Optimal Detection of Rotations about Unknown
Axes by Coherent and Anticoherent States
John Martin
1
, Stefan Weigert
2
, and Olivier Giraud
3
1
Institut de Physique Nucléaire, Atomique et de Spectroscopie, CESAM, University of Liège, B-4000 Liège, Belgium
2
Department of Mathematics, University of York, UK-York YO10 5DD, United Kingdom
3
Université Paris-Saclay, CNRS, LPTMS, 91405 Orsay, France
Coherent and anticoherent states of spin sys-
tems up to spin j = 2 are known to be opti-
mal in order to detect rotations by a known
angle but unknown rotation axis. These opti-
mal quantum rotosensors are characterized by
minimal fidelity, given by the overlap of a state
before and after a rotation, averaged over all
directions in space. We calculate a closed-form
expression for the average fidelity in terms of
anticoherent measures, valid for arbitrary val-
ues of the quantum number j. We identify
optimal rotosensors (i) for arbitrary rotation
angles in the case of spin quantum numbers up
to j = 7/2 and (ii) for small rotation angles in
the case of spin quantum numbers up to j = 5.
The closed-form expression we derive allows
us to explain the central role of anticoherence
measures in the problem of optimal detection
of rotation angles for arbitrary values of j.
1 Introduction and main result
Historically, advances in measurement techniques of-
ten are the reason for physics to progress. Over time,
metrology has developed as a subject of its own, es-
pecially in the context of defining standard units of
measurement for physical quantities.
Quantum theory provides new perspectives on mea-
surements, ranging from fundamental limitations on
measurements [1], new opportunities [2] as well as
technical challenges and even philosophical quagmires
[3]. From a practical point of view, quantum infor-
mation science requires ever better control of micro-
scopic systems and, hence, measurements which are
as accurate as possible. More specifically, quantum
metrology [4] aims at finding bounds on the achiev-
able measurement precision and at identifying states
which would be optimal for quantum measurements
or other specific tasks. The optimal transmission of a
Cartesian frame [5] or the efficient detection of inho-
mogeneous magnetic fields [6] are typical examples.
John Martin: jmartin@uliege.b e
Stefan Weigert: stefan.weigert@york.ac.uk
Olivier Giraud: olivier.giraud@universite-paris-saclay.fr
While the classical Cramér-Rao theorem [7, 8] pro-
vides a lower bound on the variance of random estima-
tors by means of the Fisher information, its quantum-
mechanical counterpart provides bounds for quan-
tum parameter estimation theory [9]. The quantum
Cramér-Rao bound is expressed as the inverse of the
quantum Fisher information, which can be geometri-
cally interpreted as the (Bures) distance between two
quantum states differing by an infinitesimal amount
in their parameter [10, 11]. It provides lower bounds
on the variance of any quantum operator whose mea-
surement aims at estimating the parameter. Optimal
measurement is achieved by maximizing the quantum
Fisher information over parameter-dependent states.
The quantum Cramér-Rao bound was calculated
for instance in the reference frame alignment prob-
lem [12]. This problem involves estimating rotations
about unknown axes. It has been shown in [13]
that spin states with vanishing spin expectation value
and isotropic variances of the spin components are
valuable for estimating such rotations, as they satu-
rate the quantum Cramér-Rao bound for any axis.
Also, recently, the problem of characterizing a rota-
tion about an unknown direction encoded into a spin-j
state has been considered in [14].
In this paper, we are interested to determine
whether a quantum system has undergone a rotation
R
n
(η) by a known angle η about an unknown axis
n. Suppose first that we apply the rotation by η to
an initial state |ψi about a known axis and perform
a measurement of the projector |ψihψ| in the rotated
state R
n
(η)|ψi. The expectation value of the observ-
able |ψihψ| is given by
F
|ψi
(η, n) = |hψ|R
n
(η)|ψi|
2
, (1)
i.e. by the fidelity between the initial state and the
final state. The fidelity F
|ψi
(η, n) equals the proba-
bility to find the quantum system in the initial state
after the rotation. Thus, the probability to detect
that the rotation has occurred is given by the quan-
tity 1 − F
|ψi
(η, n). Therefore, the measurement will
be most sensitive if the rotation is applied to states
|ψi which minimize the expression (1) for given angle
and rotation axis.
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 1
arXiv:1909.08355v2 [quant-ph] 18 Jun 2020
Next, suppose that only the rotation angle η is well-
defined while the rotation axis is not known, as de-
scribed in [15]. This situation occurs, for example,
when spins prepared in the state |ψi are—during the
measurement sequence—subjected to a magnetic field
whose direction randomly fluctuates on a time scale
much larger than the Larmor period. Measuring the
observable |ψihψ| on an ensemble of identically pre-
pared systems will now produce a value of the fidelity
(1) averaged over all possible spatial directions n.
Then, the most suitable quantum states |ψi—called
optimal quantum rotosensors in [15]—are determined
by the requirement that the average fidelity
F
|ψi
(η) =
1
4π
Z
S
2
F
|ψi
(η, n) dn , (2)
achieve its minimum, for a given value of the param-
eter η.
The fidelity (1) and its average (2) also play a role
when setting up experiments which aim to determine
an unknown rotation angle as accurately as possible.
This is explained in more detail in Appendix A.
For the spin values j = 1/2, 1, 3/2, 2, optimal quan-
tum rotosensors have been identified [15], using an
approach which combines analytical and numerical
methods. For rotation angles η close to π, the av-
erage fidelity is minimized systematically by coherent
spin states. Coherent spin states are strongly local-
ized in phase space and entirely specified by a spatial
direction into which they point on the Bloch sphere
[16]. For small rotation angles η, the average fidelity
is minimized by anticoherent states, which are char-
acterized by the fact that they do not manifest any
privileged direction; in this respect, they are as dis-
tinct as possible from coherent states [17]. The role
of anticoherent states for optimal detection of rota-
tions has also been observed and was subsequently
quantified in terms of quantum Fisher information
in [13]. Between these two extreme cases of η ∼ 0 and
η ∼ π, optimal states are neither coherent nor anti-
coherent in general. From an experimental point of
view, anticoherent and other non-classical spin states
have been created using a variety of physical sys-
tems. For instance, anticoherent states of quantum
light fields have been generated using orbital angular
momentum states of single photons with their useful-
ness for quantum metrology being established in [18].
Non-classical spin states—including Schrödinger cat
states (c.f. Sec. 4)—of highly magnetic dysprosium
atoms with spin quantum number j = 8 have been
created in order to enhance the precision of a magne-
tometer [19].
The main result of the present paper is a closed-
form expression of the average fidelity F
|ψi
(η), valid
for arbitrary values of j. A rather general argument,
based solely on the symmetries of the average fidelity
F
|ψi
(η), shows that it must be a linear combination
of the form
F
|ψi
(η) = ϕ
(j)
0
(η) +
bjc
X
t=1
ϕ
(j)
t
(η) A
t
(|ψi), (3)
as explained in detail in Sec. 2. In this expression, the
A
t
(|ψi) are the anticoherence measures of a state |ψi,
introduced in [20] and given explicitly in Eq. (10),
while the real-valued functions ϕ
(j)
t
(η) are trigono-
metric polynomials independent of |ψi, and bjc is the
largest integer smaller than or equal to j. The main
challenge is to calculate the η-dependent coefficients
ϕ
(j)
t
(η), which we do in Sec. 3.
In earlier works, the average fidelity F
|ψi
(η) had
been expressed as a sum of functions of η weighted
by state-dependent coefficients, upon representing the
state in the polarization-tensor basis [15]. The advan-
tage of relation (3) is that the average fidelity depends
on the state under consideration only through its mea-
sures of anticoherence, and thus it directly relates to
the degree of coherence or anticoherence of the state.
Expression (3) allows us to identify optimal quantum
rotosensors for spin quantum numbers up to j = 5,
thereby confirming the role played by coherent and
anticoherent states beyond j = 2. Readers mainly
interested in the optimal quantum rotosensors may
want to directly consult Sec. 4.
Let us outline the overall argument leading to the
expression of the average fidelity F
|ψi
(η) in (3). In
Sec. 2, we introduce a number of tools and concepts
feeding into the derivation of (3): first, we discuss
the symmetries built into the average fidelity F
|ψi
(η),
followed by a brief summary of the Majorana repre-
sentation which enables us to interpret spin-j states as
completely symmetric states of N = 2j qubits. This
perspective allows us to introduce, for 1 6 t 6 bjc,
the anticoherence measure A
t
(|ψi), defined as the
linear entropy of the t-qubit reduced density ma-
trix of |ψihψ|. To actually carry out the integration
in Eq. (2), we will use a tensor representation (see
Sec. 2.5) of mixed spin-j states generalizing the Bloch
representation. In addition, this representation also
enables us to exploit the symmetries of the average fi-
delity which can only depend on expressions invariant
under SU(2) rotations. As shown in Sec. 2.6, it is then
possible to establish a linear relation between these
invariants and the anticoherence measures A
t
(|ψi),
which finally leads to (3).
Section 3 is dedicated to deriving explicit expres-
sions for the functions ϕ
(j)
t
(η). This will be done in
two ways: the first one is based on the fact that an-
ticoherence measures are explicitly known for certain
states, so that the functions ϕ
(j)
t
(η) appear as solu-
tions of a linear system of equations. The second ap-
proach makes use of representations of the Lorentz
group and allows us to obtain a general closed ex-
pression. In Sec. 4 we make use of this closed-form
expression to identify the optimal quantum rotosen-
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 2
sors. We conclude with a brief summary given in Sec.
5.
2 Concepts and tools
In this section, we introduce the tools that will be
needed to address the optimality problem described
in the Introduction.
2.1 Notation
Quantum systems with integer or half-integer spin j
are described by states |ψi of the Hilbert space C
N+1
with N = 2j, which carries a (N + 1)-dimensional
representation of the group SU(2). The components
of the angular momentum operator J satisfy [J
k
, J
`
] =
iε
k`m
J
m
, k, `, m ∈ {x, y, z}, where ε
k`m
is the Levi-
Civita symbol. Denoting unit vectors in R
3
by
n =
sin θ cos φ
sin θ sin φ
cos θ
, θ ∈ [0, π] , φ ∈ [0, 2π[ , (4)
the operator
R
n
(η) = e
−iηJ· n
(5)
describes a rotation by an angle η ∈ [0, 4π[ about the
direction n.
2.2 Symmetries
By definition, the average fidelity in (2) is a positive
function of the angle η and of the state |ψi and pos-
sesses three symmetries: it is 2π-periodic in η, sym-
metric about η = π, and invariant under rotation of
|ψi.
Periodicity with period 2π comes from the fact that
R
n
(2π) = (−1)
N
. Symmetry about η = π is equiva-
lent to
F
|ψi
(η) = F
|ψi
(2π − η) , (6)
which can be shown using R
n
(2π−η) = (−1)
N
R
−n
(η)
and the fact that the set of directions averaged over in
(2) is the same irrespective of the sign of the unit vec-
tor n since the fidelity (1) is given by the the squared
modulus of the overlap between the states |ψi and
R
n
(η)|ψi.
