Demonstration of quantum controlled teleportation for discrete variables on linear optical devices

TL;DR

This paper reports an experimental demonstration of controlled quantum teleportation using linear optical devices with discrete variables, showing high fidelity and revealing that tripartite entanglement is not essential for the protocol.

Contribution

The work experimentally implements controlled quantum teleportation with discrete variables and demonstrates that biseparable states suffice, challenging the necessity of tripartite entanglement.

Findings

Achieved teleportation fidelity of around 83%.

Controlled teleportation can be performed without tripartite entanglement.

Lowered resource requirements for controlled quantum teleportation.

Abstract

We report an experimental implementation of tripartite controlled quantum teleportation on the quantum optical devices. The protocol is performed through bi- and tripartite entangled channels of discrete variables and qubits encoded in polarization of individual photons. The experimental results demonstrate successful controlled quantum teleportation with a fidelity around $83\%$, well above the classical limit. By realizing the controlled quantum teleportation through biseparable state, we show that tripartite entangled is not a necessary resource for controlled quantum teleportation and the controller's capability to allow or prohibit the teleportation cannot be considered to be a manifestation of tripartite entanglement. These results open new possibilities for further application of controlled quantum teleportation by lowering teleportation channel's requirements.

Tables

Table 1. Table 1: Measured fidelities for the linearly polarized input state | ψ 4 ⟩ = 1 2 ​ { | H 4 ⟩ + | V 4 ⟩ } ket subscript 𝜓 4 1 2 ket subscript 𝐻 4 ket subscript 𝑉 4 |\psi_{4}\rangle=\frac{1}{\sqrt{2}}\{|H_{4}\rangle+|V_{4}\rangle\} and several teleportation channels: ρ ref subscript 𝜌 ref \rho_{\textrm{ref}} denotes two-photon teleportation channel | Φ 12 + ⟩ ​ ⟨ Φ 12 + | ket subscript superscript Φ 12 bra subscript superscript Φ 12 |\Phi^{+}_{12}\rangle\langle\Phi^{+}_{12}| with photon in mode 3 3 3 serving only as trigger (see text) while ρ ​ ( p ) 𝜌 𝑝 \rho(p) is given in Eq. ( 1 ). The last two columns correspond to the process when controller allows and denies the teleportation, respectively. All uncertainties are obtain by numerical calculations assuming a Poisson distribution of the four-fold coincidences.

channel	$F_{\text{allowed}}(\%)$	$F_{\text{denied}}(\%)$
${\rho }_{\text{ref}}$	$83.1\pm 4.9$	–
$\rho \left(p=0\right)$	$77.9\pm 8.1$	$57.2\pm 5.0$
$\rho \left(p=1\right)$	$83.0\pm 7.5$	$51.8\pm 6.7$
$\rho \left(p=1/2\right)$	$80.2\pm 5.7$	$55.1\pm 5.0$

Equations

$$\rho (p)=(1-p)|{\mathcal{G}}^{(1)}⟩⟨{\mathcal{G}}^{(1)}|+p|{\mathcal{G}}^{(2)}⟩⟨{\mathcal{G}}^{(2)}|,$$

$$F=\frac{f_{\Vert }}{f_{\Vert }+f_{\perp }},$$

Full text

*Introduction.—*Quantum teleportation is considered as one of the major protocols in quantum information science. By exploiting the physical resource of entanglement, quantum teleportation has played a prominent role in the development of quantum information theory Bennett et al. (1993); Nielsen and Chuang (2000); Kim et al. (2001); Yeo and Chua (2006); Liuzzo-Scorpo et al. (2017) and represents a fundamental ingredient to the progress of many quantum technologies such as quantum gate teleportation Gottesman and Chuang (1999), quantum repeaters Briegel et al. (1998); Sangouard et al. (2011), measurement-based quantum computing Raussendorf and Briegel (2001), port-based teleportation Ishizaka and Hiroshima (2008) and quantum network teleportation (QTN) Landry et al. (2007); Valivarthi et al. (2016); Sun et al. (2016). Teleportation has also been used as a quantum simulator for ’extreme’ phenomena, such as closed timelike curves and the grandfather paradox Lloyd et al. (2011).

