Entanglement Swapping in Quantum Networks

• Entanglement swapping is a quantum protocol that generates entanglement between initially uncorrelated particles via joint Bell-state measurements on independent entangled pairs.
• It employs measurement in the Bell basis with quantification metrics such as concurrence, negativity, and I-concurrence, supporting experiments in photonic, superconducting, and continuous-variable systems.
• This technique is foundational for practical quantum applications including quantum repeaters, device-independent cryptography, and scalable quantum information networks.

Entanglement swapping is a quantum protocol enabling two particles, which have not directly interacted, to become entangled via operations on other particles with which each is individually entangled. This process is realized by performing a joint measurement, typically in the Bell basis, on two partner particles from initially independent entangled pairs. The protocol is foundational for quantum networking, quantum repeaters, and nonlocality tests, supporting the distribution of entanglement across diverse quantum architectures. Its implementations span discrete-variable, continuous-variable, hybrid, and high-dimensional systems, with direct experimental relevance for quantum communication, device-independent cryptography, and quantum information processing.

1. Protocols and Mathematical Foundations

The standard entanglement swapping protocol begins with two independent bipartite entangled pairs, labeled (1,2) and (3,4) for qubits or more generally for qudits. The initial state for maximally entangled qubits is $|\Phi^+\rangle_{12} \otimes |\Phi^+\rangle_{34}$, with $$|\Phi^+\rangle = (|00\rangle + |11\rangle)/\sqrt{2}$$. Performing a Bell-state measurement (BSM) on particles 2 and 3 projects the remaining pair (1,4) into the same Bell state as the measurement outcome (Zangi et al., 1 Aug 2025, Zangi et al., 2022). The protocol generalizes to higher dimensions, where the measurement is implemented in a maximally entangled basis of $B \otimes C$:

$$|\Psi_{u v}\rangle_{BC} = \frac{1}{\sqrt{d}} \sum_{\ell=0}^{d-1} e^{2\pi i u \ell / d} |\ell\rangle_B \otimes |\ell \oplus v\rangle_C$$

and the conditional post-measurement state of $A$ and $D$ is given by tracing out the measured subsystems and normalizing the resulting outcome (Zangi et al., 1 Aug 2025).

The quantification of swapped entanglement employs concurrence $C(\rho)$, negativity $\mathcal{N}(\rho)$, and, for high-dimensional pure states, I-concurrence:

$$C_I(|\psi \rangle) = \sqrt{2 (1 - \textrm{Tr} \rho_A^2)}$$

For noisy (mixed) inputs, the product of the initial pairs' concurrence or negativity provides an upper bound to the average swapped entanglement, and nonmaximally entangled measurements degrade the output entanglement (Zangi et al., 2022).

2. Experimental Realizations and Architectures

Photonic Implementations

Time-bin and polarization entanglement swapping has been demonstrated with fully independent sources over metropolitan-scale fiber networks, employing Hong–Ou–Mandel interference for Bell-state measurement and achieving fidelities sufficient to violate Bell inequalities (0809.3991, Sun et al., 2016). On-demand quantum dot sources now support all-photonic swapping protocols with brightness and purity compatible with atomic memories, with observed swapped-pair fidelities up to $0.81 \pm 0.04$ (Zopf et al., 2019, Beccaceci et al., 11 Dec 2025). Deterministic, remote quantum dot emitters, enabled via piezoelectric spectral tuning, have achieved swapping between independent sources with fidelities well above the classical threshold (Beccaceci et al., 11 Dec 2025).

Superconducting Circuits

Deterministic entanglement swapping in superconducting transmon arrays leverages high-fidelity Bell-state gates and multiplexed single-shot readout. Experiments yield average swapped-state concurrence above 0.75, demonstrating outcome-independent entanglement transfer, and support delayed-choice protocols wherein the choice of measurement on intermediary qubits retroactively determines whether distant qubits were entangled or separable (Ning et al., 2019).

Continuous Variables and Hybrid Systems

Swapping in continuous-variable (CV) systems relies on two-mode squeezed vacua and homodyne detection; optimal Gaussian entanglement swapping with specific gain settings preserves purity and maximizes output entanglement (Hoelscher-Obermaier et al., 2010). Hybrid DV–CV swapping protocols have robustly transferred discrete-mode entanglement via efficient CV resources, quantifying output entanglement with logarithmic negativity and demonstrating post-selection–distilled Bell-state violations (Takeda et al., 2014).

Multipartite and Hyperentangled Architectures

Swapping has been extended to multipartite entangled states—merging independent GHZ or EPR-type CV ensembles into larger entangled networks through joint measurement and classical feedforward (Su et al., 2016). Hyperentangled swapping protocols with two-level neutral atoms exploit simultaneous internal and external degrees of freedom, implemented via Bragg diffraction and resonant cavity interactions, achieving deterministic, decoherence-resistant operations with net fidelities exceeding 0.99 (Hasan et al., 2024).

3. Optimality, Deterministic Measurements, and Thresholds

Traditional entanglement swapping is probabilistic: for arbitrary inputs, only specific measurement outcomes produce entangled output states, with the remaining branches requiring post-selection or corrective unitaries. Recent work fully characterizes deterministic swapping measurements—those under which every outcome gives local-unitarily equivalent, optimally entangled states (Alimuddin et al., 13 Jan 2026):

• For arbitrary pure inputs, measurements built from a single phase-conjugation class of unbiased operators (complex Hadamard matrices) yield universal outcome-independent entanglement. For $d=2,3$, there exists a unique class; for $d=4$ an infinite family, and for $d=5$ exactly 72 inequivalent classes.
• Deterministic optimal protocols maximize G-concurrence and eliminate LOCC irreversibility of post-selection.
• In 2- and 3-dimensional repeater networks, the order in which swap nodes perform measurements does not affect the final entanglement spectrum.

For non-maximal or mixed input states, threshold values exist: only above specific input concurrence or negativity does the swapping protocol produce output entanglement above the separability bound; the probability of obtaining high-entanglement outcomes is smaller than for lower entanglement (Muñoz et al., 2013).

4. Entanglement Swapping in Complex and High-Dimensional Systems

High-Dimensional and Noisy Scenarios

Swapping extends to qudits ($d>2$), where generalized Bell measurements produce maximally entangled qudit pairs in the output. For uniform maximal input entanglement, output I-concurrence scales as $\sqrt{2(d-1)/d}$, unbounded as $d$ increases, while negativity saturates at unity (Zangi et al., 1 Aug 2025). In noisy scenarios (isotropic channel or depolarizing models), swapped entanglement persists at lower fidelity thresholds as $d$ increases, supporting enhanced robustness for networked quantum communication.

Frequency, Time, and Multipartite Modes

Frequency-resolved swapping with multimode photonic sources allows multiplexed heralding of multiple orthogonal Bell pairs, with the Schmidt number $K$ providing an upper bound on simultaneous distinguishable entanglement links (Merkouche et al., 2021). Swapping between multipartite states deterministically merges independent tripartite GHZ or EPR resources into fused, larger CV networks, quantified via van Loock–Furusawa inequalities and positive partial transpose (PPT) criteria (Su et al., 2016).

Critical Quantum Spin Chains

In many-body physics, Bell-state measurements performed across pairs in critical XXZ spin chains induce swapped entanglement between macroscopic subsystems. The entanglement entropy of the swapped regions scales logarithmically with subsystem size, with a universal coefficient determined by boundary CFT (e.g., $c_{\textrm{eff}} = 1/2$ for XXZ chains) (Hoshino et al., 2024). These behaviors are quantitatively replicated in tensor-network simulations and are accessible experimentally in Rydberg atom arrays.

5. Retrocausality, Causality, and Interpretive Frameworks

The Transactional Interpretation (TI) models quantum processes as atemporal, nonlocal “handshakes” between offer and confirmation waves propagating both forward and backward in time. In entanglement swapping, all detectors—Bell-state analyzers and local spin measurements—participate in the global transaction. The apparent retroactive correlation is understood as the block-universe realization of the transaction, with no dynamical collapse, no retrocausal signaling, and the outcome being determined by the completed absorber configuration. This reframes paradoxes of delayed-choice and outcome dependence into consistent, nonlocal causality (Marchildon, 2013).

6. Applications and Quantum Networking Relevance

Entanglement swapping is pivotal for quantum communication primitives, serving as the backbone for quantum repeater chains, network topologies, and device-independent cryptographic protocols. Swapping supports teleportation of entanglement, entanglement-based QKD immune to source and detector attacks, and enables quantum information transfer across distributed systems including high-dimensional, multipartite, CV, and hybrid architectures (Davis et al., 24 Mar 2025, Zangi et al., 2022). Performance metrics such as swapped-state fidelity, concurrence, secret key rate, and robustness to noise are crucial for evaluating and scaling practical quantum networks.

7. Fundamental Limits and Future Directions

Theoretical results indicate that, for Gaussian states, optimal swaps do not enhance effective transmission beyond direct links; entanglement distillation steps remain necessary for repeater chains (Hoelscher-Obermaier et al., 2010). High-dimensional swapping offers advantages in noise tolerance and channel capacity, suggesting new pathways for long-distance, multiplexed quantum networks. Multipartite and hyperentangled swapping protocols are progressing toward scalable, loss-resistant quantum architectures, with deterministic outcome-independent measurements poised to overcome probabilistic inefficiencies (Su et al., 2016, Hasan et al., 2024, Alimuddin et al., 13 Jan 2026).

Experimental and theoretical advances continue to push entanglement swapping into regimes of higher fidelity, greater scalability, improved resource efficiency, and broad applicability across both discrete and continuous quantum information science domains.