈N,C≥0,∀x∈Rn:∣∂αf(x)∣≤C(1+∣x∣2)k,(68)\forall\alpha\in\mathbb{N}^n, \exists k \in\mathbb{N}, C \ge0, \forall x \in\mathbb{R}^n : \lvert\partial^\alpha f(x) \rvert\le C(1+\lvert x\rvert^2)^k, \tag*{(68)}

which are precisely those functions that preserve Schwartz spaces under multiplication.

Tempered distributions In contrast to L2L^2, which is self-dual, the dual4 space of S\mathcal{S} is much bigger and contains elements like the Dirac distribution5

δx[f]=f(x)=∫Rnδ(y−x)f(y) dny(69)\delta_x[f] = f(x) = \int_{\mathbb{R}^n} \delta(y-x)f(y)\,\mathrm{d}^n y \tag*{(69)}

which cannot be defined on L2L^2 since it does not contain functions but equivalence classes of functions. The elements of this dual space S′\mathcal{S}' are called tempered distributions.

Integral and Dirac measure Taking the integral above literally requires some further qualifications, since there is no function δ:Rn→R\delta: \mathbb{R}^n \to\mathbb{R} that would satisfy eq. (85). However, δx∈S′\delta_x \in\mathcal{S}' can be multiplied by f∈S(Rn,C)f \in\mathcal{S}(\mathbb{R}^n,\mathbb{C}) (or in general OM(Rn)O_M(\mathbb{R}^n)) to give another tempered distribution, which can be evaluated at the constant function 1.

δx[f]=δx[f⋅1]=(δx⋅f)[1]=∫Rn1(fδx)(70)\delta_x[f] = \delta_x[f \cdot1] = (\delta_x \cdot f)[1] = \int_{\mathbb{R}^n} 1(f\delta_x) \tag*{(70)}

The first multiplication ⋅\cdot is between an element of OM(Rn)O_M(\mathbb{R}^n) and S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}), while the second multiplication is between S′\mathcal{S}' and S(Rn,C)\mathcal{S}(\mathbb{R}^n,\mathbb{C}) and defined in terms of the first, just as written. Note that even though 1∉S(Rn,C)1 \notin\mathcal{S}(\mathbb{R}^n,\mathbb{C}) we can evaluate δx⋅f\delta_x \cdot f on any OM(Rn)O_M(\mathbb{R}^n) function and any indicator function 1U1_U of some Borel set U⊆RnU \subseteq\mathbb{R}^n. This defines a measure

μ(U):=(δx⋅f)[1U](71)\mu(U) := (\delta_x \cdot f)[1_U] \tag*{(71)}

which can be integrated over Rn\mathbb{R}^n to give the same result.

δx[f]=∫Rndμ(72)\delta_x[f] = \int_{\mathbb{R}^n} \mathrm{d}\mu \tag*{(72)}

Approximation by integrals Another justification for using the integral notation is that S\mathcal{S} is dense in S′\mathcal{S}', i.e., every linear functional in S′\mathcal{S}' can be approximated by integrals of the form

∫Rnfi(x)g(x) dx→F[g],(73)\int_{\mathbb{R}^n} f_i(x)g(x)\,\mathrm{d}x \to F[g], \tag*{(73)}

where fi∈Sf_i \in\mathcal{S}. The δx\delta_x distribution for example can be approximated by Gaussian probability density functions with mean xx and standard deviation σ→0\sigma\to0.

Alternative proof using orthogonal de Broglie family

This is an alternative proof of theorem 20 that uses a generalized notion of orthogonality for the de Broglie family eiℏpxe^{\frac{i}{\hbar}px}.