Quantum teleportation, first proposed by Bennett et al. Bennett et al. (1993), is a scheme of quantum information processing which allows the transfer of a quantum state between remote physical systems without physical transfer of the information carrier. Specifically, an unknown quantum state of a physical system is measured and subsequently reconstructed at a remote location through the use of classical communication and quantum entanglement Massar and Popescu (1993); Popescu (1994). Without entanglement, such quantum state transfer would not be possible within the laws of quantum mechanics. For that reason, quantum teleportation is thought of as the quantum information protocol which clearly demonstrates the character of quantum entanglement as a resource.

To date, quantum teleportation has been achieved and studied in many different systems, including photonic systems, nuclear mag­netic resonance, optical modes, trapped atoms and solid-state systems (see Pirandola et al. (2015) and references therein). Naturally, most attention has been focused on teleporting the state on long-distance Yin et al. (2012); Ma et al. (2012) with the recent satellite-based implementations Ren et al. (2017). However, even though quantum teleportation is a typically bipartite process, it can be extended to multipartite quantum protocols which have not been thoroughly studied yet. Such multipartite protocols are expected to form fundamental components for larger-scale quantum communication and computation Nielsen and Chuang (2000).

An important extension of quantum teleportation to a multipartite case is known as the controlled quantum teleportation (CQT) Karlsson and Bourennane (1998) which allows for remote quantum nondemolition (QND) measurements and forms a backbone of QTN van Loock and Braunstein (2000); Yonezawa et al. (2004); Pirandola et al. (2015). In the simplest case of tripartite systems, the essential concept of the CQT scheme is that the transfer of the quantum state from sender (Alice) to receiver (Bob) needs controller’s (Charlie’s) classical information and thus, Charlie can determine success or failure of teleportation by restricting the access to his information, what is commonly thought of as a clear manifestation of tripartite entanglement Yonezawa et al. (2004). When Alice, Bob and Charlie can choose any one of them to be the sender, receiver and controller, then the CQT protocol is equivalent to QTN, a prelude for a genuine quantum internet Castelvecchi (2018). Here, it is also believed that parties must share a multipartite entangled state to allow teleportation between any two parties Yonezawa et al. (2004); Pirandola et al. (2015). Furthermore, the CQT protocol as discussed in this paper, may be applied in the processing of quantum secret sharing, a prominent quantum-information protocol Hillery et al. (1999).

Although, several implementation schemes of CQT have been proposed over time using, for instance, a Brown state via cavity QED Wang et al. (2009), quantum dots Wang et al. (2013); Heo et al. (2017), GHZ-like states Naseri et al. (2015), a GHZ state in an ion-trapped system Xu and Wang (2016), so far to the best of our knowledge, the successful experimental realization of the CQT protocol has been reported only for GHZ state of continuous variables Yonezawa et al. (2004). For such system the GHZ teleportation channel can be contracted, for instance, using three vacuum states in the limit of infinite squeezing Adesso et al. (2006). Naturally, in a real experiment, a maximally entangled GHZ state of continuous variable is not available because of finite squeezing and inherent losses. Therefore, the realistic state generated by three highly squeezed vacuum states is the non-maximally entangled GHZ-like state. Consequently, CQT of a coherent state was perform with fidelity up to $F_{CQT}=64\%\pm 2\%$ Yonezawa et al. (2004).

To overcome the limitation caused by finite squeezing, in this paper we present the first experimental verification of CQT on GHZ states of discrete variables. Using the four-photon source based on the process of spontaneous parametric down-conversion (SPDC), we generate GHZ state and perform the CQT with fidelity of $F=83.0\%\pm 7.3\%$. Our experiment is also successfully repeated for other teleportation channels based on the GHZ states, in particular, a statistical mixture of such states, demonstrating the controller’s capability of steering the teleportation process based on the classical correlations without presence of multipartite entanglement. Such result represents a universal feature of CQT and QTN which is deeply rooted in the operational definition of bipartite entanglement Barasiński et al. (2018).

*The concept of quantum controlled teleportation.—*We start by reviewing the basic tripartite CQT protocol in finite-dimensional settings Karlsson and Bourennane (1998).