Invariance under rotation of |ψi can be understood
in the following way. Let R
m
(χ) = e
−iχJ·m
be a
unitary operator representing a rotation in R
3
by an
angle χ ∈ [0, 4π[ about the direction m, acting on
a state |ψi ∈ C
N+1
. Then the average fidelities F
associated with the states |ψi and |ψ
R
i ≡ R
m
(χ)|ψi
are equal. Indeed, we have
F
|ψ
R
i
(η, n) = hψ|R
m
(χ)
†
R
n
(η)R
m
(χ)|ψi (7)
and
R
m
(χ)
†
R
n
(η)R
m
(χ) = e
−iη(R
m
(χ)
†
JR
m
(χ))·n
= e
−iη(RJ)·n
= e
−iηJ·n
R
, (8)
with n
R
≡ R
T
n the vector obtained by the rotation
R ∈ SO(3) associated with R
m
(χ). Due to the invari-
ance under rotations of the unit-ball region S
2
ap-
pearing in (2) (invariance of the Haar measure used),
the result of the integration will be the same, leading
to
F
|ψ
R
i
(η) =
1
4π
Z
S
2
F
|ψ
R
i
(η, n) dn
=
1
4π
Z
S
2
F
|ψi
(η, n) dn = F
|ψi
(η) . (9)
This invariance of the fidelity can be seen in a geo-
metrically appealing way by use of the Majorana rep-
resentation, which we consider now.
2.3 Majorana representation of pure spin
states
The Majorana representation establishes a one-to-one
correspondence between spin-j states and N = 2j-
qubit states that are invariant under permutation of
their constituent qubits (see e.g. [21, 22, 23]). It al-
lows to geometrically visualise a pure spin-j state as
N points on the unit sphere associated with the Bloch
vectors of the N qubits. The Majorana points are of-
ten referred to as stars, and the whole set of Majo-
rana points of a given state as its Majorana constella-
tion. Considering a spin-j state |ψi as an N-qubit
state, any local unitary (LU) operation U = u
⊗N
with u ∈ SU(2) transforms |ψi into a state whose
Majorana constellation is obtained by the constella-
tion of |ψi rotated by the SO(3) rotation associated
with u. Spin-coherent states take a very simple form
in the Majorana representation, as they can be seen
as the tensor product |φi
⊗N
of some spin-1/2 state
|φi. Their constellation thus reduces to an N-fold de-
generate point.
The fidelity (1) is given by the squared modulus
of the overlap between |ψi and R
n
(η)|ψi. Since the
Majorana constellation of R
n
(η)|ψi is obtained by
rigidly rotating that of |ψi, the fidelity (1) only de-
pends on the relative positions of these two sets of
points. The average transition probability F
|ψi
(η) is
obtained by integrating over all possible constellations
obtained by rigid rotations of the Majorana constel-
lation of |ψi, and therefore it must be invariant under
LU. In other words, the equality (9) takes the form
F
|ψi
(η) = F
u
⊗N
|ψi
(η).
2.4 Anticoherence measures
An order-t anticoherent state |χi is defined by the
property that hχ|(J · n)
k
|χi is independent of the vec-
tor n for all k = 1, . . . , t. In the Majorana represen-
tation, it is characterized by the fact that its t-qubit
reduced density matrix is the maximally mixed state
in the symmetric sector [24].
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 3
The degree of coherence or t-anticoherence of a
spin-j pure state |ψi can be measured by the quan-
tities A
t
(|ψi), which are positive-valued functions of
|ψi [20]. Let ρ
t
= tr
¬t
[|ψihψ|] be the t-qubit reduced
density matrix of the state |ψi interpreted as a N-
qubit symmetric state with N = 2j; it is obtained by
taking the partial trace over all but t qubits (it does
not matter which qubits are traced over since |ψi is a
symmetric state). The measures A
t
(|ψi) are defined
as the rescaled linear entropies
A
t
(|ψi) =
t + 1
t
1 − tr
ρ
2
t
, (10)
where tr
ρ
2
t
is the purity of ρ
t
. Thus, anticoher-
ence measures are quartic in the state |ψi and range
from 0 to 1, and are invariant under SU(2) rota-
tions. Spin-coherent states are characterized by pure
reduced states and thus are the only states such that
A
t
= 0. Anticoherent states to order t are character-
ized by ρ
t
= 1/(t+1) and thus are the only states such
that A
t
= 1. In particular, if a state |ψi is anticoher-
ent to some order t, then it is necessarily anticoherent
to all lower orders t
0
= 1, . . . , t since reductions of the
maximally mixed state are maximally mixed.
While for any state we have 0 6 A
t
6 1, not all
possible tuples (A
1
, A
2
, . . .) are realised by a physi-
cal state |ψi. For instance, since A
t
= 1 implies that
A
t
0
= 1 for all t
0
6 t, the choice A
2
= 1 and A
1
< 1
cannot correspond to any state. We denote the do-
main of admissible values of the measures A
t
by Ω.
2.5 Tensor representation of mixed states
We now introduce a tensor representation of an ar-
bitrary (possibly mixed) spin-j state ρ acting on a
(N + 1)-dimensional Hilbert space with N = 2j, fol-
lowing [24]. Any state can be expanded as
ρ =
1
2
N
x
µ
1
µ
2
...µ
N
S
µ
1
µ
2
...µ
N
. (11)
Here and in what follows, we use Einstein summation
convention for repeated indices, with Greek indices
running from 0 to 3 and Latin indices running from 1
to 3. Here, the S
µ
1
µ
2
...µ
N
are (N + 1) × (N + 1) Her-
mitian matrices invariant under permutation of the
indices.
The x
µ
1
µ
2
...µ
N
are real coefficients also invariant un-
der permutation of their indices, which enjoy what we
call the tracelessness property
3
X
a=1
x
aaµ
3
...µ
N
= x
00µ
3
...µ
N
, ∀ µ
3
, . . . , µ
N
. (12)
Whenever x
µ
1
µ
2
...µ
N
has some indices equal to 0, we
take the liberty to omit them, so that e.g. for a spin-
3 state x
110200
may be written x
112
(recall that the
order of the indices does not matter). In the case of
a spin-coherent state given by its unit Bloch vector
n = (n
1
, n
2
, n
3
), the coefficients in (11) are simply
given by x
µ
1
µ
2
...µ
N
= n
µ
1
n
µ
2
. . . n
µ
N
, with n
0
= 1.
In the following, we will make use of two essential
properties of the tensor representation. Namely, let
us consider a state ρ with coordinates x
µ
1
µ
2
...µ
N
in
the expansion (11). Then, the tensor coordinates of
the t-qubit reduced state ρ
t
in the expansion (11) are
simply given by x
µ
1
µ
2
...µ
t
= x
µ
1
µ
2
...µ
t
0...0
. Thus, since
we omit the zeros in the string µ
1
µ
2
. . . µ
N
, the tensor
coordinates of ρ
t
and ρ coincide for any string of k 6 t
nonzero indices.
The second property we use is that for states ρ and
ρ
0
in the form (11) with tensor coordinates respec-
tively x
µ
1
µ
2
...µ
N
and x
0
µ
1
µ
2
...µ
N
we have
tr [ρρ
0
] =
1
2
N
X
µ
1
,µ
2
,...,µ
N
x
µ
1
µ
2
...µ
N
x
0
µ
1
µ
2
...µ
N
. (13)
Note that this equality holds despite the fact that
the S
µ
1
µ
2
...µ
N
are not orthogonal; this property fol-
lows from the fact that these matrices form a 2
N
-
tight frame, see [24]. In particular, for a pure state
ρ = |ψihψ|, the equality trρ
2
= 1 translates into
X
µ
1
,µ
2
,...,µ
N
x
2
µ
1
µ
2
...µ
N
= 2
N
, (14)
while the purity of the reduced density matrix ρ
t
reads
tr
ρ
2
t
=
1
2
t
X
µ
1
,µ
2
,...,µ
t
x
2
µ
1
µ
2
...µ
t
. (15)
The normalization condition tr [ρ] = 1 imposes
x
00...0
= 1. A consequence of (12) is then that
P
3
a=1
x
aa
= 1.
2.6 SU(2)-Invariants
If u ∈ SU(2) and R ∈ SO(3) is the corresponding
rotation matrix, then the tensor coordinates of UρU
†
with U = u
⊗N
are the R
µ
1
ν
1
. . . R
µ
N
ν
N
x
ν
1
...ν
N
where
R is the 4 × 4 orthogonal matrix
R =
1 0
0 R
!
. (16)
That is, x
µ
1
µ
2
...µ
N
transforms as a tensor. Under
such transformations, x
µ
x
µ
goes into R
µν
R
µν
0
x
ν
x
ν
0
=
(R
T
R)
ν
0
ν
x
ν
x
ν
0
= x
ν
x
ν
, where the last equality comes
from orthogonality of R. Thus x
µ
x
µ
is an SU(2) in-
variant. Similarly, x
µ
x
µν
x
ν
and, more generally, any
product of the x
µ
1
µ
2
...µ
N
such that all indices are con-
tracted (i.e. summed from 0 to 3), are invariant under
SU(2) action on ρ. One can then show by induction
that products of terms x
a
1
a
2
...a
k
with k 6 N where all
indices appear in pairs and are summed from 1 to 3
are also SU(2) invariant. For instance, x
a
x
a
, x
ab
x
ab
,
x
ab
x
bc
x
ca
, x
a
x
ab
x
b
are such invariants.
Invariants of degree 1 in x are of the form x
a
1
a
2
...a
2k
,
where the a
i
appear in pairs. Since the order of in-
dices is not relevant, these invariants are in fact of
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 4
the form x
a
1
a
1
a
2
a
2
...a
k
a
k
. Because of Eq. (12), each
pair can be replaced by zeros in the string, so that
x
a
1
a
1
a
2
a
2
...a
k
a
k
= x
00...0
= 1. Therefore, there is no
invariant of degree 1. The invariants of degree 2 are
products of the form x
a
1
a
2
...a
k
x
b
1
b
2
...b
k
0
where indices
appear in pairs and are summed from 1 to 3. If the
two indices of a pair appear in the same index string
(a
1
a
2
. . . a
k
or b
1
b
2
. . . b
k
0
), then from Eq. (12), they
can again be replaced by zeros and discarded. Thus
the invariants of degree 2 are κ
1
= x
a
x
a
, κ
2
= x
ab
x
ab
,
and more generally, for 1 6 r 6 N,
κ
r
= x
a
1
a
2
...a
r
x
a
1
a
2
...a
r
. (17)
Using (10) and (15) one can express the invariants
κ
r
in terms of a linear combination of the A
t
. In-
deed, grouping together terms with the same number
of nonzero indices in (15) yields
tr
ρ
2
t
=
1
2
t
X
µ
1
,µ
2
,...,µ
t
x
2
µ
1
µ
2
...µ
t
=
1
2
t
t
X
r=0
t
r
κ
r
.
(18)
Inverting that relation via the binomial inversion for-
mula, we obtain
κ
r
=
r
X
t=0
(−1)
t+r
2
t
r
t
tr
ρ
2
t
, (19)
and by use of (10) we finally can express the SU(2)-
invariants in terms of anticoherence measures,
κ
r
=
r
X
t=0
(−1)
t+r
2
t
r
t
1 −
t
t + 1
A
t
(20)
for r = 1, . . . , N.
2.7 General form of the average fidelity
Let us now explain why the average fidelity F
|ψi
(η)
given in Eq. (3) is a linear combination of the low-
est bjc anticoherent measures A
t
. Due to its rota-
tional symmetry, the average fidelity F
|ψi
(η)—when
considered as a function of the tensor coordinates
x
µ
1
µ
2
...µ
N
—can only involve invariants constructed
from these coordinates. With F
|ψi
(η) being quadratic
in ρ = |ψihψ|, it must also be quadratic in x. As there
is no invariant of degree 1, the only invariants that
can appear in the expression of F
|ψi
(η) are the invari-
ants κ
r
defined in (17). Since the quantity F
|ψi
(η)
is quadratic it must be a linear combination of the
coefficients κ
r
which, according to Eq. (20), implies
that F
|ψi
(η) is also a linear combination of the A
t
.
Furthermore, the identity
tr
ρ
2
t
= tr
ρ
2
N−t
, (21)
which holds for any pure state, means that the antico-
herence measures A
t
for t > N/2 can be expressed in
terms of the measures A
t
for t < N/2. Therefore, (3)
is the most general form the fidelity F
|ψi
(η) can take,
with the dependence in η being only in the coefficients
of the measures A
t
.
2.8 Generalizations
It is worth stressing that the form (3) for the average
fidelity also holds for more general types of average
fidelity
1
4π
Z
S
2
|hψ|U
n
(η)|ψi|
2
dn (22)
between a state |ψi and its image under the unitary
U
n
(η) = e
−iη f(J·n)
, (23)
where f is an arbitrary real analytic function, ensur-
ing that f(J · n) is an Hermitian operator. Indeed,
from an argument similar to that of Sec. 2.6, the gen-
eralized fidelity (22) can be expressed as a function
of the κ
r
and hence of the A
t
. An interesting case is
when U
n
(η) is a spin-squeezing operator, which cor-
responds to choosing f(J · n) = (J · n)
2
. Moreover, if
we now consider the quantities
1
4π
Z
S
2
|hψ|U
n
(η)|ψi|
2k
dn (24)
with integer k > 2, the same arguments show that
they are linear combinations of higher-order invari-
ants, leading to generalizations of the relation (20).
3 Closed form of the average fidelity
In this section we derive the angular functions ϕ
(j)
t
(η),
which characterize the fidelity through (3), in two
different ways. The first method (subsection 3.1) is
based on the fact that anticoherence measures can
be evaluated explicitly for Dicke states. The second
method (subsection 3.2) exploits a tensor represen-
tation of spin states [24] which uses Feynman rules
from relativistic spin theory. These approaches are
independent and we checked, for all integers and half-
integers j up to 26, that as expected they yield the
same angular functions. Technical detail is delegated
to appendices in both cases.
3.1 Derivation based on anticoherence mea-
sures for Dicke states
In the following, we will work in the standard angu-
lar momentum basis of C
N+1
, for positive integer or
half-integer value of j = N/2. It consists of the Dicke
states {|j, mi, |m| 6 j} given by the common eigen-
states of J
2
, the square of the angular momentum
operator J, and of its z-component J
z
. In this basis,
any spin-j state |ψi can be expanded as
|ψi =
j
X
m=−j
c
m
|j, mi, (25)
with c
m
∈ C and
P
j
m=−j
|c
m
|
2
= 1.
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 5
The first derivation is based on the fact that both
the measures of t-anticoherence A
t
(|j, mi) and the av-
erage fidelities F
|j,mi
(η) can be determined explicitly
for Dicke states. Their measures of t-anticoherence
are given by
A
t
(|j, mi) =
t + 1
t
1 −
P
t
`=0
j+m
t−`
2
j−m
j−m−`
2
2j
t
2
.
(26)
They can readily be obtained from the purities tr
ρ
2
t
for a state of the form (25), which were calculated in
[20] in terms of the coefficients c
m
and read
tr
ρ
2
t
=
t
X
q,`=0
2j−t
X
k=0
c
∗
j−k−`
c
j−k−q
Γ
`q
k
2
(27)
with
Γ
`q
k
=
q
2j−k−q
t−q
2j−k−`
t−`
k+q
k
k+`
k
2j
t
. (28)
As for the fidelity, the calculation is done in Appendix
B and yields
F
|j,mi
(η) =
1
(2j + 1)
2
2j
X
`=0
(2` + 1)(C
jm
jm`0
χ
j
`
(η))
2
,
(29)
with Clebsch-Gordan coefficients C
jm
jm`0
and the func-
tions χ
j
`
(η) defined in Eqs. (77)–(78). The angular
functions ϕ
(j)
t
(η) are then solutions of the system of
linear equations
F
|j,mi
(η) = ϕ
(j)
0
(η) +
P
bjc
t=1
ϕ
(j)
t
(η) A
t
(|j, mi)
for m = j, j − 1, . . . , j − bjc.
(30)
This system can easily be solved for the lowest values
of j. A general (but formal) solution can then be
obtained by inverting the system (30).
3.2 Derivation based on relativistic Feynman
rules and tensor representation of spin states
The second approach allows us to derive a closed-form
expression for the functions ϕ
(j)
t
(η). It is based on an
expansion of the operator
Π
(j)
(q) ≡ (q
2
0
− |q|
2
)
j
e
−2θ
q
ˆ
q·J
, (31)
with tanh θ
q
= −|q|/q
0
and
ˆ
q = q/|q|, as a multi-
variate polynomial in the variables q
0
, q
1
, q
2
, q
3
. This
operator is a (N +1)-dimensional representation (with
N = 2j) of a Lorentz boost in the direction of the 4-
vector q = (q
0
, q) = (q
0
, q
1
, q
2
, q
3
). As shown in [25],
it can be written as
Π
(j)
(q) = (−1)
2j
q
µ
1
q
µ
2
. . . q
µ
2j
S
µ
1
µ
2
...µ
2j
. (32)
The identification of Eqs. (31) and (32) defines the
(N + 1) ×(N + 1) matrices S
µ
1
...µ
N
appearing in (11)
(see [24] for detail). Taking
q
0
= i cot(η/2) and q
i
= n
i
, i = 1, 2, 3 , (33)
in (31), we see that Π
(j)
(q) reduces to a rotation op-
erator,
R
n
(η) = e
−iηJ· n
=
Π
(j)
(q)
m
N
(34)
with
m
2
= q
2
0
− |q|
2
= −
1
sin
2
(η/2)
. (35)
Moreover, for a state ρ given by (11) we have
tr
h
ρ Π
(j)
(q)
i
= (−1)
N
x
µ
1
µ
2
...µ
N
q
µ
1
. . . q
µ
N
, (36)
according to Eq. (24) of [24], which holds for any 4-
vector q. Thus, with ρ = |ψihψ|, using the identity
(34) and the expansion (32) for the rotation operator
in (1) allows us to explicitly perform the integral in
Eq. (2), resulting in
F
|ψi
(η) =
1
4π
Z
S
2
|hψ|R
n
(η)|ψi|
2
dn
=
1
4π
Z
S
2
tr
ρ
Π
(j)
(q)
m
N
2
dn
= (−1)
N
x
µ
1
...µ
N
x
ν
1
...ν
N
4π
×
Z
S
2
q
µ
1
. . . q
µ
N
q
∗
ν
1
. . . q
∗
ν
N
m
2N
dn,
(37)
where ∗ denotes complex conjugation (which acts on
q
0
only because of the choice (33) and using |m|
2
=
−m
2
). Each term q
µ
1
. . . q
∗
ν
N
with 2(N − k) indices
equal to 0 is proportional to
q
2(N−k)
0
m
2N
= (−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
. (38)
For the remaining 2k nonzero indices, we have from
(33) that q
i
= n
i
, so that (37) involves an integral of
the form
1
4π
Z
S
2
n
a
1
n
a
2
. . . n
a
2k
dn , 1 6 a
i
6 3 . (39)
These integrals are performed in Appendix C. The in-
tegrals (39) are in fact precisely given by the tensor
coordinates x
(0)
a
1
a
2
...a
2k
of the maximally mixed state,
whose expression is explicitly known. One can there-
fore rewrite (37) as
F
|ψi
(η) =
N
X
k=0
(−1)
N
q
2(N−k)
0
m
2N
×
X
µ,ν
2(N−k)zeros
(−1)
nr of 0 in ν
x
(0)
µ
1
...µ
N
ν
1
...ν
N
x
µ
1
...µ
N
x
ν
1
...ν
N
,
(40)
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 6
where the sum over µ,ν runs over all strings of in-
dices (between 0 and 3) containing 2(N − k) zeros.
An explicit expression for this sum is derived in Ap-
pendix C, leading to the compact expression
F
|ψi
(η) =
N
X
k=0
sin
2k
η
2
cos
2(N−k)
η
2
N
X
t=0
a
(j)
t,k
tr
ρ
2
t
,
(41)
with numbers
a
(j)
t,k
=
4
t
(−1)
k+t
2N
2k
k
t
2N−2t
N−t
(2k + 1)
2N
N
. (42)
Note that the sum over k in (41) can start at k = t
because the factor
k
t
in a
(j)
t,k
implies that a
(j)
t,k
= 0
for t > k. Using the symmetry tr
ρ
2
t
= tr
ρ
2
N−t
we
may rewrite (41) as
F
|ψi
(η) =
N
X
k=t
sin
2k
η
2
cos
2(N−k)
η
2
×
bjc
X
t=0
a
(j)
t,k
+ a
(j)
N−t,k
1 −
δ
jt
2
tr
ρ
2
t
.
(43)
From (10) we obtain a relation between A
t
and tr
ρ
2
t
,
namely tr
ρ
2
t
= 1 −
t
t+1
A
t
, which yields the explicit
expression of the polynomials ϕ
(j)
t
(η) in Eq. (3) as
ϕ
(j)
t
(η) =
N
X
k=t
b
(j)
t,k
sin
2k
η
2
cos
2(N−k)
η
2
, (44)
with coefficients
b
(j)
t,k
=
−
t
t + 1
a
(j)
t,k
+ a
(j)
N−t,k
1 −
δ
jt
2
t 6= 0
N
k
2k + 1
t = 0 .
(45)
Note that although q
0
and m are not well-defined for
η = 0, the ratio in (38) always is, so that the expres-
sion above is valid over the whole range of values of
η. For spin-coherent states, all A
t
vanish and thus
F
|ψi
(η) = ϕ
(j)
0
(η) from Eq. (3), which coincides with
the expression obtained in [15]. For the smallest val-
ues of j, we recover the functions obtained in Section
3.1. In the following section, we will use the functions
ϕ
(j)
t
(η) given in (44) to identify optimal quantum ro-
tosensors.
4 Optimal quantum rotosensors
4.1 Preliminary remarks
We now address the question of finding the states |ψi
which minimize the average fidelity F
|ψi
(η) for fixed
rotation angles η. According to Eq. (3), the fidelity
is a linear function of the anticoherence measures A
t
with 1 6 t 6 bjc. Linearity, when combined with
the fact that the domain Ω, over which the measures
A
t
vary, is bounded implies that the fidelity must at-
tain its minimum on the boundary. The minimization
problem thus amounts to characterizing this domain
Ω. Unfortunately, even for the smallest values of j,
no simple descriptions of this domain are known.
We will first determine the states minimizing the
2π-periodic average fidelity for values of j up to
j = 7/2, with the rotation angle taking values in the
interval η ∈ [0, π] (which is sufficient due to the sym-
metry (6)). Then we will examine the limiting case
of angles η close to 0 for arbitrary values of the quan-
tum number j. Throughout this section, we will ex-
pand arbitrary states with spin j in terms of the Dicke
states, as shown in Eq. (25).
For spins up to j = 2 the states minimizing the av-
erage fidelity F
|ψi
(η) are known [15]. In Sec. 4.2, we
show that our approach based on the expression (3)
correctly reproduces these results. Then, in Sec. 4.3,
we consider the minimization problem for spin quan-
tum numbers up to j = 7/2, mainly identifying the
optimal rotosensors within various ranges of the ro-
tation angle η by numerical techniques. More specifi-
cally, for a fixed angle η, F
|ψi
(η) is a function of the
A
t
which can be parametrized by the complex coef-
ficients c
m
entering the expansion (25) of the state
|ψi in the Dicke basis (see Eq. (27)). We search nu-
merically for the minimum value of F
|ψi
(η) with re-
spect to the c
m
, taking into account the normalization
condition
P
m
|c
m
|
2
= 1. In most cases this numeri-
cal search converges towards states which have simple
analytic expressions which are the ones that we give.
For each value of j, we performed this search at about
1000 evenly spaced values of η in order to explore the
whole range of rotation angles. Whenever we find a
region of values of η in which |ψ
1
i is the optimal state
adjacent to a region where |ψ
2
i is optimal, at the crit-
ical angle separating these two regions, one should
have F
|ψ
1
i
(η) = F
|ψ
2
i
(η) because the average fidelity
F
|ψi
(η) is a continuous function of |ψi. Therefore, the
critical angle is a solution of the equation
bjc
X
t=1
ϕ
(j)
t
(η) A
t
(|ψ
1
i) =
bjc
X
t=1
ϕ
(j)
t
(η) A
t
(|ψ
2
i). (46)
4.2 Rotosensors for arbitrary rotation angles η
and j 6 2
4.2.1 j = 1/2
For a spin 1/2, all pure states are coherent: each state
|ψi can be obtained by a suitable rotation of the state
|
1
2
,
1
2
i. Since the fidelity is invariant under rotation,
all states are equally sensitive to detect rotations for
any angle η.
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 7
4.2.2 j = 1
For j = 1, the expansion (3) takes the form
F
|ψi
(η) = ϕ
(j)
0
(η) + ϕ
(1)
1
(η) A
1
, (47)
with
ϕ
(1)
0
(η) =
1
15
6 cos(η) + cos(2η) + 8
,
ϕ
(1)
1
(η) = −
1
15
2 cos(η) − 3 cos(2η) + 1
.
(48)
The first strictly positive zero of ϕ
(1)
1
(η) is given by
η
0
= arccos(−2/3). In the interval η ∈ [0, η
0
[, where
ϕ
(1)
1
(η) is negative, the fidelity F
|ψi
(η) is minimized
by states with A
1
= 1, i.e. by 1-anticoherent states.
For η = η
0
, the fidelity takes the same value for
all states |ψi, namely F
|ψi
(η
0
) = ϕ
(1)
0
(η
0
) = 7/27.
For rotation angles in the the remaining interval,
η ∈]η
0
, π], where ϕ
(1)
1
(η) is positive, F
|ψi
(η) is min-
imized for states with A
1
= 0, i.e. coherent states.
Thus, we indeed recover the results obtained in [15].
4.2.3 j = 3/2
In this case, the average fidelity (3) reads
F
|ψi
(η) = ϕ
(3/2)
0
(η) + ϕ
(3/2)
1
(η) A
1
, (49)
with
ϕ
(3/2)
0
(η) =
1
70
cos(3η) + 8 cos(2η) + 29 cos(η) + 32
,
ϕ
(3/2)
1
(η) =
3
70
3 cos(3η) + 3 cos(2η) − 4 cos(η) − 2
.
(50)
The situation is basically the same as for j = 1. The
first strictly positive zero of the coefficient ϕ
(3/2)
1
(η)
is found to be η
0
= arccos(
−9+
√
21
12
). Hence, in the
interval η ∈ [0, η
0
[ where ϕ
(3/2)
1
(η) is negative, the
fidelity F
|ψi
(η) is minimal for 1-anticoherent states.
At the value η = η
0
, the fidelity takes the same value
for all states |ψi, namely, F
|ψi
(η
0
) = ϕ
(3/2)
0
(η
0
) =
(33 + 2
√
21)/80. Otherwise, F
|ψi
(η) is minimized for
coherent states, thereby reproducing earlier results
[15].
4.2.4 j = 2
For j = 2, the fidelity (3) is a linear combination of
three terms,
F
|ψi
(η) = ϕ
(2)
0
(η) + ϕ
(2)
1
(η) A
1
+ ϕ
(2)
2
(η) A
2
, (51)
with the angular functions ϕ
(2)
k
, k = 0, 1, 2, displayed
in Appendix D. They all take negative values in the
interval η ∈ [0, η
0
], with η
0
≈ 1.2122 the first strictly
positive zero of ϕ
(2)
1
(η). The tetrahedron state
|ψ
tet
i =
1
2
|2, −2i + i
√
2 |2, 0i + |2, 2i
, (52)
whose Majorana points lie at the vertices of a regu-
lar tetrahedron, is 2-anticoherent, and for j = 2 it is
the only state (up to LU) with A
1
= A
2
= 1 [26];
hence it provides the optimal rotosensor for angles in
the interval η ∈ [0, η
0
]. For larger angles of rotation
comprised between 1.68374 and 2.44264, we find nu-
merically that an optimal state is the Schrödinger cat
state
|ψ
cat
i =
1
√
2
(|2, −2i + |2, 2i) , (53)
which is only 1-anticoherent, with A
1
= 1 and A
2
=
3/4. For values η & 2.44264, the optimal state is a
coherent state.
We thus obtain numerically three intervals with
three distinct optimal states corresponding to
(A
1
, A
2
) = (1, 1), (1, 3/4), and (0, 0), respectively. In
order to find the critical angles, we solve Eq. (46).
The angle η
1
separating the first two regions is a so-
lution of ϕ
(2)
2
(η) = 0. The first positive zero of ϕ
(2)
2
(η)
is η
1
= 2 arctan(
p
9 − 2
√
15) ≈ 1.68374, which coin-
cides with the numerically obtained value. The angle
η
2
at which the second and third region touch, is a
zero of ϕ
(2)
1
(η) +
3
4
ϕ
(2)
2
(η). Its first strictly positive
zero is given by
η
2
= 2 arctan
r
−
a + 102b
a − 38b
!
, (54)
with a = 19 6
2/3
+
3
√
6
223 − 35
√
7
2/3
and b =
3
p
223 − 35
√
7, and we have indeed η
2
≈ 2.44264. The
results we obtained are summarized in Fig. 1; they
agree with the findings of [15].
It is noteworthy that the state (53) is not the only
state with anticoherence measures A
1
= 1 and A
2
=
3/4. For instance, any state of the form
|ψi =
c
1
|2, −1i + c
2
|2, 0i − c
∗
1
|2, 1i
p
2|c
1
|
2
+ |c
2
|
2
(55)
with c
1
∈ C and c
2
∈ R come with the same measures
of anticoherence, as readily follows from Eq. (27).
These states are thus also optimal in the interval
η ∈ [η
1
, η
2
], thereby removing the uniqueness of opti-
mal rotosensors observed for j = 1 and j = 3/2.
4.3 Rotosensors for 5/2 6 j 6 7/2
4.3.1 j = 5/2
For j = 5/2, there is no anticoherent state of order
2 but only of order 1 [12]. Numerical optimization
shows that the optimal state for small angles of rota-
tion is the 1-anticoherent state with the largest mea-
sure of 2-anticoherence, that is given by
|ψi =
1
√
2
|
5
2
, −
3
2
i + |
5
2
,
3
2
i
, (56)
and has A
1
= 1 and A
2
= 99/100. This state is found
to be optimal up to a critical angle η
1
≈ 1.49697,
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 8
πη
2
η
1
π
4
0
1
0.8
0.6
0.4
0.2
0
t = 2
t = 1
η
A
t
1
0.8
0.6
0.4
0.2
0
F
|ψi
(η)
Figure 1: Average fidelity F
|ψi
(η) (top, red solid curve) and
measures of anticoherence A
t
(bottom) for optimal states
with j = 2, as functions of the rotation angle η; the values
of the measures A
t
for the optimal states are discontinuous
at the values η
1
≈ 1.68374 and η
2
≈ 2.44264 (see text for
details). The dashed curve on top shows the average fidelity
ϕ
(2)
0
(η) for coherent states. The blue (red) shaded area shows
the range of rotation angles for which anticoherent states to
order bjc (coherent states) are optimal.
which is obtained from Eq. (46) and coincides with
the first strictly positive zero of ϕ
(5/2)
2
(η). It is worth
noting that the optimal state (56) was also found to be
the most non-classical spin state for j = 5/2, both in
the sense that it maximizes the quantumness [27] and
that it minimizes the cumulative multipole distribu-
tion [28, 29]. The Majorana constellation of this state
defines a triangular bipyramid, which is a spherical
1-design [30, 31], thus corresponding to the arrange-
ment of point charges on the surface of a sphere which
minimize the Coulomb electrostatic potential energy
(solution to Thomson’s problem for 5 point charges,
see [32]).
For larger angles of rotation ranging between η
1
and
η
2
≈ 2.2521, we find that an optimal state is
|ψ
cat
i =
1
√
2
|
5
2
, −
5
2
i + |
5
2
,
5
2
i
; (57)
unlike in the case j = 2, we found this state for j =
5/2 to be the only state (up to LU) with A
1
= 1 and
A
2
= 3/4. For η ∈ [η
2
, π], we find that coherent states
are optimal. The transition occurs at the first strictly
positive zero η
2
of ϕ
(5/2)
1
(η) +
3
4
ϕ
(5/2)
2
(η). Our results
are summarized in Fig. 2.
πη
2
η
1
π
4
0
1
0.8
0.6
0.4
0.2
0
t = 2
t = 1
η
A
t
1
0.8
0.6
0.4
0.2
0
F
|ψi
(η)
Figure 2: Average fidelity F
|ψi
(η) (top, red solid curve) and
measures of anticoherence A
t
(bottom) for optimal states
with j = 5/2, as functions of the rotation angle η; the values
of the measures A
t
for the optimal states are discontinuous
at the values η
1
≈ 1.49697 and η
2
≈ 2.2521 (see text for
details). The dashed curve on top shows the average fidelity
ϕ
(5/2)
0
(η) for coherent states. Shaded areas are defined as in
Fig. 1.
4.3.2 j = 3
Anticoherent states of order 3 do exist for j = 3. They
are all connected by rotation to the octahedron state
|ψ
oct
i =
1
√
2
(|3, −2i + |3, 2i) , (58)
whose Majorana points lie at the vertices of a regular
octahedron. Therefore, the state (58) is, at small η,
the unique optimal quantum rotosensor (up to LU)
for j = 3. Numerical optimization shows that the
octahedron state is optimal up to a critical angle η
1
≈
1.3635 coinciding with the first strictly positive zero
of
1
4
ϕ
(3)
2
(η) +
1
3
ϕ
(3)
3
(η), and that, for larger angles,
the state
|ψ
cat
i =
1
√
2
(|3, −3i + |3, 3i) (59)
with A
1
= 1, A
2
= 3/4 and A
3
= 2/3 is optimal up to
a critical angle η
2
≈ 2.04367 coinciding with the first
strictly positive zero of ϕ
(3)
1
(η)+
3
4
ϕ
(3)
2
(η)+
2
3
ϕ
(3)
3
(η).
We found that this is the only spin-3 state (up to
LU) with A
1
= 1, A
2
= 3/4 and A
3
= 2/3. Coherent
states are found to be optimal for angles of rotation in
the ranges [η
2
, η
3
] and [η
4
, π] with η
3
≈ 2.35881 and
η
4
≈ 2.65576 coinciding with the second and third
strictly positive zeros of ϕ
(3)
1
(η)+ϕ
(3)
2
(η)+ϕ
(3)
3
(η). In
the range [η
3
, η
4
], the octahedron state (58) becomes
again optimal (although the three functions ϕ
(3)
k
for
k = 1, 2, 3 are not simultaneously negative in that
range). Our results are displayed in Fig. 3.
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 9
πη
4
η
3
η
2
η
1
π
4
0
1
0.8
0.6
0.4
0.2
0
t = 3
t = 2
t = 1
η
A
t
1
0.8
0.6
0.4
0.2
0
F
|ψi
(η)
Figure 3: Average fidelity F
|ψi
(η) (top, red solid curve) and
measures of anticoherence A
t
(bottom) for optimal states
with j = 3, as functions of the rotation angle η; the values
of the measures A
t
for the optimal states are discontinuous
at the values η
1
≈ 1.3635, η
2
≈ 2.04367, η
3
≈ 2.35881 and
η
4
≈ 2.65576 (see text for details). The dashed curve on
top shows the average fidelity ϕ
(3)
0
(η) for coherent states.
Shaded areas are defined as in Fig. 1.
4.3.3 j = 7/2
This is the smallest spin quantum number for which
a smooth variation of the optimal state with η is ob-
served, resulting in the complex behaviour displayed
in Figs. 4 and 5. There are no anticoherent states
to order 3 for j = 7/2, but there exist anticoherent
states to order 2. The optimal state for small angles
of rotation (by which we mean here η → 0) turns out
to be one of those. Numerical optimization yields the
state
|ψi =
q
2
9
|
7
2
, −
7
2
i −
q
7
18
|
7
2
, −
1
2
i −
q
7
18
|
7
2
,
5
2
i (60)
with measures of anticoherence A
1
= A
2
= 1 and
A
3
= 1198/1215. This is not the state with the high-
est measure of 3-anticoherence, as the state
|ψi =
1
√
2
|
7
2
, −
5
2
i + |
7
2
,
5
2
i
, (61)
has measures of anticoherence A
1
= 1, A
2
= 195/196
and A
3
= 146/147 > 1198/1215. The latter state is
found to be optimal for η ∈ [η
1
, η
2
] with η
1
≈ 0.71718
(not identified) and η
2
≈ 1.24169 coinciding with the
first strictly positive zero of
12
49
ϕ
(7/2)
2
(η)+
16
49
ϕ
(7/2)
3
(η).
The state
|ψ
cat
i =
1
√
2
|
7
2
, −
7
2
i + |
7
2
,
7
2
i
(62)
with A
1
= 1, A
2
= 3/4 and A
3
= 2/3 is found
to be optimal for η ∈ [η
2
, η
3
] and η ∈ [η
4
, η
5
] with
η
3
≈ 1.60141 and η
4
≈ 1.88334 coinciding with the
third and fourth strictly positive zeros of ϕ
(7/2)
1
(η)
and η
5
≈ 2.41684 with the first strictly positive zero
of ϕ
(7/2)
1
(η) +
3
4
ϕ
(7/2)
2
(η) +
2
3
ϕ
(7/2)
3
(η). In the interval
[η
5
, π], coherent states are found to be optimal.
πη
5
η
4
η
3
η
2
η
1
0
1
0.8
0.6
0.4
0.2
0
t = 3
t = 2
t = 1
η
A
t
1
0.8
0.6
0.4
0.2
0
F
|ψi
(η)
Figure 4: Average fidelity F
|ψi
(η) (top, red solid curve) and
measures of anticoherence A
t
(bottom) for optimal states
with j = 7/2, as functions of the rotation angle η. The
dashed curve on top shows the average fidelity ϕ
(7/2)
0
(η) for
coherent states. Shaded areas are defined as in Fig. 1.
0.8η
1
0.60.40.20
1
195
196
146
147
0.99
1198
1215
t = 3
t = 2
t = 1
η
A
t
Figure 5: Measures of anticoherence A
t
for optimal states
with j = 7/2, as functions of the rotation angle η ∈ [0, 0.8].
4.4 Rotosensors for small rotation angles η and
arbitrary values of j
4.4.1 Angular functions at small angles
According to Secs. 4.2 and 4.3 optimal rotosensors
for integer values of spin (j = 1, 2, 3) are given by
j-anticoherent states while for half-integer spin (j =
3/2, 5/2, 7/2) the fidelity is optimized by states which
are anticoherent of order t = 1, 1, 2, respectively, and
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 10
possess large anticoherence measures A
t
for values of t
up to t = bjc. This fact can be understood quite gen-
erally through the behaviour of the functions ϕ
(j)
t
(η)
at small η for arbitrary values of j. In the vicinity of
η = 0, the functions ϕ
(j)
t
(η) given in Eq. (44) take the
form
ϕ
(j)
t
(η) =
b
(j)
t,t
2
2t
η
2t
+ O(η
2t+2
), (63)
with coefficients b
(j)
t,t
given by Eq. (45). These co-
efficients are strictly negative for all t > 1 and all
j = N/2, since a
(j)
t,t
> 0 and a
(j)
N−t,t
is either 0 for
t < N/2 or positive for t = N/2. This implies that all
functions ϕ
(j)
t
(η) are negative in some interval around
η = 0. Thus, the fidelity F
|ψi
(η) is a linear combina-
tion of the A
t
with negative coefficients in that inter-
val. Since 0 6 A
t
6 1, it follows that if there exists a
state with A
t
= 1 for all t 6 bjc—that is, an antico-
herent state to order bjc—then this state provides an
optimal quantum rotosensor for η in that interval.
This interval can be made more specific, at least
for the lowest values of j. Let η
0
denote the first
zero of ϕ
(j)
1
(η). Numerical results up to j = 85 in-
dicate that all functions ϕ
(j)
t
(η) for t = 1, . . . , bjc
are negative for η ∈ [0, η
0
], so that an anticoherent
state to order bjc (if it exists) is optimal in the whole
interval [0, η
0
]. As shown in Fig. 6, η
0
is found to
scale as 3π/(4j) for large j. A simple explanation for
this is that the expansion of the function ϕ
(j)
1
(η) as
P
k
a
k
cos(kη) is dominated by the term a
2j
cos(2jη)
(note however that η
0
is even better approximated by
9/(4j)). Conversely, the states maximizing F
|ψi
(η)
for small angles of rotation are the states with A
t
= 0
for all t, i.e. coherent states.
15
2
7
13
2
6
11
2
5
9
2
4
7
2
3
5
2
2
3
2
1
2.5
2
1.5
1
0.5
0
3π
4j
j
η
0
Figure 6: First zero η
0
of the functions ϕ
(j)
1
(η) (blue dots)
as a function of j: for j = 1 and for j > 5/2, the values are
well approximated by η
0
≈ 3π/(4j) (pink dashes).
To see whether any general pattern emerges, we
now identify optimal small-angle rotosensors for the
next few values of the spin quantum numbers.
4.4.2 j = 4
For j = 4, there is no anticoherent state to order
t = 4. We find that the optimal state for small angles
of rotation is the 3-anticoherent state
|ψi =
q
5
24
|4, −4i −
q
7
12
|4, 0i −
q
5
24
|4, 4i, (64)
with A
1
= A
2
= A
3
= 1 and A
4
= 281/288.
4.4.3 j = 9/2
For j = 9/2, there is no anticoherent state to order
t > 3. The anticoherent states of order t = 2 with the
largest A
3
are found to be of the form
|ψi =
√
13
8
|
9
2
, −
9
2
i + e
iχ
q
15
32
|
9
2
, −
1
2
i −
√
21
8
|
9
2
,
7
2
i,
(65)
with χ ∈ [0, π/2]. Their measures of antico-
herence are A
1
= A
2
= 1, A
3
= 2347/2352
and A
4
= 5
355609 + 175
√
273 cos(2χ)
/1806336.
Among these states, the one with χ = 0 has the
largest value of A
4
and numerical results suggest that
this is the optimal state for small angles of rotation.
4.4.4 j = 5
For j = 5, there is no anticoherent state to order
t > 4. We find that the optimal state for small angles
is the 3-anticoherent state
|ψi =
q
5
16
|5, −4i +
q
3
8
|5, 0i −
q
5
16
|5, 4i, (66)
with A
1
= A
2
= A
3
= 1, A
4
= 895/896 and A
5
=
1097/1120.
4.4.5 Arbitrary values of j
As was mentioned earlier, if an anticoherent state to
order bjc exists for a given j, then this state gives rise
to an optimal quantum rotosensor for η ∈ [0, η
0
]. This
applies to values j = 1, 3/2, 2 and j = 3, which are
the only cases where existence of anticoherent states
to order t = bjc has been established (see e.g. [33, 20]).
The situation is less straightforward if such a state
is not known to exist from the outset. The only gen-
eral conclusion one can draw is that minimizing the
average fidelity F
|ψi
(η) for a fixed angle η ∈ [0, η
0
]
corresponds to maximizing the measures A
t
within
the domain Ω (by definition, Ω is the set of all reach-
able A
t
so that by changing |ψi, we will remain within
Ω). In this sense, the more anticoherent a state is, the
more sensitive it will be as a quantum rotosensor. In
general, varying |ψi will change all anticoherence mea-
sures simultaneously. The challenge is to determine
whether a state with given values of the measures A
t
exists and, if it does, to identify it.
The maximal order of anticoherence that a spin-j
state can display is generally much smaller than bjc,
typically t ∼ 2
√
j for large spins j [33]. Numerical
results for j . 100 seem to suggest that the pairs (t, j)
for which a t-anticoherent spin-j state exists coincide
with those for which a 2j-points spherical t-design
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 11
j |ψ
optimal
i A
t
Interval
1
|ψ
cat
i
any state
|j, ji
A
1
= 1
0 6 A
1
6 1
A
1
= 0
η ∈ [0, η
0
[
η = η
0
η ∈ [η
0
, π]
3/2
|ψ
cat
i
any state
|j, ji
A
1
= 1
0 6 A
1
6 1
A
1
= 0
η ∈ [0, η
0
[
η = η
0
η ∈ [η
0
, π]
2
|ψ
tet
i
|ψ
cat
i
|j, ji
A
1
= A
2
= 1
A
1
= 1, A
2
= 3/4
A
1
= A
2
= 0
η ∈ [0, η
1
], η
1
≈ 1.68374
η ∈ [η
1
, η
2
]
η ∈ [η
2
, π], η
2
≈ 2.44264
5/2
Eq. (56)
|ψ
cat
i
|j, ji
A
1
= 1, A
2
= 99/100
A
1
= 1, A
2
= 3/4
A
1
= A
2
= 0
η ∈ [0, η
1
], η
1
≈ 1.49697
η ∈ [η
1
, η
2
]
η ∈ [η
2
, π], η
2
≈ 2.2521
3
|ψ
oct
i
|ψ
cat
i
|j, ji
A
1
= A
2
= A
3
= 1
A
1
= 1, A
2
= 3/4, A
3
= 2/3
A
1
= A
2
= A
3
= 0
η ∈ [0, η
1
] ∪ [η
3
, η
4
], η
3
≈ 2.35881
η ∈ [η
1
, η
2
], η
1
≈ 1.3635, η
2
≈ 2.04367
η ∈ [η
2
, η
3
] ∪ [η
4
, π], η
4
≈ 2.65576
7/2
Eq. (60)
−
|ψ
cat
i
−
|j, ji
A
1
= A
2
= 1, A
3
= 1198/1215
195
196
6 A
2
6 1,
1198
1215
6 A
3
6
146
147
, see Fig. 5
A
1
= 1, A
2
= 3/4, A
3
= 2/3
see Fig. 4
A
1
= A
2
= A
3
= 0
η → 0
η ∈ [0, η
1
], η
1
≈ 0.71718
η ∈ [η
2
, η
3
] ∪ [η
4
, η
5
], η
2
≈ 1.24169
η ∈ [η
3
, η
4
], η
3
≈ 1.60141, η
4
≈ 1.88334
η ∈ [η
5
, π], η
5
≈ 2.41684
Table 1: Summary of the results of Secs. 4.2 and 4.3 on optimal states for 1 6 j 6 7/2. Here, η
0
denotes the first strictly
positive zero of ϕ
(j)
1
(η), |ψ
tet
i defined for j = 2 is given by Eq. (52), |ψ
oct
i defined for j = 3 is given by Eq. (58), and
|ψ
cat
i =
1
√
2
(|j, −ji + |j, ji) for any j. The state |j, ji has been taken as an example of coherent state. Note that optimal
states given here are not necessarily unique (states not related by a rotation can have the same A
t
).
exists in three dimensions [34]. The latter have been
tabulated up to j = 50 [31]. For example, the first
pairs (t, j) for j 6 4 are given by (1, 1), (1, 3/2), (2, 2),
(1, 5/2), (3, 3), (2, 7/2), (3, 4).
5 Summary and conclusions
The main result of this work is a closed-form expres-
sion (3) for the fidelity F
|ψi
(η) between a state and its
image under a rotation by an angle η about an axis
n, averaged over all rotation axes. The expression
takes the form of a linear combination of anticoher-
ence measures A
t
, with explicit η-dependent coeffi-
cients. It follows that not only spin-j states which
are related by a global rotation of the axes come with
the same average fidelity, but more generally all states
with identical purities of their reduced density matri-
ces (calculated for any subset of their 2j constituent
spin-1/2 in the Majorana representation). This gives
an explanation for the observation of [15] that optimal
states are not necessarily unique. Moreover, since the
fidelity is linear in the anticoherence measures, opti-
mal states correspond to values of A
t
on the boundary
of the domain Ω of admissible values. This shows the
relevance of characterizing the domain Ω.
The expression (3) allows us to characterize states
which optimally detect rotations by their degree of
coherence or anticoherence. At small angles η 6 η
0
,
where the coefficients of the measures A
t
are all nega-
tive, optimality of detection of rotations goes hand in
hand with high degrees of anticoherence. For angles
close to η = π, however, numerical results support the
claim that optimality is achieved throughout by spin
coherent states.
We also performed a systematic investigation of
states minimizing the average fidelity for small values
of j, for all integers and half-integers from j = 1/2 to
j = 5. Table 1 summarizes our findings for the low-
est values of j. At small rotation angle, all optimal
states were found to have a maximal lowest antico-
herence measure: A
1
= 1. These states, which are
anticoherent to order 1, exist for any value of j, and
one may conjecture that they should, in fact, be op-
timal for arbitrary values of j. More generally, for all
values of j investigated and for η 6 η
0
, the optimal
states turned out to have, for each t > 1, the largest
admissible anticoherence measure A
t
compatible with
fixed values of the lower measures A
1
, A
2
, . . . , A
t−1
.
Whether this property holds in general remains an
open question.
Note that natural generalizations of this problem,
such as maximization of the average fidelity, can also
be addressed by our approach. For instance, for small
rotation angles η ∈ [0, η
0
], where all ϕ
(j)
t
(η) with t > 1
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 12
are negative, the average fidelity is maximal for co-
herent states. For rotation angles close to η = π, nu-
merical results indicate that the 1-anticoherent state
|ψ
cat
i =
1
√
2
(|j, −ji + |j, ji) is optimal for all j up to
17/2.
Acknowledgments
OG and SW thank the hospitality of the University
of Liège, where this work has been initiated.
A Fidelity in parameter estimation the-
ory of rotations
It was shown in [13] that minimizing the uncertainty
in the measurement of an unknown angle about a
known rotation axis is equivalent to identifying the
states which minimize the fidelity F
|ψi
(η, n), assum-
ing small rotation angles and using parameter esti-
mation theory. To see this, first expand the fidelity
as
|hψ|R
n
(η)|ψi|
2
= 1 − η
2
(∆J
n
)
2
+ O(η
4
) , (67)
where (∆J
n
)
2
is the variance of J
n
≡ J· n in the state
|ψi. Solving (67) for the angle η will, upon measuring
the fidelity in any state |ψi, result in an estimate of
η. Second, the accuracy of this value depends on the
initial state |ψi: using error propagation, one finds
that the variance of the estimator is approximately
given by
(∆η)
2
≈
1
(2∆J
n
)
2
. (68)
Thus, states |ψi with large values of the variance
(∆J
n
)
2
are seen to minimize the uncertainty of the
angle η. According to Eq. (67), these states also min-
imize the fidelity F
|ψi
(η, n).
Let us generalize the argument to the case in which
the rotation axis is unknown. We will see that the
states producing the most reliable results—i.e. the
smallest variance in the angle estimator—are those
which minimize the average fidelity F
|ψi
(η). It is
convenient to describe the randomness in the rota-
tion axis in terms of a quantum channel (see for in-
stance [35] for the use of channels in quantum esti-
mation theory). Suppose we prepare the pure initial
state ρ
0
= |ψihψ| and send it through the η-dependent
channel Λ
η
(·),
ρ
η
= Λ
η
(ρ
0
) =
1
4π
Z
S
2
R
n
(η)ρ
0
R
†
n
(η) dn , (69)
which describes rotations by η about all possible ro-
tation axes. Next, we measure the projector ρ
0
=
|ψihψ|. Assuming the rotation angle to be small,
η 1, the probability to still find the propagated
state ρ
η
in the initial state ρ
0
is given by
hρ
0
i
η
= tr [ρ
0
ρ
η
] = 1 − η
2
V + O(η
4
) , (70)
where V is the variance of J
n
≡ J · n, averaged over
all directions,
V =
1
4π
Z
S
2
hψ|J
2
n
|ψi − hψ|J
n
|ψi
2
dn . (71)
Using ρ
2
0
= ρ
0
and the relation (70), the variance of
the measurement outcomes is found to be
(∆ρ
0
)
2
η
= tr
ρ
2
0
ρ
η
− (tr [ρ
0
ρ
η
])
2
= η
2
V + O(η
4
) .
(72)
Now using again the error propagation formula in-
strumental in the derivation of Eq. (68) about known
rotation axes, its generalization to unknown axes is
given by
(∆η)
2
=
(∆ρ
0
)
2
η
|∂hρ
0
i
η
/∂η|
2
+ O(η
2
) ≈
1
4V
. (73)
This result concludes the argument we wish to pro-
vide: it is of physical interest to minimize the average
fidelity
F
|ψi
(η) ≡ tr [ρ
0
ρ
η
] ≈ 1 − η
2
V , (74)
since the states which do so are those states which al-
low one to most accurately estimate a (small) rotation
angle about unknown axes.
B Average fidelity for Dicke states
For Dicke states |j, mi (common eigenstates of J
2
and
J
z
), the average fidelity (2) reads
F
|j,mi
(η) =
1
4π
Z
S
2
|hj, m|R
n
(η)|j, mi|
2
dn
=
1
4π
Z
S
2
|U
j
mm
(η, n)|
2
dn
(75)
with U
j
mm
(η, n) ≡ U
j
mm
a matrix element of the rota-
tion operator in the angle-axis parametrization given
by
U
j
mm
=
√
4π
2j + 1
X
λ,µ
(−i)
λ
√
2λ + 1χ
j
λ
(η)C
jm
jmλµ
Y
m
λ
(n)
(76)
where C
jm
jmλµ
are Clebsch-Gordan coefficients, Y
m
λ
(n)
are spherical harmonics and χ
j
λ
(η) are the generalized
characters of order λ of the irreducible representations
of rank j of the rotation group [36]. These are defined
by
χ
j
λ
(η) =
q
(2j+1)(2j−λ)!
(2j+λ+1)!
sin
λ
η
2
d
d cos
η
2
λ
χ
j
(η)
(77)
with the characters
χ
j
(η) =
(4j + 2)!!
2(4j + 1)!!
P
1
2
,
1
2
2j
cos
η
2
(78)
where P
(α,β)
n
are Jacobi polynomials. Taking the
modulus squared of (76) and integrating over all di-
rections by using orthonormality of the spherical har-
monics, we readily get Eq. (29).
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 13
C Explicit calculation of the angular
functions ϕ
(j)
t
(η)
C.1 Matrices S
µ
1
µ
2
...µ
N
The matrices S
µ
1
µ
2
...µ
N
with N = 2j appearing in
the expansion (11) can be obtained by expanding the
(j, 0) representation of a Lorentz boost,
Π
(j)
(q) ≡ (q
2
0
− |q|
2
)
j
e
−2θ
q
ˆ
q·J
, (79)
with θ
q
= arctanh(−|q|/q
0
) and
ˆ
q = q/|q|. This
expansion takes the form of a multivariate polynomial
in the variables q
0
, q
1
, q
2
, q
3
,
Π
(j)
(q) = (−1)
2j
q
µ
1
q
µ
2
. . . q
µ
2j
S
µ
1
µ
2
...µ
2j
, (80)
where the coefficients are the (N+1)×(N +1) matrices
S
µ
1
µ
2
...µ
N
[24].
C.2 Tensor coordinates of the maximally mixed
state
The maximally mixed state ρ
0
= 1/(N + 1) can be
expanded along (11) with coefficients x
(0)
µ
1
µ
2
...µ
N
. The
coherent state decomposition of the maximally mixed
state, ρ
0
=
1
4π
R
S
2
|nihn|dn, yields the identity
x
(0)
µ
1
µ
2
...µ
N
=
1
4π
Z
S
2
n
µ
1
n
µ
2
. . . n
µ
N
dn. (81)
Using our convention not to write indices when they
are equal to 0, we have, irrespective of spin size, x
(0)
0
=
1, x
(0)
aa
= 1/3, x
(0)
aaaa
= 1/5 and x
(0)
aabb
= 1/15 for a 6=
b. More generally, the coefficients of the maximally
mixed state are given by the polynomial identity (cf.
Eq. (27) of [24])
x
(0)
µ
1
µ
2
...µ
N
q
µ
1
. . . q
µ
N
=
j
X
k=0
N
2k
2k + 1
q
N−2k
0
|q|
2k
, (82)
which leads to
x
(0)
a
1
a
2
...a
N
=
1
N + 1
j
p
1
/2,p
2
/2,p
3
/2
N
p
1
,p
2
,p
3
, (83)
where p
i
denotes the number of i in {a
1
, a
2
, . . . , a
N
}
and the terms in the fraction are multinomial coeffi-
cients (by convention the right-hand side evaluates to
zero if some p
i
is not even).
C.3 Average fidelity in terms of tensor coordi-
nates
According to Eq. (40), the average fidelity can be writ-
ten as a double sum,
F
|ψi
(η) =
N
X
k=0
(−1)
N
q
2(N−k)
0
m
2N
×
X
µ,ν
2(N−k)zeros
(−1)
nr of 0 in ν
x
(0)
µ
1
...µ
N
ν
1
...ν
N
x
µ
1
...µ
N
x
ν
1
...ν
N
.
(84)
We now wish to show that the second sum which runs
over all strings of indices (between 0 and 3) containing
2(N −k) zeros can evaluated explicitly leading to the
simpler form for F
|ψi
(η) given in Eq. (99) at the end
of this section.
The sum runs over terms containing 2(N −k) zeros,
that is, 2k non-zero indices. We split it into terms
containing r nonzero indices in µ and 2k −r in ν . At
fixed k we have
X
µ,ν
2(N−k)zeros
(−1)
nr of 0 in ν
x
µ
1
...µ
N
x
ν
1
...ν
N
x
(0)
µ
1
...µ
N
ν
1
...ν
N
=
N
X
r=2k−N
(−1)
N−2k+r
N
r
N
2k − r
×
×
X
a
i
,b
i
x
a
1
...a
r
x
b
1
...b
2k−r
x
(0)
a
1
...a
r
b
1
...b
2k−r
. (85)
We now evaluate the sums
P
a
i
,b
i
x
a
1
...a
r
x
b
1
...b
2k−r
x
(0)
a
1
...a
r
b
1
...b
2k−r
. We may
suppose that r 6 2k − r. Using (83), we see that the
nonzero indices a
i
and b
i
must occur in pairs. Indices
a
i
are either paired with indices a
k
or indices b
k
. We
can then split the sum according to the number of
pairings of the form (a
i
, b
i
) (all other pairings are
then within the a
i
or within the b
i
). Let us first
consider the case k = r. We are going to show that
X
a
i
,b
i
x
a
1
...a
r
x
b
1
...b
r
x
(0)
a
1
...a
r
b
1
...b
r
=
λ
0
X
a
i
x
2
a
1
...a
r
+ λ
1
X
a
i
X
b
x
a
1
...a
r−2
bb
!
2
+ λ
2
X
a
i
X
b
1
,b
2
x
a
1
...a
r−4
b
1
b
1
b
2
b
2
2
+ . . .
(86)
with
λ
q
=
2
r−2q
r!
2
(2r + 1)!
r
r − 2q, q, q
. (87)
We first use the explicit expression (83) of x
(0)
a
1
...a
r
b
1
...b
r
to get an equation equivalent to (86), namely
X
c
i
x
c
1
...c
r
x
c
r+1
...c
2r
2r
r
r
p
1
/2 p
2
/2 p
3
/2
2r
p
1
p
2
p
3
= 2
r
X
a
i
x
2
a
1
...a
r
+ 2
r−2
r
r − 2, 1, 1
X
a
i
X
b
x
a
1
...a
r−2
bb
!
2
+ ···
+ 2
r−2q
r
r − 2q, q, q
X
a
i
X
b
x
a
1
...a
r−2q
b
1
b
1
...b
q
b
q
!
2
+ ··· , (88)
where p
i
is the number of i in {c
1
, c
2
, . . . , c
2r
} and
terms with p
i
odd are zero. In order to prove Eq. (88),
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 14
we just observe that it represents two different ways
of counting the same quantity. Indeed, let η
i
=
{a
i
,
i
,
0
i
} for 1 6 i 6 r be triplets with 1 6 a
i
6 3
and 0 6
i
,
0
i
6 1. To a given set η = {η
1
, . . . , η
r
} we
associate a term of the form x
c
1
...c
r
y
c
r+1
...c
2r
where
the c
i
occur in pairs (a
1
, a
1
), (a
2
, a
2
), . . . , (a
r
, a
r
). In
a pair (a
i
, a
i
), the first a
i
is assigned to be an index of
x if
i
= 0, of y if
i
= 1 (and similarly the second a
i
in
the pair is an index of x if
0
i
= 0, of y otherwise). For
instance, η = (a, 0, 0) corresponds to a term x
aa...
y
...
and η = (a, 0, 1) corresponds to a term x
a...
y
a...
. In or-
der that x and y have the same number r of indices we
need to have
P
i
(
i
+
0
i
) = r, so that among the
i
,
0
i
there are r 0’s and r 1’s. Each η = {η
1
, . . . , η
r
} such
that
P
i
(
i
+
0
i
) = r then corresponds to a unique term
of the form x
c
1
...c
r
y
c
r+1
...c
2r
. Consider now, for some
q 6 r, all η with
P
i
(
i
+
0
i
) = r for which
i
=
0
i
= 0
for exactly q values of i. These correspond to terms
x
c
1
...c
r
y
c
r+1
...c
2r
such that exactly q pairs (a
i
, a
i
) ap-
pear as indices of x, q pairs appear as indices of y, and
r − 2q are distributed over x and y, i.e. terms of the
form x
a
1
a
2
...a
r−2q
b
1
b
1
b
2
b
2
...b
q
b
q
y
a
1
a
2
...a
r−2q
c
1
c
1
c
2
c
2
...c
q
c
q
.
Replacing y by x, all these terms are those appear-
ing in the right-hand side of (88). In fact, each sum
on the right-hand side of (88) can be interpreted as
the sum over all η
i
such that
P
i
(
i
+
0
i
) = r and
i
=
0
i
= 0 for exactly q values of i. For instance the
first term on the right-hand side of (88) corresponds
to terms q = 0, where all pairs (a
i
, a
i
) are distributed
over the two different strings of indices (and then of
course replacing y by x). The prefactors correspond
to the ways of choosing the positions of a given set of
pairs: the multinomial coefficient corresponds to the
choice of positions of the indices among the r indices
of x
a
1
a
2
...a
r−2q
b
1
b
1
b
2
b
2
...b
q
b
q
. The factor 2
r−2q
corre-
sponds to choosing between x and y for the r − 2q
indices a
i
which are distributed over x and y.
The same sum can be expressed as the left-hand-
side of (88) if we now first sum over all strings
c
1
6 c
2
6 ··· 6 c
2r
, which implies dividing by the
number of permutations
2r
p
1
,p
2
,p
3
, then consider all
possible positions of the a
i
over the r pairs, which
implies multiplying by the number of permutations of
the pairs
r
p
1
/2,p
2
/2,p
3
/2
, and finally choose the r en-
tries among the
i
and
0
i
that will take the value 0,
hence the factor
2r
r
. Thus (88) holds, which proves
(86). The tracelessness condition (12) then allows us
to reduce the sums over b in (86) to invariants κ
r
. We
simply get
X
a
i
,b
i
x
a
1
...a
r
x
b
1
...b
r
x
(0)
a
1
...a
r
b
1
...b
r
= λ
0
κ
r
+ λ
1
κ
r−2
+ λ
2
κ
r−4
+ . . .
(89)
Following exactly the same procedure from (86) to
(89) in the case where the strings of indices of x and
y have different lengths, we obtain the more general
expression
X
a
i
,b
i
x
a
1
...a
r
x
b
1
...b
2k−r
x
(0)
a
1
...a
r
b
1
...b
2k−r
=
r!(2k − r)!
(2k + 1)!
b
r
2
c
X
q=0
2
r−2q
k
r − 2q, q, q + k − r
κ
r−2q
.
(90)
From (84) we finally get
F
|ψi
(η) =
N
X
k=0
(−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
2k
X
r=0
(−1)
r
N!
2
(N − r)!(N −2k + r)!(2k + 1)!
×
b
r
2
c
X
q=0
2
r−2q
k
r − 2q, q, q + k − r
κ
r−2q
.
(91)
Changing the summation over r to a summation over s = r − 2q, we get
F
|ψi
(η) =
N
X
k=0
(−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
2k
X
s=0
(−2)
s
κ
s
×
k−b
s+1
2
c
X
q=0
N
s + 2q
N
2k − 2q − s
(s + 2q)!(2k − 2q − s)!
(2k + 1)!
k
s, q, k − q − s
.
(92)
Because of the multinomial coefficient at the end of (92), the sum over s can be restricted to s 6 k and the sum
over q to q 6 k − s, yielding
F
|ψi
(η) =
N
X
k=0
(−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
N!
2
k!
(2k + 1)!
×
k
X
s=0
(−2)
s
s!(2N − 2k)!(k − s)!
k−s
X
q=0
2N − 2k
N − s − 2q
k − s
q
κ
s
.
(93)
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 15
Grouping the κ
s
together by changing the order of the sum we get
F
|ψi
(η) = N!
2
N
X
s=0
(−2)
s
κ
s
s!
N
X
k=s
(−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
k!
(2k + 1)!
×
k−s
X
q=0
1
(N − s − 2q)!(N − 2k + s + 2q)!(k − s − q)!q!
.
(94)
Because of the sum over q from 0 to k −s, we can make the sum over k start at 0. We then use (19) to express
the κ
s
in terms of tr
ρ
2
t
. This gives
F
|ψi
(η) = N!
2
N
X
t=0
(−2)
t
t!
tr
ρ
2
t
N
X
k=0
(−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
k!
(2k + 1)!
×
N
X
s=t
2
s
(s − t)!
k−s
X
q=0
1
(N − s − 2q)!(N − 2k + s + 2q)!(k − s − q)!q!
.
(95)
It turns out that the sums in the second line of this expression can be performed. Indeed, the identity
N
X
s=t
2
s
(s − t)!
k−s
X
q=0
1
(N − s − 2q)!(N − 2k + s + 2q)!(k − s − q)!q!
=
2
t
(2N − 2t)!
(N − t)!
2
(k − t)!(2N −2k)!
(96)
holds for arbitrary N, t, k. This can be proved as follows. First change variables N → N − t, k → k − t and
s → s −t, so that showing (96) amounts to showing
k
X
s=0
2
s
s!
k−s
X
q=0
1
(N − s − 2q)!(N − 2k + s + 2q)!(k − s − q)!q!
=
(2N)!
N!
2
k!(2N − 2k)!
(97)
(the upper bound of the sum over s can be changed from N to k since terms s > k do not contribute).
Equation (97) can be rewritten
k
X
s=0
k−s
X
q=0
2
s
k
s
k − s
q
2N − 2k
N − s − 2q
=
2N
N
. (98)
Such an identity can be proven by writing (1 + x)
2N
= (1 + 2x + x
2
)
k
(1 + x)
2N−2k
for any k and any x, and
expanding the first factor using multinomial coefficients and the second one using binomial coefficients:
(1 + x)
2N
=(1 + 2x + x
2
)
k
(1 + x)
2N−2k
=
X
s,q
k
s, q, k − s − q
(2x)
s
(x
2
)
q
X
u
2N − 2k
u
x
u
=
X
s,q,u
2
s
k
s
k − s
q
2N − 2k
u
x
u+s+2q
(the boundaries of the sums are taken care of by the binomial coefficients which vanish outside a certain range
of parameters). Identifying the coefficients of the term in x
N
readily gives (98).
Using (96), Eq. (95) finally reduces to
F
|ψi
(η) =
1
2N + 1
1
2N
N
N
X
t=0
(−4)
t
2N − 2t
N − t
tr
ρ
2
t
N
X
k=0
(−1)
k
sin
2k
η
2
cos
2(N−k)
η
2
2N + 1
2k + 1
k
t
. (99)
D Angular functions for j = 2
Evaluating the expression (44) for j = 2 leads to these three angular functions:
ϕ
(2)
0
(η) =
1
315
(130 cos(η) + 46 cos(2η) + 10 cos(3η) + cos(4η) + 128) ,
ϕ
(2)
1
(η) = −
4
315
(10 cos(η) − 11 cos(2η) + 16 cos(3η) − 20 cos(4η) + 5) ,
ϕ
(2)
2
(η) = −
64
105
sin
4
η
2
(10 cos(η) + 5 cos(2η) + 6) .
(100)
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 16
E Sample code
We give here a short sample code written in Mathematica
TM
to find an optimal state for j = 5/2 and η = 0.5.
1 (* An gular functions , see Eqs . (42) , (45) and (44) *)
2 a[n_ , t_ , k_ ]:=(( -1) ^( k+t)*4^ t) /(2* k +1) *( B inomial [2*n ,2* k ]* Binomial [2* n -2*t ,n - t]
Binomial [k,t ]) / Binomial [2*n,n ];
3 b[n_ , t_ , k_ ]:= If [t==0 , Binomi al [n,k ]/(1+ 2* k ) , -( t /( t +1) ) * If[t == n /2 ,1/2 ,1]*( a[n,t,k ]+ a[n,
n -t ,k ]) ];
4 phi [n_ , t_ , eta_ ]:= Sum [ Sin [ eta / 2]^(2* k )*Cos [ eta /2]^(2*(n- k))*b[n ,t ,k ] ,{k ,0 ,n }];
5
6 (* Measu res of a ntico heren ce of a pure state of the form (25) ,
7 see Eqs . (10) and (27) *)
8 A[cm__ ,n_ , t_ ]:=( t +1) /t *(1 - Sum [ Abs [ Sum [ Conjugate [ cm [[ k+l +1]]]* cm [[ k+q +1]]* Sqrt [
Binomial [ k+l,k ]* Binomial [n -k -l ,t - l ]* Binomi al [ k+q,k ]* Binomial [n -k -q ,t - q ]]/ Binom ial
[n ,t] ,{k ,0 ,n- t }]^2] ,{ q ,0 , t } ,{ l ,0 , t }])
9
10 (* Av erage fidelity , see Eq . (3) *)
11 F[cm__ ,n_ , eta_ ]:= phi [n ,0 , eta ]+ Sum [ phi [n ,t, eta ]* A[cm ,n,t ] ,{ t ,1 , Floor [ n /2] }];
12
13 (* Normali zed state expressed in the Dicke basis for j =5/2 *)
14 j =5/2; n=2* j;
15 cm = N ormalize@ ( Array [r , n +1 ,0]+ I* Array [i ,n+1 ,0]) ;
16
17 (* Search for an optimal state for eta =0.5 *)
18 eta =0.5;
19 f= Sim pli fy [ Comp lexEx pand [ F[ cm ,n, eta ]]];
20 sol = NMin imize [f, Array [r , n +1 ,0]~ Join ~ Array [i ,n+1 ,0] , Ac cura cyGo al - >25 , P reci sionGoal
- >25];
21
22 (* Mi nimal avera ge f ide lity *)
23 Re@sol [[1]]
24
25 (* Op timal state in the Dicke basis *)
26 cmsol = cm /. sol [[2]]
27
28 (* Measu res of a ntico heren ce of order 1 and 2 *)
29 A[cmsol ,n ,1]
30 A[cmsol ,n ,2]
The evaluation of the code with Mathematica 12.0
yields the output
1 0. 453 337
2
3 { 0.189 461+0.4 8194 I, -0.15 5904 -0.0488534
I , 0.082 86 66+0.00 44 0845 I ,
0. 374917+ 0.583 96 7 I ,
0.257018 -0.0504367 I,
-0.165557+0.347373 I }
4
5 1.
6
7 0.99
The state that is found, with measures of antico-
herence A
1
= 1 and A
2
= 99/100, can be shown to
be related by a rotation to the state (56).
References
[1] W. Heisenberg, Über den anschaulichen In-
halt der quantentheoretischen Kinematik und
Mechanik, Z. Phys. 43, 172 (1927).
[2] V. Giovannetti, S. Lloyd, L. Maccone, Advances
in quantum metrology, Nature Photonics 5, 222
(2011).
[3] P. Busch, P. Lahti, J.-P. Pellonpää, K. Ylinen:
Quantum Measurement, Springer 2016.
[4] W. Nawrocki, Introduction to Quantum Metrol-
ogy, Springer Nature 2019.
[5] A. Peres and P. F. Scudo, Transmission of a
Cartesian Frame by a Quantum System, Phys.
Rev. Lett. 87, 167901 (2001).
[6] H. Hakoshima and Y. Matsuzaki, Efficient
detection of inhomogeneous magnetic fields
from the single spin with Dicke states,
arXiv:2003.12524.
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 17
[7] C. R. Rao, Information and the accuracy attain-
able in the estimation of statistical parameters,
Bulletin of the Calcutta Mathematical Society
37, 81 (1945).
[8] H. Cramér, Mathematical Methods of Statistics
(PMS-9), Princeton University Press 1946.
[9] C. W. Helstrom, Quantum Detection and Esti-
mation Theory, volume 123 of Mathematics in
Science and Engineering, Elsevier (1976).
[10] M. Hübner, Explicit computation of the Bu-
res distance for density matrices, Phys. Lett. A
163, 239 (1992).
[11] M. Hübner, Computation of Uhlmann’s paral-
lel transport for density matrices and the Bu-
res metric on three-dimensional Hilbert space,
Phys. Lett. A 179, 226 (1993).
[12] P. Kolenderski and R. Demkowicz-Dobrzanski,
Optimal state for keeping reference frames
aligned and the Platonic solids, Phys. Rev. A
78, 052333 (2008).
[13] A. Z. Goldberg and D. F. V. James, Quantum-
limited Euler angle measurements using antico-
herent states, Phys. Rev. A 98, 032113 (2018).
[14] Y. Mo and G. Chiribella, Quantum-enhanced
learning of rotations about an unknown direc-
tion, New J. Phys. 21, 113003 (2019).
[15] C. Chryssomalakos and H. Hernández-
Coronado, Optimal quantum rotosensors,
Phys. Rev. A 95, 052125 (2017).
[16] F. T. Arecchi, E. Courtens, R. Gilmore, and H.
Thomas, Atomic Coherent States in Quantum
Optics, Phys. Rev. A 6, 2211 (1972).
[17] J. Zimba, “Anticoherent” Spin States via the
Majorana Representation, Electr. J. Theor.
Phys. 3, 143 (2006).
[18] F. Bouchard, P. de la Hoz, G. Björk, R. W.
Boyd, M. Grassl, Z. Hradil, E. Karimi, A.
B. Klimov, G. Leuchs, J. Rehacek, and L. L.
Sánchez-Soto, Quantum metrology at the limit
with extremal Majorana constellations, Optica
4, 1429 (2017).
[19] T. Chalopin, C. Bouazza, A. Evrard, V.
Makhalov, D. Dreon, J. Dalibard, L. A.
Sidorenkov, and S. Nascimbene, Quantum-
enhanced sensing using non-classical spin states
of a highly magnetic atom, Nature Communica-
tions 9, 4955 (2018).
[20] D. Baguette and J. Martin, Anticoherence mea-
sures for pure spin states, Phys. Rev. A 96,
032304 (2017).
[21] L. C. Biedenharn and J. D. Louck, Angular Mo-
mentum in Quantum Physics, Cambridge Uni-
versity Press 1984.
[22] I. Bengtsson and K. Życzkowski, Geometry of
Quantum States : An Introduction to Quan-
tum Entanglement, 2nd ed. Cambridge Univer-
sity Press 2017.
[23] B. Coecke, A Representation for a Spin-S En-
tity as a Compound System in R
3
Consisting of
2S Individual Spin-1/2 Entities, Foundations of
Physics 28, 1347 (1998).
[24] O. Giraud, D. Braun, D. Baguette, T. Bastin,
and J. Martin, Tensor representation of spin
states, Phys. Rev. Lett. 114, 080401 (2015).
[25] S. Weinberg, Feynman Rules for Any Spin,
Phys. Rev. 133, B1318 (1964).
[26] D. Baguette, T. Bastin, and J. Martin, Mul-
tiqubit symmetric states with maximally mixed
one-qubit reductions, Phys. Rev. A 90, 032314
(2014).
[27] O. Giraud, P. Braun, and D. Braun, Quantify-
ing Quantumness and the Quest for Queens of
Quantum, New J. Phys. 12, 063005 (2010).
[28] G. Björk, A. B. Klimov, P. de la Hoz, M. Grassl,
G. Leuchs, L. L. Sánchez-Soto, Extremal quan-
tum states and their Majorana constellations,
Phys. Rev. A 92, 031801(R) (2015).
[29] G. Björk, M. Grassl, P. de la Hoz, G. Leuchs and
L. L. Sánchez-Soto, Stars of the quantum Uni-
verse: extremal constellations on the Poincaré
sphere, Phys. Scr. 90, 108008 (2015).
[30] P. Delsarte, J. M. Goethals, J. J. Seidel, Spher-
ical codes and designs, Geometriae Dedicata 6,
363 (1977).
[31] R. H. Hardin and N. J. A. Sloane, McLaren’s
Improved Snub Cube and Other New Spheri-
cal Designs in Three Dimensions, Discrete and
Computational Geometry 15, 429 (1996).
[32] R. E. Schwartz, The Five-Electron Case of
Thomson’s Problem, Experimental Mathemat-
ics 22, 157 (2013).
[33] D. Baguette, F. Damanet, O. Giraud, and J.
Martin, Anticoherence of spin states with point-
group symmetries, Phys. Rev. A 92, 052333
(2015).
[34] See e.g. http://polarization.markus-grassl.de/,
http://www.oq.ulg.ac.be and Refs. [20, 28].
[35] J. S. Sidhu and P. Kok, A Geometric Perspec-
tive on Quantum Parameter Estimation, AVS
Quantum Science 2, 014701 (2020).
[36] D. A. Varshalovich, A. N. Moskalev, V. K.
Khersonskii, Quantum Theory Of Angular Mo-
mentum, World Scientific (1988).
Accepted in Quantum 2020-06-16, click title to verify. Published under CC-BY 4.0. 18