Proof. Plugging def. 11 into eq. (21) yields

Kψ(x,y)=12πℏ∫R(∫Re−iℏpzψ(x+y+z2)ψ(x+y−z2)‾ dz)dp(74)K_{\psi}(x,y)=\frac{1}{2\pi\hbar}\int_{\mathbb{R}}\left(\int_{\mathbb{R}}e^{-\frac{i}{\hbar}pz}\psi\left(\frac{x+y+z}{2}\right)\overline{\psi\left(\frac{x+y-z}{2}\right)}\,\mathrm{d}z\right)\mathrm{d}p \tag*{(74)}

where the inner integral over zz can be interpreted as the application of the tempered distributions ηp=⟨eiℏpz|⋅⟩\eta_p=\left\langle e^{\frac{i}{\hbar}pz}\middle|\cdot\right\rangle parametrized by pp to the function a(z)=ψ(x+y+z2)ψ(x+y−z2)‾a(z)=\psi\left(\frac{x+y+z}{2}\right)\overline{\psi\left(\frac{x+y-z}{2}\right)}. The result is a function of pp evaluated by the tempered distribution ηx−y\eta_{x-y}. In formulas,

Kψ(x,y)=12πℏηx−y[p↦ηp∗[a]].(75)K_{\psi}(x,y)=\frac{1}{2\pi\hbar}\eta_{x-y}\left[p\mapsto\eta_p^*[a]\right]. \tag*{(75)}

By delaying the evaluation at aa we can generalize the inner product

Kψ(x,y)=12πℏ⟨η∣ηx−y⟩[a]=δx−y[a]=a(x−y)(76)K_{\psi}(x,y)=\frac{1}{2\pi\hbar}\langle\eta\mid\eta_{x-y}\rangle[a]=\delta_{x-y}[a]=a(x-y) \tag*{(76)}

and use the fact that η\eta is an orthogonal family in that generalized sense (see Refs. [3, 4]) to simplify this to

Kψ(x,y)=ψ(x+y+x−y2)ψ(x+y−(x−y)2)‾=ψ(x)ψ(y)‾(77)K_{\psi}(x,y)=\psi\left(\frac{x+y+x-y}{2}\right)\overline{\psi\left(\frac{x+y-(x-y)}{2}\right)}=\psi(x)\overline{\psi(y)} \tag*{(77)}

as expected.

The conjugation identity

We justify the identity

Uα†eiℏsXUα=eiℏsUα†XUα.(78)U_{\alpha}^{\dagger}e^{\frac{i}{\hbar}sX}U_{\alpha}=e^{\frac{i}{\hbar}sU_{\alpha}^{\dagger}XU_{\alpha}}. \tag*{(78)}

First, conjugation respects products:

Uα†XnUα=(Uα†XUα)n∀n∈N0.(79)U_{\alpha}^{\dagger}X^{n}U_{\alpha}=(U_{\alpha}^{\dagger}XU_{\alpha})^{n}\qquad\forall n\in\mathbb{N}_{0}. \tag*{(79)}

Second, the exponential of an operator is defined by the power series

eX=∑n=0∞Xnn!.(80)e^{X}=\sum_{n=0}^{\infty}\frac{X^{n}}{n!}. \tag*{(80)}

Applying conjugation term by term, we get

Uα†eiℏsXUα=∑n=0∞1n!Uα†(iℏsX)nUα=∑n=0∞1n!(iℏsUα†XUα)n,(81)U_{\alpha}^{\dagger}e^{\frac{i}{\hbar}sX}U_{\alpha}=\sum_{n=0}^{\infty}\frac{1}{n!}U_{\alpha}^{\dagger}\left(\frac{i}{\hbar}sX\right)^{n}U_{\alpha}=\sum_{n=0}^{\infty}\frac{1}{n!}\left(\frac{i}{\hbar}sU_{\alpha}^{\dagger}XU_{\alpha}\right)^{n}, \tag*{(81)}

which is exactly

eiℏsUα†XUα.(82)e^{\frac{i}{\hbar}sU_{\alpha}^{\dagger}XU_{\alpha}}. \tag*{(82)}

This argument is fully legitimate on S(R)\mathcal{S}(\mathbb{R}), since S(R)\mathcal{S}(\mathbb{R}) is stable under XX, HH, UαU_{\alpha}, and all powers appearing in the series.

Alternative proof

Alternatively, use the fact that XX has an eigenbasis of Dirac delta δq\delta_q when extended to S′\mathcal{S}'.

