Pith. sign in

REVIEW 5 major objections 4 minor 1 cited by

Constructing two completely independent spanning trees in the dual-cube

T0 review · 5 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The n-dimensional dual-cube F_n contains two completely independent spanning trees for every n ≥ 5, and a recursive algorithm constructs them with diameters 5n+5 and 5n+7 for n ≥ 6.

desk verdict New existence result for two CIST in dual-cubes with a genuinely useful cluster-based framework — but the load-bearing base case is under-verified as written. read the letter →

arxiv 2607.25917 v1 pith:Y3BGE7UD submitted 2026-07-28 math.CO

Mohammed Lalou , Nader Mbarek , Abdallah Skender , Olivier Togni This is my paper · ORCID
classification math.CO MSC 05C0505C3805C4068R10
keywords dual-cubecompletelyindependentspanningtreeshypercubeHamiltonianpathscompletebipartitegraphinterconnectionnetworksfault-tolerantroutingdiameter
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Completely independent spanning trees are pairs of spanning trees that share no edges and no vertex that is internal to more than one tree; they give two openly-disjoint routes between any two vertices. The paper establishes that the n-dimensional dual-cube F_n — a hypercube-like network with the same degree but fewer edges per vertex — contains such a pair for every n ≥ 5. The proof decomposes F_n into hypercube clusters, uses the fact that the clusters form a complete bipartite graph, threads two edge-disjoint Hamiltonian paths through that quotient, and then fills each cluster with two local spanning trees whose prescribed internal vertices match the cross-edge endpoints. A recursive algorithm extends the construction to all n ≥ 6 with diameters 5n+5 and 5n+7, and the authors conjecture that k CIST in the hypercube imply k CIST in the dual-cube. If correct, this gives the dual-cube the same fault-tolerant routing backbone as the hypercube at lower wire density.

What carries the argument

The cluster decomposition of F_n is the load-bearing structure: it turns a large graph into a complete bipartite quotient K_{2^{n-1},2^{n-1}} whose vertices are hypercubes. Lemma 1 ('generator pairs') guarantees local CIST inside Q_{n-1} with any two prescribed vertices internal in each tree; Lemma 2 supplies two edge-disjoint Hamiltonian paths on the quotient. The combination fixes, for each cluster, which vertices must be internal in T_1 and T_2, so gluing local trees along the paths yields two spanning trees that are edge-disjoint and share no internal vertex. The recursive diameter control selects cross-edges close to local centers, producing the 5n+5 / 5n+7 diameters.

What would settle it

Re-run the integer-linear-programming model for all 455 (v,x,y) configurations in Q_4 with the generator constraints; if any instance is infeasible, Lemma 1's base case fails and Theorem 2 collapses. Separately, instantiate Lemma 2's construction with m=8 and primes p=3, q=5 to see that the two Hamiltonian paths in K_{8,8} share edges, contradicting the lemma as stated.

Watch

Extended reading notes

Core claim

The central claim is that the n-dimensional dual-cube F_n contains two completely independent spanning trees for every n ≥ 5. The proof decomposes F_n into its (n−1)-dimensional hypercube clusters, treats the quotient of clusters as a complete bipartite graph, and shows that two edge-disjoint Hamiltonian paths on that quotient can be combined with local CIST inside each cluster. The key lemma asserts that for any n ≥ 4 and any two disjoint vertex pairs in Q_n, there exist two CIST of Q_n in which the prescribed pairs are internal vertices; the base case n=4 is verified exhaustively by an integer-linear-programming search. The recursive algorithm F-2CIST assembles the trees from an optimized

Load-bearing premise

The construction rests on a computer search over all 455 configurations in the 4-dimensional hypercube base case, which the paper reports as feasible without releasing the program or certificates.