The protocol considers three remote parties – Alice, Bob and Charlie – who share pure three-qubit entangled state in advance. In the perfect scheme, the shared entangled state is taken to be a maximally entangled GHZ state, $|\mathcal{G}^{(1)}\rangle=\frac{1}{\sqrt{2}}\{|H_{1}H_{2}H_{3}\rangle+|V_{1}V_{2}V_{3}\rangle\}$, where we use the polarization degree of freedom of the photons generated in optical setup with $|H\rangle$ and $|V\rangle$ denoting the horizontal and vertical polarization states, respectively. Initially, Alice is in possession of a qubit in mode no. $1$ of the GHZ state and a single qubit in mode $4$ in the input quantum state $|\psi_{4}\rangle$ which she wants to teleport. In our experiment, the input state is the polarization of an arbitrary single photon: $|\psi_{4}\rangle=\alpha|H_{4}\rangle+\beta|V_{4}\rangle$, where $|\alpha|^{2}+|\beta|^{2}=1$. Suppose now that Alice applies a specific joint quantum measure­ment which projects photons in modes $1$ and $4$ into the maximally entangled Bell state $|\psi^{-}_{14}\rangle=\frac{1}{\sqrt{2}}\{|H_{1}\rangle|V_{4}\rangle-|V_{1}\rangle|H_{4}\rangle\}$. As a result, the state of remaining two qubits is simultaneously projected into $|\psi_{23}\rangle=\alpha|H_{2}\rangle|V_{3}\rangle-\beta|V_{2}\rangle|H_{3}\rangle$ which can be further decomposed in the new basis $\mathcal{B}_{\pm}=\{|+\rangle,|-\rangle\}$ as $|\psi_{23}\rangle=\frac{1}{\sqrt{2}}\{\alpha|H_{2}\rangle-\beta|V_{2}\rangle\}|+_{3}\rangle-\frac{1}{\sqrt{2}}\{\alpha|H_{2}\rangle+\beta|V_{2}\rangle\}|-_{3}\rangle$, where $|\pm_{3}\rangle=\frac{1}{\sqrt{2}}\{|H_{3}\rangle\pm|V_{3}\rangle\}$. In the next step, Charlie (the controller) applies a von Neumann measurements on qubit in mode $3$ in the basis $\mathcal{B}_{\pm}$. Consequently, the final state of qubit in mode $2$, kept by Bob, is equal to $|\psi_{2}\rangle=\alpha|H_{2}\rangle+\beta|V_{2}\rangle$ up to a unitary operation that depends on the outcomes of Charlie’s measurements. In contrary, if Charlie decides to apply a von Neumann measurements on mode $3$ in the basis $\mathcal{B}_{HV}=\{|H\rangle,|V\rangle\}$ then the resulting quantum state of Bob’s qubit is either $|\psi_{2}\rangle=|H_{2}\rangle$ or $|\psi_{2}\rangle=|V_{2}\rangle$. These two scenarios clearly show Charlie’s power to determine success and failure of CQT.

Now, it is important to note that the above-mentioned $|\psi^{-}_{14}\rangle$ state is only one of four possible Bell states which can be obtained by Alice. In general, the composed state of qubits in modes $1$ and $4$ can be projected into four different states $(\mathcal{P}_{k}\otimes I)|\psi^{-}_{14}\rangle$, where $\mathcal{P}_{k}$ is an appropriate Pauli operator Nielsen and Chuang (2000) and $k=0,1,2,3$. When this happens, the state of particles in modes $2$ and $3$ becomes $\rho_{23}=\mathcal{P}^{\dagger}_{k}|\psi_{23}\rangle\langle\psi_{23}|\mathcal{P}_{k}$. Bob can then recover the input state by applying an accordingly chosen transformation that requires a classical communication with both Alice and Charlie. Although in the above-mentioned scheme only one of four Bell states is distinguished, teleportation is still successfully achieved, albeit only in a quarter of the cases. Moreover, it should be noted that the complete Bell state measurement which is based on nonlinear processes, requires hyperentanglement or feed-forward techniques Knill et al. (2001) and hence, it remains an experimentally challenging problem which usually causes the reduction of the signal intensity Kim et al. (2001); Grice (2011). Therefore, the antisymmetric structure of the state $|\psi^{-}_{14}\rangle$ makes this state the most useful in the experimental implementation of teleportation protocols as discussed in Bouwmeester et al. (1997); Kim et al. (2001). In this paper, we also take the advantage of this property and limit our Bell state measurement only to $|\psi^{-}_{14}\rangle$.