Xδq=qδq(83)X\delta_q=q\delta_q \tag*{(83)}

and can be rotated by the unitary evolution into an eigenbasis βα,q=Uα†δq\beta_{\alpha,q}=U_\alpha^\dagger\delta_q for (the extension of) the conjugated operator X(α)=Uα†XUαX(\alpha)=U_\alpha^\dagger XU_\alpha.

X(α)βα,q=Uα†XUαUα†δq=Uα†Xδq=Uα†qδq=qβα,q(84)X(\alpha)\beta_{\alpha,q}=U_\alpha^\dagger XU_\alpha U_\alpha^\dagger\delta_q=U_\alpha^\dagger X\delta_q=U_\alpha^\dagger q\delta_q=q\beta_{\alpha,q} \tag*{(84)}

It is now easy to see that the left- and right-hand side of eq. (94) act in the same way on this basis (and therefore everywhere).

The left-hand side evaluates to

Uα†eiℏsXUαβα,q=Uα†eiℏsXUαUα†δq(85)U_\alpha^\dagger e^{\frac{i}{\hbar}sX}U_\alpha\beta_{\alpha,q}=U_\alpha^\dagger e^{\frac{i}{\hbar}sX}U_\alpha U_\alpha^\dagger\delta_q \tag*{(85)}
=Uα†eiℏsXδq(86)=U_\alpha^\dagger e^{\frac{i}{\hbar}sX}\delta_q \tag*{(86)}
=Uα†eiℏsqδq(87)=U_\alpha^\dagger e^{\frac{i}{\hbar}sq}\delta_q \tag*{(87)}
=eiℏsqUα†δq(88)=e^{\frac{i}{\hbar}sq}U_\alpha^\dagger\delta_q \tag*{(88)}
=eiℏsqβα,q(89)=e^{\frac{i}{\hbar}sq}\beta_{\alpha,q} \tag*{(89)}

The right-hand side trivially simplifies to

eiℏsUα†XUαβα,q=eiℏsX(α)βα,q=eiℏsqβα,q(90)e^{\frac{i}{\hbar}sU_\alpha^\dagger XU_\alpha}\beta_{\alpha,q}=e^{\frac{i}{\hbar}sX(\alpha)}\beta_{\alpha,q}=e^{\frac{i}{\hbar}sq}\beta_{\alpha,q} \tag*{(90)}

since βα,q\beta_{\alpha,q} is an eigenvector of X(α)X(\alpha) with eigenvalue qq.

Solution of the Heisenberg equation for the harmonic oscillator

A=12(X2+P2),Uα=e−iℏαA(91)A=\frac{1}{2}\left(X^2+P^2\right),\qquad U_\alpha=e^{-\frac{i}{\hbar}\alpha A} \tag*{(91)}
X(α)=UαXUα†=cos⁡α X+sin⁡α P.(92)X(\alpha)=U_\alpha XU_\alpha^\dagger=\cos\alpha\,X+\sin\alpha\,P. \tag*{(92)}

The first derivative is

dXdα=ddαUαXUα†=ddαe−iℏαAXeiℏαA(93)\frac{\mathrm{d}X}{\mathrm{d}\alpha}=\frac{\mathrm{d}}{\mathrm{d}\alpha}U_\alpha XU_\alpha^\dagger=\frac{\mathrm{d}}{\mathrm{d}\alpha}e^{-\frac{i}{\hbar}\alpha A}Xe^{\frac{i}{\hbar}\alpha A} \tag*{(93)}
=−iℏA(e−iℏαAXeiℏαA)+(e−iℏαAXeiℏαA)iℏA(94)=-\frac{i}{\hbar}A\left(e^{-\frac{i}{\hbar}\alpha A}Xe^{\frac{i}{\hbar}\alpha A}\right)+\left(e^{-\frac{i}{\hbar}\alpha A}Xe^{\frac{i}{\hbar}\alpha A}\right)\frac{i}{\hbar}A \tag*{(94)}
=[X(α),iℏA]=iℏ[X(α),A].(95)=\left[X(\alpha),\frac{i}{\hbar}A\right]=\frac{i}{\hbar}[X(\alpha),A]. \tag*{(95)}