Editorial extensions

If this is right

  • Every dual-cube F_n with n ≥ 5 supports two CIST, so between any two vertices there are two internally vertex-disjoint paths, one in each tree.
  • For n ≥ 6, the F-2CIST-Recursive algorithm yields two explicit CIST with diameters 5n+5 and 5n+7, growing linearly with the dimension.
  • At the same vertex count, the dual-cube's CIST diameters stay close to the hypercube's 2n−1, so the link-sparse topology remains competitive for fault-tolerant routing.
  • If Conjecture 1 holds, k CIST in Q_n would produce k CIST in F_{n+1}, potentially multiplying the number of independent routing backbones.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The quotient-plus-local-generators scheme is a reusable template: any network built from hypercube clusters wired as a complete bipartite graph should inherit CIST from edge-disjoint Hamiltonian structures on the quotient, so other hypercube variants with cluster layouts are natural targets.
  • The per-dimension diameter increment of 5 is tied to the chosen inter-cluster shortest-path trees; adjusting the local tree shapes or the cross-edge selection rule could reduce the constant.
  • The base case n=4 is checked by an unreleased ILP, so a human-readable constructive proof for Q_4 would remove the computational dependency from the whole induction.
  • The text reports F_5 diameters as 29 and 31 in the introduction and 28 and 31 in Theorem 3; resolving this discrepancy is needed before benchmarking the recursive formula.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 4 minor

Summary. The paper proves that for every n ≥ 5 the n-dimensional dual-cube F_n admits two completely independent spanning trees (CIST), and gives a recursive algorithm, F-2CIST-Recursive, claimed to produce two CIST with diameters 5n+5 and 5n+7 for n ≥ 6. The proof strategy is to decompose F_n into clusters isomorphic to Q_{n-1}, first finding two edge-disjoint Hamiltonian paths in the quotient graph K_{2^{n-1},2^{n-1}}, then constructing local CIST inside every cluster with prescribed internal vertices matched to the cross-edge endpoints. A base case for n = 5 is obtained by ILP plus a local optimization step.

Significance. If the proof is repaired, the result is a natural and useful contribution: it establishes the first CIST existence theorem for the dual-cube, a sparser hypercube variant, and it gives an explicit recursive construction with linear diameter. The conceptual decomposition into an inter-cluster Hamiltonian-path problem and a local hypercube CIST problem with prescribed internal vertices is elegant and likely to be reusable. The paper also proposes a plausible conjecture about lifting k CIST from Q_n to F_{n+1}. However, the proof as written is not reliable: the base-case verification is undercounted, a central induction case is missing, Lemma 2 is false as stated, and the diameter formulas in Theorem 4 are asserted rather than proved. These are load-bearing issues, though they appear repairable within the existing framework.

major comments (5)
  1. [Section 3, Lemma 1, base case] The claimed exhaustive ILP verification is not exhaustive. Fixing u = 0000, v has 15 choices and {x,y} is an unordered pair from the remaining 14 vertices, so the number of configurations is 15·C(14,2) = 1365, not C(15,3) = 455. Enumerating unordered triples {v,x,y} covers only one role assignment per triple; the three assignments are not equivalent under automorphisms fixing 0 because the internal Hamming distance of the second pair depends on which vertex is v. This is load-bearing because Lemma 1 is invoked for every cluster Q_{n-1} in Theorem 2. Please correct the enumeration, run all 1365 (or provide a rigorous reduction), and make the ILP instances or feasibility certificates available.
  2. [Section 3, Lemma 1, induction step] The induction omits the 2+2 distribution in which {u,v} lie in one copy of Q_n and {x,y} lie in the other. Cases 1–3 cover 4+0, 3+1, and 2+2 with one vertex of each tree in each copy, but not u_i=v_i=0, x_i=y_i=1. Consequently the assertion in Case 4 that the only remaining pattern is u_i=v_i≠x_i=y_i for every i is false. In addition, Case 3 is internally inconsistent: after stating that the relevant cross-pairs are (u^{i,0},v^{i,1}) and (x^{i,0},y^{i,1}), the subcases switch to (u,y) and (x,v), and Subcase 3.2 assumes u0–y1 is an edge but then connects copies using (v0,v1) and (x0,x1), never using the assumed edge. A coherent repair of this induction is needed before Lemma 1 can support Theorem 2.
  3. [Section 3, Lemma 2] The construction P_p and P_q is not edge-disjoint for arbitrary distinct primes p,q. In K_{m,m} the paths share edges whenever p ≡ ±q (mod m). For example, with m=16, p=3 and q=13, the edge 7_0–10_1 lies on both P_3 and P_13. Since m=2^{n-1}, such pairs of primes exist in the stated range. The underlying existence claim is true — K_{m,m} has edge-disjoint Hamiltonian cycles — but the lemma as stated is false. Because the explicit P_p/P_q are used to determine the required internal vertices in each cluster, the construction in Theorems 2 and Algorithm 1 must either use a correct edge-disjoint pair of Hamiltonian paths or add a hypothesis on p,q.
  4. [Section 4, Theorem 4] The proof of Theorem 4 asserts diam(T_1)=5n+5 and diam(T_2)=5n+7 for n≥6, but no induction calculation is given. 'By repeating the same construction recursively, we obtain the following equality' is not a proof: one must account for the distances within the four copies, the lengths contributed by the three cross-edges, and the choice of the extremal copies at each level. The base values in §4.2 are 29 and 31, so the recurrence would give 35 and 37 for n=6, but the paper does not show how these are computed from the local tree data. Please provide a genuine inductive distance analysis.
  5. [Section 4 and Appendix B] The explicit base-case data for F_5 is internally inconsistent. In Table 4, the row for cluster 14^0 contains the parent value 20 in a 16-vertex Q_4 cluster, which is outside the vertex set. Several vectors in Tables 3 and 4 have fewer than 16 entries (e.g., rows 7^0 and 11^0 in Table 3). Since these local trees are the basis for the diameter claims and the recursive algorithm, the tables need to be regenerated and corrected, or the code/solver output must be supplied so the reader can verify the claimed F_5 construction.
minor comments (4)
  1. [Section 4, Theorem 3] Theorem 3 states diameters 28 and 31, while §4.2 computes diam(T_1)=29 and diam(T_2)=31, and the abstract/introduction say 29 and 31. Please reconcile.
  2. [Section 4, Algorithm 3] Algorithm 3 calls F-2CIST-Base for the F_5 copies, but the proof of Theorem 4 uses trees produced by F-2CIST-Base and F-2CIST-Enhanced. Clarify which algorithm is actually the base of the recursion.
  3. [Section 3, Lemma 1, Case 3.1] There is a typo: 'Q^{i,0}_1' should be 'Q^{i,0}_n'.
  4. [Appendix B] The formatting of Tables 3 and 4 makes the columns (cluster, generators, center, parent vector) hard to parse; a legend explaining the row format would help.

Circularity Check

0 steps flagged · score 0.0 of 10

No circular dependency between the construction and the claimed conclusions.

full rationale

The derivation chain is not circular. Lemma 1 is proved by induction with a base case checked by an ILP that searches for witness trees; the ILP is not used to fit any parameter to the conclusion. Lemma 2 offers an explicit (though possibly flawed) modular construction for edge-disjoint Hamiltonian paths in the contracted graph, relying on standard Hamiltonian decomposition facts rather than on the target result. Theorem 2 assembles local CIST inside hypercube clusters with inter-cluster paths; nothing in the assembly assumes the existence of the final trees. Section 4's diameter formulas are computed from the constructed trees, not used as input constraints. There are no self-citations invoked as load-bearing external theorems: the ILP model is given in Appendix A, and the only overlapping-author citation ([17]) is the source of an ILP formulation whose constraints are restated in full. The known gaps — the undercount of all pairings in the n=4 enumeration and the apparent false claim that any two distinct primes give edge-disjoint Hamiltonian paths — are correctness/verification concerns, not instances where an output is equivalent to an input by construction.

