The solution to the Einstein-Podolsky-Rosen paradox

Roman Schnabel1, ∗
1
Institut für Laserphysik & Zentrum für Optische Quantentechnologien,
Universität Hamburg, Luruper Chaussee 149, 22761 Hamburg, Germany
(Dated: August 31, 2022)
In 1935, Albert Einstein, Boris Podolsky, and Nathan Rosen (‘EPR’) reported on a thought ex-
periment that they believed showed that quantum theory provided an incomplete description of
reality. Today it is known that quantum theory is both complete and correct. The EPR thought
experiment, while also correct, had no natural physical explanation until now and has since been
considered a paradox. Here I solve the paradox. I do not refer to any interpretation of quantum
theory but present a simple thought experiment and show that EPR’s seemingly logical statement
on which they base their conclusion is false. I show that with corrected logic, the EPR thought
arXiv:2208.13831v1 [quant-ph] 29 Aug 2022

experiment supports the opposite, namely the non-existence of ‘local realism’. I also show that
EPR experiments are a direct consequence of the existence of true randomness and the conserva-
tion of energy. I see my work as an important building block in finally making quantum physics
understandable.

When the quantum theory was formulated in the three main potential loopholes have since been closed by
1920s, one of its most prominent detractors was Albert experiments [10, 14–16].
Einstein. Together with his two co-authors Boris Podol- Since the quantum theory is proven to be complete
sky and Nathan Rosen, he brought his criticism to the and the EPR thought experiment correct, since it is pre-
point in 1935, when they published a manuscript enti- dicted by quantum theory, EPR’s general logical state-
tled “Can Quantum-Mechanical Description of Physical ment about local realism must be incorrect. But until
Reality Be Considered Complete? ”[1]. It is today one of now, the error in the EPR logic is still a blur. The
Einstein’s most cited works. In this work they formulated EPR thought experiment has been experimentally re-
a general and, as they thought, logical statement about alised many times in many quantum physics laboratories,
the reality of physical properties. They presented a quan- and the measurements are still seen as one of the most
tum physical thought experiment and finally arrived at intriguing aspects of quantum physics. Remarkably, the
the conclusion that physical systems have more precisely associated EPR entanglement is already a valuable re-
defined properties than suggested by the Heisenberg un- source in quantum technology [19–22].
certainty relation. Since the wave function of quantum Here I present a simple thought experiment, which
theory does not describe these properties, EPR concluded demonstrates the logical flaw in EPR’s statement about
their work by stating “While we have thus shown that proving the reality of physical properties and reveals
the wave function does not provide a complete descrip- the correct logic. If and only if the value of a physical
tion of the physical reality, we left open the question of property appears in a truly random fashion, it did not
whether or not such a description exists. We believe, have reality at earlier times, but the value can never the
however, that such a theory is possible.” Niels Bohr and less be precisely inferred (‘predicted’) from the also truly
Erwin Schrödinger immediately published counter state- random value of another physical property. In this case,
ments [2, 3]. While they agreed on the correctness of the two truly random processes are correlated with each
their thought experiment, they expressed their belief that other. I present the physical explanation of the EPR
quantum theory was complete. Apart from this, EPR’s experiments actually performed.
work did not attract further attention for more than 25
True randomness — (Quantum) physics is unable to
years. It changed in the 1960s when John Bell [4] suc-
precisely predict the timing of a spontaneous decay and
ceeded in formulating an inequality that made it possible
the amount of energy that is redistributed when one
to test experimentally whether physical systems could
physical system interacts with another. In other words,
have ‘hidden variables’ that are not covered by quantum
there are quantum uncertainties. They also show up
theory [5, 6]. Since the 1970s and up to recent years,
in measurement results. The monitoring of a single ra-
so-called ‘Bell tests’ have repeatedly drawn a great deal
dioactive atom will detect a decay event that occurs with
of attention [7–17]. They disproved the existence of local
equally distributed probabilities before or after the half-
hidden variables, i.e. ‘local realism’, and thus have com-
life, notably, regardless of when the actual monitoring
pellingly shown that the quantum theory is complete. To
begins.
many physicists this result has been surprising and the
question arose whether Bell tests have loopholes [18]. But One might falsely assume that a causal description of
to promote potential loopholes, strange and implausible when precisely the radioactive decay happens should be
assumptions about Nature have to be made. Anyway, all possible. The experimental violations of Bell inequalities
[7, 8, 10, 14–16], however, provided evidence that this is
 2