The second derivative is

d2Xdα2=ddαiℏ[X(α),A]=iℏ[ddαX(α),A](96)\frac{\mathrm{d}^2X}{\mathrm{d}\alpha^2}=\frac{\mathrm{d}}{\mathrm{d}\alpha}\frac{i}{\hbar}[X(\alpha),A]=\frac{i}{\hbar}\left[\frac{\mathrm{d}}{\mathrm{d}\alpha}X(\alpha),A\right] \tag*{(96)}
=[[X(α),iℏA],iℏA](97)=\left[\left[X(\alpha),\frac{i}{\hbar}A\right],\frac{i}{\hbar}A\right] \tag*{(97)}
=−1ℏ2[[X(α),A],A].(98)=-\frac{1}{\hbar^2}[[X(\alpha),A],A]. \tag*{(98)}

The relevant commutators are

[X,P]=iℏ(99)[X,P] = i\hbar \tag*{(99)}
[X,P2]=P[X,P]+[X,P]P=Piℏ+iℏP=2iℏP(100)\begin{aligned} [X,P^2] &= P[X,P] + [X,P]P \\ &= Pi\hbar+ i\hbar P = 2i\hbar P \tag*{(100)} \end{aligned}
[X,A]=[X,12(X2+P2)]=12[X,X2+P2]=12[X,P2]=iℏP(101)\begin{aligned} [X,A] &= \left[X,\frac{1}{2}(X^2+P^2)\right] = \frac{1}{2}[X,X^2+P^2] \\ &= \frac{1}{2}[X,P^2] = i\hbar P \tag*{(101)} \end{aligned}
[P,X2]=X[P,X]+[P,X]X=−Xiℏ−iℏX=−2iℏX(102)\begin{aligned} [P,X^2] &= X[P,X] + [P,X]X \\ &= -Xi\hbar- i\hbar X = -2i\hbar X \tag*{(102)} \end{aligned}
[P,A]=[P,12(X2+P2)]=12[P,X2+P2]=12[P,X2]=−iℏX(103)\begin{aligned} [P,A] &= \left[P,\frac{1}{2}(X^2+P^2)\right] = \frac{1}{2}[P,X^2+P^2] \\ &= \frac{1}{2}[P,X^2] = -i\hbar X \tag*{(103)} \end{aligned}
[Uα,A]=0=[Uα†,A](104)[U_\alpha,A] = 0 = [U_\alpha^\dagger,A] \tag*{(104)}
[UαOUα†,A]=UαOUα†A−AUαOUα†=UαOAUα†−UαAOUα†=Uα[O,A]Uα†(105)\begin{aligned} [U_\alpha O U_\alpha^\dagger,A] &= U_\alpha O U_\alpha^\dagger A - A U_\alpha O U_\alpha^\dagger\\ &= U_\alpha O A U_\alpha^\dagger- U_\alpha A O U_\alpha^\dagger\\ &= U_\alpha[O,A]U_\alpha^\dagger \tag*{(105)} \end{aligned}
⟹[X(α),A]=Uα[X,A]Uα†=iℏUαPUα†≔iℏP(α)(106)\Longrightarrow\quad[X(\alpha),A] = U_\alpha[X,A]U_\alpha^\dagger= i\hbar U_\alpha P U_\alpha^\dagger\coloneqq i\hbar P(\alpha) \tag*{(106)}
[P(α),A]=Uα[P,A]Uα†=−iℏUαXUα†=−iℏX(α).(107)[P(\alpha),A] = U_\alpha[P,A]U_\alpha^\dagger= -i\hbar U_\alpha X U_\alpha^\dagger= -i\hbar X(\alpha). \tag*{(107)}