Assumptions & free parameters 0 free parameters · 6 assumptions · 0 invented entities

No free parameters are fit to data. The proof rests on standard results plus a computer-assisted base case (not shipped) and a flawed constructive detail in Lemma 2 that can be repaired via the cited Hamiltonian decomposition.

assumptions (6)
  • standard math Hasunuma characterization (Theorem 1): CIST iff edge-disjoint and each vertex internal in at most one tree.
    Used in proofs of Theorems 2 and 3 to verify CIST conditions.
  • standard math K_{m,m} with m=2^{n-1} contains two edge-disjoint Hamiltonian paths (folklore, [13]).
    Foundation for inter-cluster connections in Theorem 2; Lemma 2 attempts a constructive proof.
  • ad hoc to paper Exhaustive ILP verification for the base case n=4 of Lemma 1 is correct.
    The paper asserts feasibility for all 455 instances but provides no solver files or certificates.
  • ad hoc to paper ILP model in Appendix A correctly encodes the CIST existence problem.
    The construction of local trees, including Lemma 1 base case and F5, depends on this encoding.
  • standard math Vertex-transitivity of the hypercube permits fixing one vertex in the ILP enumeration.
    Reduces Q4 base-case enumeration to 455 instances.
  • domain assumption Recursive construction of F_n from four copies of F_{n-1} as described (Section 2) is the intended dual-cube structure.
    Used in Lemma 3/Theorem 4 for recursive CIST assembly and diameter recurrence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Constructing two completely independent spanning trees in the dual-cube." pith.science (2026). https://pith.science/paper/Y3BGE7UD

@misc{pith2026260725917,
  author       = {Pith},
  title        = {Pith review of: Constructing two completely independent spanning trees in the dual-cube},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Y3BGE7UD}},
  note         = {Machine review of arXiv:2607.25917}
}
abstract

In this paper, we prove the existence of two completely independent spanning trees in the $n$-dimensional dual-cube $F_n$, a variant of the hypercube, for every $n \geq 5$. To this end, we use the hypercube structure of the clusters of $F_n$ to extend the construction of CIST from the $(n-1)$-dimensional hypercube to the dual-cube. In addition, we propose a recursive algorithm that builds the two trees while improving their diameters. Finally, we propose a conjecture concerning the existence of $k$ completely independent spanning trees in the dual-cube.

Figures

Figures reproduced from arXiv: 2607.25917 by the authors.

Figure 1
Figure 1. Recursive construction of Fn two CIST in this graph. Graph Degree # links Diameter Avg. distance Bisection width Qn n 2 nn n n 2 2 n 2 F n+1 2 n+1 2 2 n(n+1) 2 n + 1 n 2 + 1 − 1 2 n−1 2 2 n 4 [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Two CIST of Q4 with the generator pairs {u = 0, v = 7} for T1 and {x = 3, y = 10} for T2. Extending the construction from Qn to Qn+1: Assume that the statement holds for Qn for some n ≥ 4, we prove that it also holds for Qn+1: Recall that, for any 1 ≤ i ≤ n, Qn+1 can be decomposed into two copies of Qn, denoted by Qi,0 n and Qi,1 n , such that Qi,j n contains the vertices whose i-th coordinate is j for j = 0, 1. For… view at source ↗
Figure 3
Figure 3. Illustration of the construction of Pp in KFn ℓ 0 k 1 m1 k m ℓ ℓ [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (6 more)
Figure 4
Figure 4. Figure 4: Subpath of Pi induced by the clusters k 1 , ℓ 0 , and m1 Since each cluster is isomorphic to Qn−1 and n ≥ 5, by Lemma 1, two CIST can be constructed inside each C, denoted T1(C) and T2(C) with the required internal vertices. Let Ti be the union of all local trees Ti(C)…
Figure 5
Figure 5. Figure 5: Two edge-disjoint hamiltonian paths (P1 in red and P2 in blue) of F5 To construct the local CIST inside each cluster, we use the ILP formulation presented in Appendix A. Moreover, in our construction, we additionally impose the generator constraints described in the pr…
Figure 6
Figure 6. Figure 6: Clusteral and complete diametral chains of [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: Recursive construction of Fn and cross-edges used for the two CIST 14 [PITH_FULL_IMAGE:figures/full_fig_p014_7.png]
Figure 8
Figure 8. Figure 8: The weighted graph G1 and the tree T1 (plain red edges) that minimizes the diameter (eccentricity from cluster 3 0 to cluster 1 1 through the center 6 1 ) 23 [PITH_FULL_IMAGE:figures/full_fig_p023_8.png]
Figure 9
Figure 9. Figure 9: The weighted graph G2 and the tree T2 (plain blue edges) that minimizes the diameter (eccentricity from cluster 4 1 to cluster 6 1 through the center 5 0 ) 24 [PITH_FULL_IMAGE:figures/full_fig_p024_9.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. An explicit construction of two completely independent spanning trees in the four-dimensional dual-cube

    math.CO 2026-08 accept novelty 6.0 of 10

    F_4, the four-dimensional dual-cube, admits two completely independent spanning trees, completing the classification for all n≥4.

Reference graph

Works this paper leans on

31 extracted references · 15 canonical work pages · cited by 1 Pith paper

  1. [1]

    T. Araki. Dirac’s Condition for Completely Independent Spanning Trees.Journal of Graph Theory, 77(3):171–179, 2014. doi:10.1002/jgt.21780

    doi:10.1002/jgt.21780 2014

  2. [2]

    Barabde, S

    R. Barabde, S. A. Mane, and S. A. Kandekar. Construction and conditions for completely independent spanning trees in hypercubes and regular bipartite graphs, Mar. 2025

    2025

  3. [3]

    Chang, K.-J

    Y.-H. Chang, K.-J. Pai, C.-C. Hsu, J.-S. Yang, and J.-M. Chang. Constructing dual- CISTs of folded divide-and-swap cubes.Theoretical Computer Science, 856:75–87, Feb

  4. [4]

    Chen, S.-M

    Y.-H. Chen, S.-M. Tang, K.-J. Pai, and J.-M. Chang. Constructing dual-CISTs with short diameters using a generic adjustment scheme on bicubes.Theoretical Computer Science, 878-879:102–112, July 2021. doi:10.1016/j.tcs.2021.05.031

    work page doi:10.1016/j.tcs.2021.05.031 2021

  5. [5]

    Chen, K.-J

    Y.-H. Chen, K.-J. Pai, H.-J. Lin, and J.-M. Chang. Constructing tri-CISTs in shuffle-cubes. Journal of Combinatorial Optimization, 44(5):3194–3211, Dec. 2022. doi:10.1007/s10878- 022-00863-0

    work page doi:10.1007/s10878- 2022

  6. [6]

    Cheng, J

    B. Cheng, J. Fan, D. Wang, and J. Yang. A Reliable Broadcasting Algorithm in Locally Twisted Cubes. In2015 IEEE 2nd International Conference on Cyber Security and Cloud Computing, pages 323–328, Nov. 2015. doi:10.1109/CSCloud.2015.13

    work page doi:10.1109/cscloud.2015.13 2015

  7. [7]

    Cheng, D

    B. Cheng, D. Wang, and J. Fan. Constructing completely independent spanning trees in crossed cubes.Discrete Applied Mathematics, 219:100–109, Mar. 2017. doi:10.1016/j.dam.2016.11.019

    work page doi:10.1016/j.dam.2016.11.019 2017

  8. [8]

    Cheng, D

    B. Cheng, D. Wang, and J. Fan. Independent spanning trees in networks: A survey.ACM Computing Surveys, 55(14S):1–29, 2023. doi:10.1145/3591110

    doi:10.1145/3591110 2023

Show all 31 references
  1. [9]