randomness according to the quantum theory. If the min-

imal range of true randomness has a Gaussian shape, it
is quantified by the Heisenberg uncertainty principle [24–
27]. Its most cited form considers the position x and the
momentum p of the centre of mass of any physical system
and reads

∆x̂ · ∆p̂ ≥ ~/2 , (1)

where ∆x̂ and ∆p̂ are the standard deviations of the re-
spective uncertainties, and ~ is the reduced Planck con-
stant. It is important to realize that the only role of
the Heisenberg uncertainty principle in the EPR thought
experiment is to quantify a range of guaranteed true ran-
FIG. 1. Predicting the outcome of thrown dice – The
domness according to the quantum theory. For a given
(pseudo)-random number on the bottom of the cube can al-
ways be predicted by observing another random number - ∆p̂, the scatter of position measurement values of at least
namely the one on the top of the same cube. Without looking ~/(2∆p̂) must have a truly random character.
at the bottom faces, we know that the hidden numbers are 6, The EPR thought experiment deals with pairs of sub-
1, and 2, reading from back to front. The illustrated thought systems ‘A’ and ‘B’ that are in a maximally position-
experiment refutes EPR’s seemingly logical statement “A suf- momentum entangled state [3]. The measurements of
ficient condition for the reality of a physical quantity is the x̂A , x̂B , p̂A , and p̂B clearly obey Eq. (1) with subscripts
possibility of predicting it with certainty, without disturbing
either ‘A’ or ‘B’. But whenever a pair is used for two po-
the system.” [1]. My representation is distinct from previous
ones [23] because it considers the dice also before they come sition measurements, the difference of the two outcomes
to rest. Credit: Figure by Alexander Franzen. does not show any uncertainty. And whenever the two
momenta are measured, the sum of the two outcomes
does not show any uncertainty, i.e.
impossible. Quantum uncertainties thus represent ranges
within which events or measurement values happen on a ∆(x̂A − x̂B ) = ∆(p̂A + p̂B ) = 0 , (2)
truly random basis, i.e. without cause.
EPR correctly pointed out that a measurement of x̂A
I suspect that the majority of physicists agrees on this allows to precisely infer x̂B , despite its true randomness.
statement. However, there is a large group of physicists The alternative measurement of p̂A allows to precisely
that goes beyond physics and explores on a metaphys- infer p̂B . According to their logic, the true randomness
ical level whether there are still ways to avoid true in quantum theory was disproven by quantum theory
randomness. This is about ‘interpretations’ of quantum itself and they concluded quantum theory be incomplete.
physics. My manuscript is not about interpretations.
My manuscript shows that true randomness together The flaw in EPR’s logic — In their article, EPR
with suitable physical boundary conditions directly make the following statement. “A sufficient condition
lead to the observation made in the EPR (thought) for the reality of a physical quantity is the possibility
experiment. of predicting it with certainty, without disturbing the
About the EPR thought experiment — By 1935, EPR system.” This statement causes the EPR paradox. It
were convinced that quantum theory itself provided an seems to be a correct statement, but it is not. It’s easy
argument against the existence of true randomness. The to overlook that the statement is unclear about what
EPR thought experiment is a statistical experiment and the words “predicting it” refer to. Since no information
it obeys quantum theory. It consists of many repetitions is provided in the EPR thought experiment about
of exactly the same experiment to show that quantum how the system is generated, “predicting it” can only
theory allows individual measurement results to be in- mean ‘predicting its measurement result’. With this
ferred (‘predicted’) with a precision better than the range clarification, my simple thought experiment in Fig. 1
of the corresponding quantum uncertainty without influ- makes clear that the EPR statement is false.
encing it. According to the false logic of EPR, this pos- Here, ideal dice are thrown the same way many times.
sibility disproved the quantum uncertainty as a range of The statistics of the observed numbers on the top face
true randomness. Being a thought experiment, the in- show the expected flat probability distribution of 1/6
ference was arbitrarily precise, and EPR concluded that for the possible outcomes from one through six. The
there was no true randomness at all in Nature. randomness in this thought experiment is not true
In an actual EPR experiment, the precision of the in- randomness, just pseudo-randomness. In reality, the
ference (‘prediction’) does not need to be perfect. It only individual results are mainly due to specific initial
needs to be more precise than the minimal range of true conditions and differences in the dynamic of the tosses.
 3