The operator ODE is therefore

dXdα=iℏ[X(α),A]=P(α)(108)\frac{\mathrm{d}X}{\mathrm{d}\alpha} = \frac{i}{\hbar}[X(\alpha),A] = P(\alpha) \tag*{(108)}
d2Xdα2=dPdα=1ℏ2[iℏP(α),A]=−iℏ[P(α),A]=−X(α).(109)\begin{aligned} \frac{\mathrm{d}^2X}{\mathrm{d}\alpha^2} &= \frac{\mathrm{d}P}{\mathrm{d}\alpha} = \frac{1}{\hbar^2}[i\hbar P(\alpha),A] = -\frac{i}{\hbar}[P(\alpha),A] \\ &= -X(\alpha). \tag*{(109)} \end{aligned}

Let us propose the following Ansatz: X(α)=a(α)X+b(α)PX(\alpha) = a(\alpha)X + b(\alpha)P.

dXdα=dadαX+dbdαP=P(α)(110)\frac{\mathrm{d}X}{\mathrm{d}\alpha} = \frac{\mathrm{d}a}{\mathrm{d}\alpha}X + \frac{\mathrm{d}b}{\mathrm{d}\alpha}P = P(\alpha) \tag*{(110)}
d2Xdα2=d2adα2X+d2bdα2P=−X(α)=−a(α)X−b(α)P(111)\frac{\mathrm{d}^2X}{\mathrm{d}\alpha^2} = \frac{\mathrm{d}^2a}{\mathrm{d}\alpha^2}X + \frac{\mathrm{d}^2b}{\mathrm{d}\alpha^2}P = -X(\alpha) = -a(\alpha)X - b(\alpha)P \tag*{(111)}

We therefore arrive at the following ODE and initial conditions for aa and bb.

d2adα2=−a(α)d2bdα2=−b(α)(112)\frac{\mathrm{d}^2a}{\mathrm{d}\alpha^2} = -a(\alpha) \qquad \frac{\mathrm{d}^2b}{\mathrm{d}\alpha^2} = -b(\alpha) \tag*{(112)}
dadα∣α=0=0dbdα∣α=0=1(113)\left.\frac{\mathrm{d}a}{\mathrm{d}\alpha}\right|_{\alpha=0} = 0 \qquad \left.\frac{\mathrm{d}b}{\mathrm{d}\alpha}\right|_{\alpha=0} = 1 \tag*{(113)}
a(0)=1b(0)=0(114)a(0) = 1 \qquad b(0) = 0 \tag*{(114)}

The solutions are well-known:

a(α)=cos⁡(α)b(α)=sin⁡(α)(115)a(\alpha) = \cos(\alpha) \qquad b(\alpha) = \sin(\alpha) \tag*{(115)}
⟹X(α)=cos⁡(α)X+sin⁡(α)P(116)\Longrightarrow\quad X(\alpha) = \cos(\alpha)X + \sin(\alpha)P \tag*{(116)}

Solution of the Heisenberg equation for the free Hamiltonian

Define the Hamiltonian H=12mP2H = \frac{1}{2m}P^2 and Ut=eiℏtHU_t = e^{\frac{i}{\hbar}tH}. In the Heisenberg picture, operators evolve in time according to

dXdt=1iℏ[H,X(t)],(117)\frac{\mathrm{d}X}{\mathrm{d}t} = \frac{1}{i\hbar}[H,X(t)], \tag*{(117)}

which is solved by Ut†X(t=0)UtU_t^\dagger X(t=0)U_t. Here we want to show explicitly that X(t)=X(t=0)+1mPX(t)=X(t=0)+\frac{1}{m}P.

Instead of following the same technique as for the harmonic oscillator we use here the following trick: [f(P),X]=∂f∂p(P)[P,X][f(P),X]=\frac{\partial f}{\partial p}(P)[P,X].