Finally, we note that the faithfulness of the CQT protocol shall not change if one applies a local bit flip operation on the GHZ state shared by Alice, Bob and Charlie, say $|\mathcal{G}^{(2)}\rangle=\frac{1}{\sqrt{2}}\{|H_{1}H_{2}V_{3}\rangle+|V_{1}V_{2}H_{3}\rangle\}$. A particularly interesting scenario, however, occurs if one takes a statistical mixture of such two GHZ states,

[TABLE]

where $0\leq p\leq 1$. Then, for the equivalently balanced probabilities the state $\rho(p=1/2)$ belongs to the biseparable class and can be decomposed as $\rho(p=1/2)=\frac{1}{2}\{|\chi^{+}\rangle\langle\chi^{+}|+|\chi^{-}\rangle\langle\chi^{-}|\}$, where $|\chi^{\pm}\rangle=\frac{1}{2}\{|H_{1}H_{2}\rangle\pm|V_{1}V_{2}\rangle\}\otimes\{\pm|H_{3}\rangle+|V_{3}\rangle\}$. This means that there are no other correlations between Charlie and the rest of the system besides the classical ones. Despite that Charlie’s capability of controlling the teleportation protocol remains unchanged Barasiński et al. (2018).

To emphasize the significant role of tripartite (fully-entangled and biseparable) states in the CQT protocol, let us discuss the difference between CQT and the classical control of ordinary teleportation. Suppose that Alice and Bob share either of two Bell states $|\phi^{+}_{12}\rangle=\frac{1}{\sqrt{2}}\{|H_{1}H_{2}\rangle+|V_{1}V_{2}\rangle\}$ or $|\phi^{-}_{12}\rangle=\frac{1}{\sqrt{2}}\{|H_{1}H_{2}\rangle-|V_{1}V_{2}\rangle\}$ with equal probability and the information which Bell state is truly shared belongs only to Charlie. Then, the teleportation between Alice and Bob is successfully performed when Charlie broadcasts the information he has and is forbidden otherwise. Analogously, for the quantum protocol the GHZ state can be written in the form $|\mathcal{G}^{(1)}\rangle=\frac{1}{\sqrt{2}}(|\phi^{+}_{12}\rangle|+_{3}\rangle+|\phi^{-}_{12}\rangle|-_{3}\rangle)$, where the information “which Bell state” is encoded in the Charlie’s qubit via an QND-type interaction, i.e. in the basis of $|+_{3}\rangle$ and $|-_{3}\rangle$. However, in this case the information is quantum-mechanically possessed by Charlie and hence, any measurement in the logical basis implies that the teleportation is forbidden principially and not just by Bob’s ignorance of Charlie’s outcome (e.g. it can not be restored by any eyevesdropping). This is the main difference with the classical counterpart, valid even for the biseparable mixture $\rho(p=1/2)$. Note that measurement in any other basis than $|0/1\rangle$ allows to restore the teleportation at least probabilistically by implementing an appropriate filtering.

*Experimental implementation.—*The experimental setup consists of a four-photon source, GHZ preparation stage and three stations operated by Alice, Bob and Charlie (see Fig. 1). Photons are generated in a BBO crystal cascade Kwiat et al. (1999) by means of spontaneous parametric down-conversion pumped by femto-second pulses at $413$ nm. First pair of photons (modes $1$ and $2$) is generated while the pumping pulse propagates through the crystals in the forward direction. Subsequently it gets reflected on a mirror and generates a second pair of photons (modes $3$ and $4$) on its way back. Pump beam polarization is controlled by a half-wave plate (HWP) and a polarization dispersion line (BDA) to correct for polarization group velocity dispersion Nambu et al. (2002). Proper polarization of the pumping beam allows to generate photons in modes $1$ and $2$ in an entangled state $|\Phi^{+}_{12}\rangle=\frac{1}{\sqrt{2}}\{|H_{1}H_{2}\rangle-i|V_{1}V_{2}\rangle\}$. Photons $3$ and $4$ are collected only from one of the crystals obtaining thus a separable state $|H_{3}H_{4}\rangle$. After being subjected to procedures described below, the photons are collected to single-mode optical fibers and detected by a set of four avalanche photodetectors. Simultaneous four-fold coincidence detections are recorded.

In the next step, we generate the GHZ state $|\mathcal{G}^{(1)}\rangle$ in Eq. (1). To achieve that, polarization of the photon in mode $3$ is changed to circular $|R_{3}\rangle=\frac{1}{\sqrt{2}}\{|H_{3}\rangle+i|V_{3}\rangle\}$ and then it overlaps with photon in mode $2$ on the polarizing beamsplitter $PBS$. With success probability of $\frac{1}{2}$, these two photons leave $PBS$ by different output ports and together with the photon in mode $1$ form the GHZ state $|\mathcal{G}^{(1)}\rangle$ Lu et al. (2008). Although we have not repeated the GHZ state preparation testing in the same way as the authors of Ref. Bussières et al. (2014), we have performed testing of individual component blocks of our setup, namely we have observed purities of about $90\%$ of the generated Bell state (modes $1$ and $2$) together with similarly pure Bell state preparation with photons in mode $2$ and $3$ on the $PBS$ (note: this was tested when projecting mode $1$ onto the $|H\rangle$ state). We label the photon modes leading to Bob’s and Charlie’s apparatus by nos. $2$ and $3$ respectively.

Alice subsequently encodes the to-be-teleported qubit into the polarization state of the photon in mode $4$ using a HWP and QWP. Then she projects the state of photons in modes $1$ and $4$ onto a singlet Bell state $|\psi^{-}_{14}\rangle$ by post-selecting on photon anti-bunching behind a balanced fiber beamsplitter.

At this point, Charlie decides whether to allow or deny the teleportation. In order to allow it, Charlie projects the state of the photon in mode $3$ to circular polarization. Similarly, to deny the teleportation, Charlie projects his photon onto horizontal polarization. Due to the nature of coincidence-based measurement, Alice’s and Charlie’s actions happen simultaneously.

Bob receives the teleported qubit encoded in the state of the photon in mode $2$. He then subjects this photon to a polarization projection measurement using a sequence of HWP, QWP and a polarizer. To evaluate the performance of teleportation, we measure the fidelity of the tele­ported state $F=\langle\psi_{4}|\rho_{2}|\psi_{4}\rangle$, where $\rho_{2}$ is the resulting state of photon in mode $2$ (mixed in general). Based on the coincidence counts observed for different combinations of input states encoded by Alice and Bob’s projection measurement, $F$ is calculated as Boschi et al. (1998)

[TABLE]

where $f_{\|}$ stands for coincidence rate observed when Bob projects on the state identical to Alice’s encoding choice. Likewise, $f_{\perp}$ stands for coincidence rate observed when Bob projects on an orthogonal state.

*Experimental results.—*We test the CQT protocol on a linearly polarized balanced ($\alpha=\beta$) input state i.e. $|\psi_{4}\rangle=\frac{1}{\sqrt{2}}\{|H_{4}\rangle+|V_{4}\rangle\}$. This choice of $|\psi_{4}\rangle$ is quite natural i.e. by the very description of the GHZ state preparation as both $|H\rangle$ and $|V\rangle$ polarization can be considered as preferred directions in the experiment. Therefore, the input state polarized at $45^{\circ}$ represents one of the most challenging tasks and other commonly analyzed states yield the teleportation fidelity approximately equal or greater Bouwmeester et al. (1997); Bussières et al. (2014); Pfaff et al. (2014); Ren et al. (2017).

For the purposes of preliminary testing, in the first experiment we have operated our setup in the regime of ordinary uncontrolled quantum teleportation Bennett et al. (1993). To achieve that, polarization of the third photon is kept horizontal to be directly transmitted on $PBS$ on to Charlie’s detector. As for Bob’s projection measurement, it is performed using the combination of $HWP$ and $PBS$. Again, four-fold coincidences are registered, this time with photon in mode 3 serving only as a trigger. Based on Eq. (1) the faithfulness of uncontrolled quantum teleportation has been found to be $F=83.1\%\pm 4.9\%$ which is in line with recent experiments on photonic qubits (e.g. Bussières et al. (2014); Ren et al. (2017)).

Probabilistic nature of our four-photon source causes undesired higher-order SPDC terms to contribute to the detected signal. Presented fidelities therefore need to be corrected for these imperfections of the source to be faithful characteristics of the protocol implementation itself. A detailed analysis of these corrections is presented in the Supplementary material supplement .

In the second experiment, we have performed the teleportation on the GHZ state. In order to do this, we set back the polarization of the photon in mode $3$ to circular, $|R_{3}\rangle$, thus generating the $|\mathcal{G}^{(1)}\rangle$ channel. By proper operating of the QWP and the polarizer we analyze two scenarios of teleportation. In the first one, when Charlie allows for the teleportation the fidelity calculated from the four-fold coincidences $F_{\textrm{allowed}}=77.9\%\pm 8.1\%$. This result exceeds the classical limit of $66.7\%$ and thus certifies the quantum nature of our teleportation experiment. In the second scenario, i.e. without Charlie’s permission, the fidelity of $F_{\textrm{denied}}=57.2\pm 5.0\%$ meets the second condition of CQT. This results clearly shows the success of CQT through the GHZ state of discrete variables. Similar measurements have been performed for $|\mathcal{G}^{(2)}\rangle$ channel. This kind of GHZ state can be prepared by slight modification of the experimental setup. Specifically, both HWPs in the mode 2 and the Bob’s analyzer part, are rotated by $\pi/4$. The corresponding fidelities are presented in Table 1 and visualized in Fig. 2. What is important, in this configuration the CQT is realized with even greater fidelity of around $83\%$ with simultaneous decrease of $F_{\textrm{denied}}$.

Now we perform the CQT through the statistical mixture $\rho(p)$ given in Eq. (1) when $p=1/2$. To emulate this, we have simply summed up the respective coincidence counts obtained for $|\mathcal{G}^{(1)}\rangle$ and $|\mathcal{G}^{(2)}\rangle$. We found the resulting fidelity to be $F_{\textrm{allowed}}=80.2\%\pm 5.7\%$ when Charlie permits the teleportation and $F_{\textrm{denied}}=55.1\%\pm 5.0\%$ otherwise. This means that both conditions of CQT are satisfied also for $\rho(p=1/2)$ channel, despite it belongs to the biseparable class and there is no entanglement between Charlie’s photon and the remaining two photons. In order to verify this fact experimentally one can use the well know methods such as $\theta-$protocol McCutcheon et al. (2016) or $XY-$protocol Pappa et al. (2012). In fact, both protocols have been successfully applied recently for experimental detection of tripartite entanglement in the GHZ states McCutcheon et al. (2016). Since we use the entangled-photon source with the same efficiency as in Ref. McCutcheon et al. (2016) the outcome of multipartite-entanglement-detection protocol is similar and it is out of the scope of this paper. To the best of our knowledge, this is the first demonstration of CQT based on the biseparable states.

*Conclusions.—*In summary, we have presented a proof-of-principle experimental demonstration of CQT through various kinds of GHZ states of discrete variables with the fidelities well above the classical limit. Our experiment shows that tripartite entanglement is not a necessary recourse for CQT. In fact, the classical correlation between controller and joined "sender-receiver" subsystem is sufficient in order to allow or forbid the teleportation. In a broader context, our results open new possible ways of implementation of CQT lowering requirements for a state preparation and preservation what is of practical importance in realizing more complicated quantum computation and quantum communications among many parties. In particular, one can consider the three-qubit Werner state ($\rho_{W}=q|\mathcal{G}^{(1)}\rangle\langle\mathcal{G}^{(1)}|+(1-q)I/8$) which can be thought of as an imperfect preparation of the GHZ quantum channel. The ability to perform CQT through biseparable Werner states (i.e. for $1/3<q\leq 3/7$) implies that the CQT is less fragile against noise than the tripartite entanglement, as described in Barasiński et al. (2018). Furthermore, as the three-qubit Werner states are invariant under qubits permutation, the CQT can be successfully performed no matter how we split the qubits between Alice, Bob and Charlie. In our experiment, despite the probabilistic nature of the GHZ state preparation and the teleportation itself, the roles of Alice, Bob and Charlie can also be swapped (see Supplementary material for detailed analysis supplement ). In other words, fundamentals of our experiment can be easily used in demonstration of a QTN for biseparable states. This conclusion is in contrast with common opinion "Only if we use a fully inseparable tripartite entangled state can we succeed in teleportation between an arbitrary pair in the network" Yonezawa et al. (2004). Explanation of this phenomenon is based on the concept of localizable entanglement Verstraete et al. (2004) which plays a central role in CQT and QTN Barasiński et al. (2018). Our experiment shows non-trivial application of localizable entanglement leading to results which cannot be predicted by standard quantifiers of multipartite entanglement.