Nevertheless, there is practically no way to predict the

measurement results before the cubes come to rest and
before any kind of observation has been taken place.
The pseudo-randomness is therefore sufficient for this
representation.
Visible in Fig. 1 are example results of the ‘first’ ensemble
measurement (‘A’). The ‘second’ ensemble measurement
(‘B’) is pending and is the observation of the number on
the bottom. This number is just as random as the one
above. However, it is clear that we can always predict
the bottom number with certainty based on the top
number. The reason is the strict constraint that opposite
numbers add up to seven. The flaw in EPR’s work is
their assumption that measurement results from truly
random events cannot be predicted (better: ‘inferred’).
The solution to the EPR paradox — Its derivation re-
quires the analysis of how the EPR entangled states are
produced in actual experiments that show the EPR para-
dox. The first experiment with Gaussian uncertainties
was performed with optical fields in 1992 [28]. The type
of EPR entangled states produced were subsequently
used to demonstrate quantum teleportation [29, 30], to
improve interferometric measurements [20, 31], to im-
plement one-sided device independent quantum key dis-
tribution [21], and to research the fundamentals of the
Heisenberg uncertainty principle [32]. In all quoted ex-
FIG. 2. Illustration of the completed EPR thought ex-
periments, the EPR entangled quantities were the depths
periment – The upper half shows the emergence of entangled
of amplitude and phase quadrature modulations of quasi- states from two squeezed states, described by the respective
monochromatic laser light, the so-called quadrature am- Gaussian uncertainties. Two independent input systems a
plitudes X̂f,∆f and Ŷf,∆f [33], where f is the frequency and b in pure squeezed states (ellipses) mutually distribute
of the modulation and ∆f its resolution bandwidth. An their X̂-Ŷ -phase space uncertainties in the course of a bal-
illustration of the quadrature amplitudes for monochro- anced energy exchange. ◦ and + label the positive range of
matic modulations (∆f → 0) can be found in the sup- the input uncertainties. Dashed circles represent ground state
plementary information of Ref. [32]. The quadrature am- uncertainties. The 180◦ -phase space flip is enforced by energy
conservation. The ensemble measurements of X̂A,B and ŶA,B
plitudes are proportional to electric field strengths but
on the output systems A and B reveal Gaussian distributions
normalised to be dimensionless and in such a way that that correspond to the convolution of the input distributions.
the corresponding Heisenberg uncertainty relation reads Since hidden variables do not exist, every single measurement
result must be a truly random number within the Gaussian
∆X̂ · ∆Ŷ ≥ 1 , (3) distribution. (The mean values are set to zero here.) However,
if the same observables are measured on A and B (example
where the subscript f, ∆f is now omitted. In the above measurements 1 to 6), every single measurement value at A
quoted experiments, the quantum correlations in the allows a precise prediction of the measurement outcome at B,
measurement values of the quadrature amplitudes of and vice versa. The remaining uncertainty of the prediction
is given by the squeeze strength of the input uncertainties and
beam A and beam B were of finite strength but with
indicated by the dashed lines in the measurement histograms
canonical symmetry: X̂A ≈ +X̂B and ŶA ≈ −ŶB (or with of B. The figure clarifies why the apparently logical EPR as-
swapped plus and minus signs). With this symmetry, the sumption that a truly random event cannot be predicted is
EPR paradox and thus EPR entanglement is certified if a fallacy. In addition, the figure represents the physics that
the product of the following inequality is violated [34–36] solves the EPR paradox, see main text.

∆(X̂A − X̂B ) · ∆(ŶA + ŶB ) ≥ 1 . (4)

states [33, 38–40]. The latter states have a pair of (mutu-
The stronger the violation, the more significant is the ally uncorrelated) X and Y values that obey the Heisen-
EPR paradox. The experiment reported in [37] reached berg uncertainty principle according to Eq. (3). Most im-
for the first time a value ten times smaller than 1. portantly, measurement values of X and Y of any two
Gaussian EPR entangled states (‘two-mode squeezed single-mode squeezed states do not violate Eq. (4).
states’) are generated by pairs of single-mode squeezed The generation of two laser beams ‘A’ and ‘B’ that
 4

carry EPR entangled states requires two laser beams dice (Fig. 1) that — similar to the EPR thought experi-
‘a’ and ‘b’ of the identical wavelength that carry am- ment — allows for the prediction of a truly random result
plitude squeezed states (∆X̂a < 1) and phase quadra- with certainty. It clearly shows that the prerequisite for
ture squeezed states (∆Ŷb < 1), respectively. (Accord- such a ‘prediction’ is a measured value from a different
ing to Eq. (3), the orthogonal quadratures need to obey but correlated random process. Although the ‘true ran-
∆Ŷa > 1/∆X̂a and ∆X̂b > 1/∆Ŷb .) The two squeezed domness’ in my thought experiment is per definition, it
beams need to be spatially overlapped on a balanced makes obvious that the implication “predictability ⇒ no
beam splitter, see the top part of Fig. 2. The quantum true randomness” is wrong. In the context of a complete
uncertainties of the two beam splitter output beams need theory, only the following implications are correct: “no
to obey the overall symmetry in the setup as well as en- true randomness ⇒ predictability” and “unpredictabil-
ergy conservation. The beam splitter input-output rela- ity ⇔ true randomness”. My thought experiment shows
tion is given by classical wave physics. For two Gaus- that the wrong implication in the EPR publication would
sian input uncertainties without correlations, the out- even be wrong in a ‘semi-classical’ physics, which com-
put uncertainties are also Gaussian and the new stan- bines classical physics with the existence of true random-
dard deviations are weighted sums of those of √ the in- ness.
puts, yielding ∆X̂A = ∆X̂B = √ (∆ X̂a + ∆ X̂b )/ 2 and Referring again to my thought experiment, I note that
∆ŶA = ∆ŶB = (∆Ŷa + ∆Ŷb )/ 2. The quantum uncer- the EPR paradox is not about a prediction concerning a
tainties of the resulting states ‘A’ and ‘B’ are illustrated point in time before any part of the EPR entangled state
in Fig. 2 by the two large blurred circles. Energy con- has interacted with the environment. At the time of the
servation puts a restriction on the input-output phase ‘prediction’ in the EPR thought experiment, the ‘dice’
relations. Energy is conserved if a 180◦ phase rotation have already come to rest. My thought experiment does
applies for one of the reflected quadratures. This choice not motivate local realism at the time before the dice
is representative for all other energy conserving choices. settle. In the same way, the EPR thought experiment
The consequence is the emergence of correlations in the also does not require local realism for times before any
quantum uncertainties of ‘A’ and ‘B’ as pointed out by measurement is taken. The only thing that needs to be
◦ and +. accepted is the existence of true randomness, which is
The four Gaussian probability distributions at the bot- what EPR tried to disprove. I thus show that the EPR
tom of Fig. 2 represent the measurements of the EPR thought experiment does not provide any motivation for
thought experiment including the possibility to predict local realism.
measurement outcomes at B via measurement results at I analyse the process of formation of EPR entangled
A. For perfect input squeezing (∆X̂a , ∆Ŷb → 0) the pre- states in EPR experiments and show that the measure-
diction becomes arbitrarily precise. I note that the entire ment results from them must be (i) correlated but also
spectra of measurement results are due to quantum un- (ii) truly random, based on quantum theory as being
certainties, and thus truly random, although the uncer- complete. Neither the EPR thought experiment, nor
tainty areas of the EPR entangled states are much larger my thought experiment, and not the actually performed
than the lower bound of Eq. (3). EPR experiments [20, 21, 28–32] motivate the assump-
Fig. 2 completes the EPR thought experiment by tion of local realism. Instead, I show that the strict con-
adding the generation of the EPR entangled states. It straint required to make predictions about the outcome
resembles the quantum theory and the EPR thought at ‘B’ based on the measurement at ‘A’ was given in
experiment, and is self-evident. It explains why EPR the experiments by energy conservation in the genera-
entanglement necessarily is generated when two orthog- tion process of the EPR entangled states. This solves
onally squeezed states are superimposed such that the the EPR paradox.
anti-squeezed uncertainty of one squeezed state adds to I would like to draw the following conclusions from
the squeezed uncertainty of the other state. Fig. 2 shows my work. First, EPR’s work was based on a logical
the physics that solves the EPR paradox. flaw. With corrected logic, the EPR thought experiment
even supports the completeness of quantum theory and
Summary and conclusions — The experimental vio- the non-existence of ‘local realism’. While this reduces
lations of Bell-type inequalities already proved that the the motivation to perform Bell tests, the experimental
1935 work of Einstein, Podolsky, and Rosen was flawed. violation of Bell’s inequalities confirmed the corrected
The violations refuted EPR’s conclusion that system logic, the completeness of quantum theory, and the non-
properties should have higher precision than that given existence of ‘local realism’. Second, my work increases
by quantum theory. They thus clarified that the error in the likelihood of finding reasonable physical explanations
the EPR publication was related to its definition of ‘phys- for any remaining strange quantum phenomena.
ical reality’. The EPR thought experiment was correct,
as it was based on quantum theory.
I propose a simple thought experiment with idealised
 5

Acknowledgments — RS thanks Mikhail Ko- Schouten, C. Abellán, W. Amaya, V. Pruneri, M. W.

robko and Ralf Riedinger for useful comments on the Mitchell, M. Markham, D. J. Twitchen, D. Elkouss,
manuscript. This work was performed within the Euro- S. Wehner, T. H. Taminiau, and R. Hanson, “Loophole-
pean Research Council (ERC) Project MassQ (Grant free Bell inequality violation using electron spins sepa-
rated by 1.3 kilometres,” Nature, vol. 526, pp. 682–686,
No. 339897) and within Germany’s Excellence Strategy oct 2015.
– EXC 2056 ‘Advanced Imaging of Matter’, project ID [16] L. K. Shalm, E. Meyer-Scott, B. G. Christensen, P. Bier-
390715994 and EXC 2121 ‘Quantum Universe’, project horst, M. A. Wayne, M. J. Stevens, T. Gerrits, S. Glancy,
ID 390833306, which are financed by the Deutsche D. R. Hamel, M. S. Allman, K. J. Coakley, S. D.
Forschungsgemeinschaft. Dyer, C. Hodge, A. E. Lita, V. B. Verma, C. Lam-
brocco, E. Tortorici, A. L. Migdall, Y. Zhang, D. R.
Kumor, W. H. Farr, F. Marsili, M. D. Shaw, J. A.
Stern, C. Abellán, W. Amaya, V. Pruneri, T. Jennewein,
M. W. Mitchell, P. G. Kwiat, J. C. Bienfang, R. P.
Mirin, E. Knill, and S. W. Nam, “Strong Loophole-Free
∗ Test of Local Realism,” Physical Review Letters, vol. 115,
[email protected] p. 250402, dec 2015.
[1] A. Einstein, B. Podolsky, and N. Rosen, “Can Quantum-
[17] P.-A. Moreau, E. Toninelli, T. Gregory, R. S. Aspden,
Mechanical Description of Physical Reality Be Consid-
P. A. Morris, and M. J. Padgett, “Imaging Bell-type non-
ered Complete?,” Physical Review, vol. 47, pp. 777–780, local behavior,” Science Advances, vol. 5, jul 2019.
may 1935. [18] A. Aspect, “Bell’s inequality test: more ideal than ever,”
[2] N. Bohr, “Can Quantum-Mechanical Description of
Nature, vol. 398, no. March, p. 189, 1999.
Physical Reality be Considered Complete?,” Physical Re-
[19] A. M. Marino, R. C. Pooser, V. Boyer, and P. D.
view, vol. 48, pp. 696–702, oct 1935.
Lett, “Tunable delay of Einstein-Podolsky-Rosen entan-
[3] E. Schrödinger, “Discussion of Probability Relations be- glement,” Nature, vol. 457, no. 7231, pp. 859–862, 2009.
tween Separated Systems,” Mathematical Proceedings of [20] S. Steinlechner, J. Bauchrowitz, M. Meinders, H. Müller-
the Cambridge Philosophical Society, vol. 31, pp. 555–
Ebhardt, K. Danzmann, and R. Schnabel, “Quantum-
563, oct 1935.
dense metrology,” Nature Photonics, vol. 7, pp. 626–630,
[4] J. S. Bell, “On the Einstein Podolsky Rosen Paradox,” jun 2013.
Physics, vol. 1, pp. 195–200, 1964. [21] T. Gehring, V. Händchen, J. Duhme, F. Furrer, T. Franz,
[5] J. F. Clauser, M. A. Horne, A. Shimony, and R. A. Holt,
C. Pacher, R. F. Werner, and R. Schnabel, “Implemen-
“Proposed experiment to test local hidden-variable the-
tation of continuous-variable quantum key distribution
ories,” Physical Review Letters, vol. 23, p. 880, 1969.
with composable and one-sided-device-independent secu-
[6] J. F. Clauser and A. Shimony, “Bell’s theorem: experi- rity against coherent attacks,” Nature Communications,
mental tests and implications,” quant-ph, vol. 41, p. 1881, vol. 6, p. 8795, oct 2015.
1978.
[22] R. Schnabel, “ ‘Quantum Weirdness’ in Exploitation by
[7] S. J. Freedman and J. F. Clauser, “Experimental Test of
the International Gravitational-Wave Observatory Net-
Local Hidden-Variable Theories,” Physical Review Let- work,” Annalen der Physik, vol. 532, p. 1900508, mar
ters, vol. 28, pp. 938–941, apr 1972. 2020.
[8] A. Aspect, P. Grangier, and G. Roger, “Experimental
[23] J. S. Bell, “Bertlmann’s socks and the nature of reality,”
Tests of Realistic Local Theories via Bell‘s Theorem,”
Le Journal de Physique Colloques, vol. 42, pp. C2–41–
Physical Review Letters, vol. 47, p. 460, 1981.
C2–62, mar 1981.
[9] J. Maddox, “Non-locality bursts into life,” Nature, [24] W. Heisenberg, “Über den anschaulichen Inhalt
vol. 352, pp. 277–279, 1991. der quantentheoretischen Kinematik und Mechanik,”
[10] G. Weihs, T. Jennewein, C. Simon, H. Weinfurter, and
Zeitschrift für Physik, vol. 43, pp. 172–198, mar 1927.
A. Zeilinger, “Violation of Bell’s Inequality under Strict
[25] E. H. Kennard, “Zur Quantenmechanik einfacher Bewe-
Einstein Locality Conditions,” Physical Review Letters, gungstypen,” Zeitschrift für Physik, vol. 44, pp. 326–352,
vol. 81, pp. 5039–5043, dec 1998. apr 1927.
[11] A. Aspect, “To be or not to be local,” Nature, vol. 446,
[26] H. Weyl, “Quantenmechanik und Gruppentheorie,”
p. 866, apr 2007.
Zeitschrift für Physik, vol. 46, pp. 1–46, nov 1927.
[12] M. Ansmann, H. Wang, R. C. Bialczak, M. Hofheinz,
[27] H. P. Robertson, “The Uncertainty Principle,” Physical
E. Lucero, M. Neeley, A. D. O’Connell, D. Sank, M. Wei- Review, vol. 34, pp. 163–164, jul 1929.
des, J. Wenner, A. N. Cleland, and J. M. Martinis, “Vi- [28] Z. Y. Ou, S. F. Pereira, H. J. Kimble, and K. C. Peng,
olation of Bell’s inequality in Josephson phase qubits,”
“Realization of the Einstein-Podolsky-Rosen paradox for
Nature, vol. 461, no. 7263, pp. 504–506, 2009.
continuous variables,” Physical Review Letters, vol. 68,
[13] N. Gisin, “Quantum nonlocality: How Does Nature Do pp. 3663–3666, jun 1992.
It?,” Science, vol. 326, pp. 1357–1359, 2009. [29] A. Furusawa, J. L. Sørensen, S. L. Braunstein, C. A.
[14] M. Giustina, A. Mech, S. Ramelow, B. Wittmann,
Fuchs, H. J. Kimble, and E. S. Polzik, “Unconditional
J. Kofler, J. Beyer, A. Lita, B. Calkins, T. Gerrits, S. W.
quantum teleportation,” Science, vol. 282, pp. 706–9, oct
Nam, R. Ursin, and A. Zeilinger, “Bell violation using en-
1998.
tangled photons without the fair-sampling assumption.,” [30] W. P. Bowen, N. Treps, B. C. Buchler, R. Schnabel, T. C.
Nature, vol. 497, pp. 227–30, may 2013. Ralph, H.-A. Bachor, T. Symul, and P. K. Lam, “Ex-
[15] B. Hensen, H. Bernien, A. E. Dréau, A. Reiserer, N. Kalb,
perimental investigation of continuous-variable quantum
M. S. Blok, J. Ruitenberg, R. F. L. Vermeulen, R. N.
teleportation,” Physical Review A, vol. 67, p. 032302,
 6

mar 2003. [36] W. P. Bowen, R. Schnabel, and P. K. Lam, “Experi-

[31] M. Ast, S. Steinlechner, and R. Schnabel, “Reduc- mental Investigation of Criteria for Continuous Variable
tion of Classical Measurement Noise via Quantum-Dense Entanglement,” Physical Review Letters, vol. 90, no. 4,
Metrology,” Physical Review Letters, vol. 117, p. 180801, p. 43601, 2003.
oct 2016. [37] T. Eberle, V. Händchen, and R. Schnabel, “Stable Con-
[32] J. Zander and R. Schnabel, “Full monitoring of ensemble trol of 10 dB Two-Mode Squeezed Vacuum States of
trajectories with 10 dB-sub-Heisenberg imprecision,” npj Light,” Optics Express, 2013.
Quantum Information, vol. 7, p. 148, dec 2021. [38] D. Stoler, “Equivalence Classes of Minimum Uncertainty
[33] R. Schnabel, “Squeezed states of light and their applica- Packets,” Physical Review D, vol. 1, pp. 3217–3219, jun
tions in laser interferometers,” Physics Reports, vol. 684, 1970.
pp. 1–51, apr 2017. [39] D. F. Walls, “Squeezed states of light,” Nature, vol. 306,
[34] L. Duan, G. Giedke, J. Cirac, and P. Zoller, “Insepara- pp. 141–146, nov 1983.
bility criterion for continuous variable systems,” Physical [40] G. Breitenbach, S. Schiller, and J. Mlynek, “Measure-
review letters, vol. 84, pp. 2722–5, mar 2000. ment of the quantum states of squeezed light,” Nature,
[35] R. Simon, “Peres-Horodecki Separability Criterion for vol. 387, pp. 471–475, may 1997.
Continuous Variable Systems,” Physical Review Letters,
vol. 84, p. 2726, 2000.