X(t)=Ut†XUt=Ut†([X,Ut]+UtX)=Ut†[X,Ut]+Ut†UtX(118)X(t)=U_t^\dagger X U_t=U_t^\dagger\left([X,U_t]+U_tX\right)=U_t^\dagger[X,U_t]+U_t^\dagger U_tX \tag*{(118)}
=X−Ut†[Ut,X]=X−Ut†[exp⁡(1iℏt2mP2),X](119)=X-U_t^\dagger[U_t,X]=X-U_t^\dagger\left[\exp\left(\frac{1}{i\hbar}\frac{t}{2m}P^2\right),X\right] \tag*{(119)}
=X−Ut†1iℏtmP[P,X]⏟−iℏ(120)=X-U_t^\dagger\frac{1}{i\hbar}\frac{t}{m}P\underbrace{[P,X]}_{-i\hbar} \tag*{(120)}
=X+Ut†UttmP=X+tmP(121)=X+U_t^\dagger U_t\frac{t}{m}P=X+\frac{t}{m}P \tag*{(121)}

where we have used the fact that exp⁡(1iℏt2mP2)\exp\left(\frac{1}{i\hbar}\frac{t}{2m}P^2\right) commutes with PP.

References

  1. [1]J. Bertrand and P. Bertrand. “A Tomographic Approach to Wigner’s function”. In: Foundations of Physics 17.4 (1987). DOI: 10.1007/BF00733376.DOI
  2. [2]Gabriele Carcassi, Francisco Calderón, and Christine A. Aidala. “The unphysicality of Hilbert spaces”. In: Quantum Studies: Mathematics and Foundations 12.1 (Jan. 2025), p. 13. ISSN: 2196-5617. DOI: 10.1007/s40509-024-00357-0.DOI
  3. [3]David Carfì. “Dirac-orthogonality in the space of tempered distributions”. In: Journal of Computational and Applied Mathematics 153.1 (June 2001). Proceedings of the 6th International Symposium on Orthogonal Polynomials, Special Functions and their Applications, pp. 99–107. ISSN: 0377-0427. DOI: 10.1016/S0377-0427(02)00634-9.
  4. [4]David Carfì and Alessia Donato. Dirac Calculus and Schwartz Linear Algebra: Dirac inner product and Dirac orthonormality. Available as a Researchgate Paper. Oct. 2015. DOI: 10.13140/RG.2.1.3137.2885.DOI
  5. [5]G. Cassinelli et al. “Group theoretical quantum tomography”. In: Journal of Mathematical Physics 41.12 (Dec. 2000), pp. 7940–7951. ISSN: 0022-2488. DOI: 10.1063/1.1323497.DOI
  6. [6]Harald Cramér and Herman Wold. “Some Theorems on Distribution Functions”. In: Journal of the London Mathematical Society s1-11.4 (1936), pp. 290–294. DOI: 10.1112/jlms/s1-11.4.290.DOI
  7. [7]Gerald B. Folland. Harmonic Analysis in Phase Space. (AM-122). Princeton, NJ: Princeton University Press, 2016. DOI: 10.1515/9781400882427.
  8. [8]Gabriele Carcassi for the AoP Collaboration. Equivalence of sup-norm, L¹, and L² seminorms on the Schwartz space. Assumptions of Physics, Technical Brief 010, 2026. URL: https://assumptionsofphysics.org/autogen/briefs/010-SchwartzNorms.pdf.
  9. [9]J Kiukas, P Lahti, and J-P Pellonpää. “A proof for the informational completeness of the rotated quadrature observables”. In: Journal of Physics A: Mathematical and Theoretical 41.17 (Apr. 2008), p. 175206. DOI: 10.1088/1751-8113/41/17/175206.DOI
  10. [10]U. Leonhardt and M.G. Raymer. “Observation of Moving Wave Packets Reveals Their Quantum State”. In: Physical Review Letters 76.12 (1996). DOI: 10.1103/PhysRevLett.76.1985.DOI
  11. [11]Russell Lyons and Kevin Zumbrun. “A calculus proof of the Cramér-Wold theorem”. In: Proc. Amer. Math. Soc. 146 (2018), pp. 1331–1334. DOI: 10.1090/proc/13794.DOI
  12. [12]K. Vogel and H. Risken. “Determination of quasiprobability distributions in terms of probability distributions for the rotated quadrature phase”. In: Physical Review A 40.5 (1989). DOI: 10.1103/PhysRevA.40.2847.DOI

Paper details

vlnc:W_m3r4u4c2
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents