Binary Tree Repeater Networks

• Binary tree repeater networks are loopless, hierarchical quantum communication structures that employ unique-path entanglement swapping for state transfer between nonadjacent nodes.
• They use Werner-state links and closed-form fidelity expressions, averaging maximum teleportation fidelities across unique paths to precisely evaluate network performance.
• Performance critically depends on branching symmetry and directionality, with the directed symmetric binary tree achieving lower quantum advantage thresholds and higher resource efficiency.

Binary tree repeater networks are loopless, hierarchical quantum communication networks in which repeater nodes are arranged in a binary branching structure. In the recent arXiv literature, their principal operational role is the distribution of entanglement and the support of teleportation or state transfer between nonadjacent nodes. The dominant analytical framework evaluates such networks by the average of the maximum teleportation fidelities between all source–target pairs, $F^{\rm max}_{\rm avg}$, rather than by local link quality alone (Mylavarapu et al., 2024, Roy et al., 14 Aug 2025). Within the broader class of tree topologies, binary trees are intermediate between chains and stars: they retain the tractability of unique-path routing, shorten typical path lengths relative to chains, and yet remain degree-bounded, which prevents the star-like asymptotics that are optimal among the loopless topologies studied so far (Mylavarapu et al., 2024).

1. Operational definition and fidelity criterion

The central resource-theoretic question for a binary tree repeater network is not whether each edge is entangled, but whether the network as a whole enables high-fidelity teleportation between arbitrary node pairs. For a two-qubit resource state $\rho$, the optimal teleportation fidelity recalled in the tree-network analysis is

$F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$

with $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$, where $T$ is the correlation matrix of $\rho$ (Mylavarapu et al., 2024). The operational benchmark is $2/3$: $F^{\rm max}>2/3$ is the condition for genuine quantum advantage over classical teleportation strategies.

In the binary-tree repeater setting of Werner-state links, each edge is modeled as

$$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$

with $p\in[0,1]$. For a single Werner pair, the teleportation fidelity is

$\rho$0

which exceeds the classical limit when $\rho$1 (Roy et al., 14 Aug 2025). For a path $\rho$2 connecting source $\rho$3 to target $\rho$4, entanglement swapping multiplies the Werner parameters along the path, yielding

$\rho$5

When all links have the same parameter $\rho$6 and the path length is $\rho$7, this reduces to

$\rho$8

Because a binary tree is loopless, every node pair is connected by a unique path. Accordingly,

$\rho$9

so the global resource measure can be computed as an average over paths rather than over competing routes (Mylavarapu et al., 2024, Roy et al., 14 Aug 2025). This unique-path property is not a minor technical convenience; it is the structural reason that binary trees admit closed-form fidelity expressions.

2. Binary-tree topologies within the landscape of tree repeater networks

The general tree-network analysis identifies the star as the best and the chain as the worst among the loopless topologies explicitly studied, with “intermediate flowers” interpolating between them (Mylavarapu et al., 2024). In that language, a binary tree is another intermediate tree-like network whose degrees are constrained to at most $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$0. This places binary trees in a regime where branching materially improves performance over chain-like structures, while finite degree still limits the attainable average fidelity relative to a hub-dominated star.

The dedicated binary-tree study classifies four repeater topologies by two binary attributes—directionality and symmetry (Roy et al., 14 Aug 2025).

Topology	Structural characterization	Abbreviation
Directed asymmetric binary tree	Directed, more chain-like, fewer branching possibilities	DABT
Directed symmetric binary tree	Directed, every internal node branches into two children at each level	DSBT
Undirected asymmetric binary tree	Undirected, more chain-like, fewer branching possibilities	UABT
Undirected symmetric binary tree	Undirected, every internal node branches into two children at each level	USBT

In this classification, asymmetric trees are closer to chains, while symmetric trees grow exponentially in the number of nodes at each depth. Directed trees constrain traversable paths by edge orientation; undirected trees permit traversal in both directions. The key point is not merely topological labeling but the induced path-length distribution: symmetry changes how many paths exist at each depth, and directionality changes which paths contribute to the average (Roy et al., 14 Aug 2025).

A common simplification is to assume that undirected trees must dominate directed ones because they allow more routing options. The reported analysis rejects that simplification: more admissible paths do not automatically improve $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$1, because longer, weaker paths also enter the average (Roy et al., 14 Aug 2025).

3. Analytical fidelity structure for binary trees

For the binary-tree families with uniform link parameter $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$2, closed-form expressions for the average teleportation fidelity are derived for all four topologies (Roy et al., 14 Aug 2025). The DSBT formula is especially transparent: $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$3 This representation makes explicit that the network fidelity is a weighted average over path lengths. The weights encode topology; the factors $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$4 encode multiplicative Werner degradation with distance.

The same logic governs the other binary trees. For DABT and UABT, the closed forms show corrections to the baseline $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$5 that scale as $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$6 in the large-network limit. For DSBT, the decay is slower,

$F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$7

while for USBT the asymptotic correction is regime-dependent: $F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$8

These formulae formalize a general principle already visible in the broader tree study: teleportation fidelity in loopless repeater networks is governed by multiplicative decay along unique paths, so path-length statistics are the decisive topological variable (Mylavarapu et al., 2024). In binary trees, the distinction between symmetric and asymmetric growth controls those statistics directly.

4. Quantum advantage, thresholds, and topology ranking

The network-wide quantum-advantage condition is

$F_\rho^{\rm max}=\frac{1+\mathcal N(\rho)/3}{2},$9

The binary-tree analysis shows that local link usefulness is not sufficient for this global criterion. Even though a single Werner link is useful when $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$0, an entire binary-tree network may still fail to exceed $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$1 on average (Roy et al., 14 Aug 2025).

For $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$2, the reported thresholds are approximately $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$3 for DABT, $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$4 for DSBT, $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$5 for UABT, and $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$6 for USBT. For $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$7, they rise to approximately $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$8 for DABT, $$\mathcal N(\rho)=\mathrm{Tr}\!\left(\sqrt{T^\dagger T}\right)$$9 for DSBT, $T$0 for UABT, and $T$1 for USBT (Roy et al., 14 Aug 2025).

Topology	Threshold $T$2 at $T$3	Threshold $T$4 at $T$5
DABT	$T$6	$T$7
DSBT	$T$8	$T$9
UABT	$\rho$0	$\rho$1
USBT	$\rho$2	$\rho$3

At the single-link teleportation boundary $\rho$4, the network averages remain below $\rho$5 in all four cases. For $\rho$6, the reported values are $\rho$7 for DABT, $\rho$8 for DSBT, $\rho$9 for UABT, and $2/3$0 for USBT; for $2/3$1, they are $2/3$2, $2/3$3, $2/3$4, and $2/3$5, respectively (Roy et al., 14 Aug 2025). This is one of the clearest corrections to a common misconception: edge-wise quantum usefulness does not imply network-wide quantum usefulness.

Among the four binary-tree topologies, the directed symmetric binary tree is reported as the most beneficial. The study attributes this to the combination of high branching structure and directionality, which reduces the inclusion of long, weak paths in the averaging process. The supporting path-length data are explicit: for $2/3$6, the average path lengths are $2/3$7 for DABT, $2/3$8 for DSBT, $2/3$9 for UABT, and $F^{\rm max}>2/3$0 for USBT; for $F^{\rm max}>2/3$1, they are $F^{\rm max}>2/3$2, $F^{\rm max}>2/3$3, $F^{\rm max}>2/3$4, and $F^{\rm max}>2/3$5, respectively (Roy et al., 14 Aug 2025). In the reported comparisons, DSBT therefore couples the shortest average path length with the lowest threshold for quantum advantage.

5. Maximally entangled links, disorder, and asymptotic limits

The broader tree-network treatment distinguishes three scenarios for link quality: uniform noisy links, mixed perfect and noisy links, and random links (Mylavarapu et al., 2024). In the mixed case, a fraction $F^{\rm max}>2/3$6 of the links are maximally entangled, and perfect links act as zero-length segments in effective path counting. The binary-tree study uses the same idea to examine how a small set of maximally entangled links can restore or enlarge the quantum-advantage regime (Roy et al., 14 Aug 2025).

For $F^{\rm max}>2/3$7 at $F^{\rm max}>2/3$8, the number of maximally entangled links needed to reach quantum advantage is reported as $F^{\rm max}>2/3$9 for DSBT, $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$0 for DABT, $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$1 for UABT, and $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$2 for USBT (Roy et al., 14 Aug 2025). The topological ordering therefore persists even under resource injection: DSBT remains the most resource-efficient structure.

In the heterogeneous-link setting, the binary-tree study takes independent

$$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$3

for each of the $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$4 edges and repeats the assignment $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$5 times. Because $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$6, the disorder-averaged fidelities are reported to be close to those of the homogeneous $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$7 model. For $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$8, the comparisons are DABT $$\rho_{\text{wer}}=\frac{1-p}{4}\mathbb{I}+p\ket{\Psi^-}\bra{\Psi^-},$$9 vs $p\in[0,1]$0, DSBT $p\in[0,1]$1 vs $p\in[0,1]$2, UABT $p\in[0,1]$3 vs $p\in[0,1]$4, and USBT $p\in[0,1]$5 vs $p\in[0,1]$6; for $p\in[0,1]$7, they are DABT $p\in[0,1]$8 vs $p\in[0,1]$9, DSBT $\rho$00 vs $\rho$01, UABT $\rho$02 vs $\rho$03, and USBT $\rho$04 vs $\rho$05 (Roy et al., 14 Aug 2025). The reported implication is that the homogeneous $\rho$06 model is a good proxy for the disorder-averaged behavior under uniform random noise.

Despite the performance differences among the four binary-tree topologies, all are reported to satisfy

$\rho$07

(Roy et al., 14 Aug 2025). This aligns binary trees with the large-$\rho$08 behavior of chain-like loopless networks rather than with stars, for which the broader tree study reports retention of quantum advantage in the large-$\rho$09 limit whenever

$\rho$10

(Mylavarapu et al., 2024). This suggests that bounded-degree branching alone does not prevent asymptotic collapse of average teleportation fidelity; what matters is how branching modifies path growth relative to the hub-dominated star.

6. Related frameworks and adjacent binary-tree communication models

A second line of work places binary-tree repeater networks in a renormalization framework for arbitrary fractal quantum repeater networks. The key construction is coupled renormalization (CR): box-covering coarse-graining, replacement of each box by a hub, addition of shortcuts between hubs of connected boxes, and recursive coupling of the resulting coarse-grained network back to the original one (Wei et al., 2011). The diameter after $\rho$11 CR levels obeys

$\rho$12

with critical depth

$\rho$13

and at that depth the diameter becomes logarithmic in $\rho$14, yielding a small-world regime. The paper does not focus on binary trees specifically, but it states that a binary tree is conceptually very close to this recursive picture. A plausible implication is that a pure binary-tree repeater backbone is not automatically small-world; rather, recursive shortcut generation is the mechanism that converts a hierarchical tree into a scalable quantum small-world network (Wei et al., 2011).

A distinct but related model studies binary-tree spin networks for quantum state transfer under local dissipation, using a weak measurement (WM) on the sender qubit and a quantum measurement reversal (QMR) on the destination qubit (Behzadi et al., 2016). The network has $\rho$15 generations and $\rho$16 qubits, and a unitary transformation reduces the tree Hamiltonian to an effective chain-like invariant subspace. With optimal reversal strength

$\rho$17

the averaged transfer fidelity increases with the WM strength $\rho$18 and approaches $\rho$19 as $\rho$20, at the cost of reduced success probability. The same protocol also enhances entanglement distribution across the binary tree. This is not the same task as teleportation-fidelity averaging in Werner-state repeater networks, but it establishes that binary-tree architectures can also support repeater-like routing in Hamiltonian transport models (Behzadi et al., 2016).

A further combinatorial development studies growing binary trees via synchronized branching and extinction rules, with active frontier nodes (“anchors”) replaced at each step either by dead leaves or by internal nodes with two new anchors (Bodini et al., 26 Mar 2026). The model is not a quantum repeater model, but it provides exact counting, height bounds, profile constraints, and an entropy-optimal sampler for binary trees with prescribed profile. A plausible implication for repeater-network theory is that such growth/extinction formalisms could supply precise stochastic priors for hierarchical branching backbones, especially when active frontier repeaters may either terminate or split.

7. Conceptual synthesis

Binary tree repeater networks occupy a structurally important middle ground in quantum-network theory. Like all loopless trees, they are analytically governed by unique-path entanglement swapping, so the relevant fidelity for a source–target pair is the product of link Werner parameters along that path. Unlike chains, they exploit branching to reduce typical path lengths; unlike stars, they cannot concentrate connectivity into a single high-degree hub. The resulting performance is therefore topological in a strong sense: it depends on the entire path-length distribution rather than on local link quality alone (Mylavarapu et al., 2024, Roy et al., 14 Aug 2025).

The current binary-tree literature supports several stable conclusions. First, topology and edge quality are inseparable variables in any operational assessment. Second, symmetric branching is advantageous, but symmetry alone is insufficient: the reported best performer is the directed symmetric binary tree, not the undirected symmetric one (Roy et al., 14 Aug 2025). Third, even if every edge is individually above the single-link teleportation threshold, the network average may remain below the $\rho$21 quantum-advantage boundary. Fourth, maximally entangled links can compensate for weak uniform noise, but the number required is topology-dependent. Fifth, in the large-$\rho$22 limit, all four studied binary-tree families approach $\rho$23, so branching without additional multiscale shortcut structure does not by itself produce asymptotically scalable teleportation performance (Roy et al., 14 Aug 2025).

Taken together, these results define binary tree repeater networks as a mathematically tractable but nontrivial class of hierarchical quantum networks. They are sufficiently rich to exhibit strong topology dependence, threshold phenomena, and clear resource tradeoffs, yet sufficiently structured to admit exact fidelity expressions, asymptotic analysis, and connections to renormalization, open-system transport, and combinatorial tree growth.