    G. Fan, Y. Hong, and Q. Liu. Ore’s condition for completely independent spanning trees. Discrete Applied Mathematics, 177:95–100, Nov. 2014. doi:10.1016/j.dam.2014.06.002. 17

    2014 doi

  2. [10]

    Hasunuma

    T. Hasunuma. Completely independent spanning trees in the underlying graph of a line digraph.Discrete Mathematics, 234(1):149–157, 2001. doi:https://doi.org/10.1016/S0012- 365X(00)00377-0

    2001 doi

  3. [11]

    Hasunuma

    T. Hasunuma. Completely Independent Spanning Trees in Maximal Planar Graphs. In G. Goos, J. Hartmanis, J. van Leeuwen, and L. Kuˇ cera, editors,Graph-Theoretic Concepts in Computer Science, pages 235–245, Berlin, Heidelberg, 2002. Springer. doi:10.1007/3-540- 36379-3 21

    2002 doi

  4. [12]

    Hong and H

    X. Hong and H. Zhang. A Hamilton sufficient condition for completely independent spanning tree.Discrete Applied Mathematics, 279:183–187, May 2020. ISSN 0166- 218X. doi:10.1016/j.dam.2019.08.013. URLhttps://www.sciencedirect.com/science/ article/pii/S0166218X19303865

    2020 doi

  5. [13]

    Laskar and B

    R. Laskar and B. Auerbach. On decomposition of r-partite graphs into edge-disjoint hamilton circuits.Discrete Mathematics, 14(3):265–268, Jan. 1976. doi:10.1016/0012-365X(76)90039- X

    1976 doi

  6. [14]

    Y. Li, S. Peng, and W. Chu. Hamiltonian cycle embedding for fault tolerance in dual-cube. InThe IASTED International Conference on Networks, Parallel and Distributed Processing, and Applications, pages 1–6, 2002

    2002

  7. [15]

    Ma and J

    J. Ma and J. Cai. Fan’s condition for completely independent spanning trees, Feb. 2025. URLhttp://arxiv.org/abs/2502.11522. arXiv:2502.11522 [math]

    2025 arXiv

  8. [16]

    S. A. Mane, S. A. Kandekar, and B. N. Waphare. Constructing spanning trees in aug- mented cubes.Journal of Parallel and Distributed Computing, 122:188–194, Dec. 2018. doi:10.1016/j.jpdc.2018.08.006

    2018 doi

  9. [17]

    Moinet, B

    A. Moinet, B. Darties, N. Gastineau, J.-L. Baril, and O. Togni. Completely indepen- dent spanning trees for enhancing the robustness in ad-hoc Networks. In2017 IEEE 13th International Conference on Wireless and Mobile Computing, Networking and Com- munications (WiMob), pages 6...

    2017

  10. [18]

    K.-J. Pai. Designing an Algorithm to Improve the Diameters of Completely Independent Spanning Trees in Crossed Cubes. InNew Trends in Computer Technologies and Applica- tions, pages 433–439, Singapore, 2019. Springer. doi:10.1007/978-981-13-9190-3 46

    2019 doi

  11. [19]

    Pai and J.-M

    K.-J. Pai and J.-M. Chang. Constructing two completely independent spanning trees in hypercube-variant networks.Theoretical Computer Science, 652:28–37, Nov. 2016. doi:10.1016/j.tcs.2016.08.024

    2016 doi

  12. [20]

    Pai and J.-M

    K.-J. Pai and J.-M. Chang. Improving the diameters of completely independent span- ning trees in locally twisted cubes.Information Processing Letters, 141:22–24, Jan. 2019. doi:10.1016/j.ipl.2018.09.006

    2019 doi

  13. [21]

    Pai, J.-S

    K.-J. Pai, J.-S. Yang, S.-C. Yao, S.-M. Tang, and J.-M. Chang. Completely Independent Spanning Trees on Some Interconnection Networks.IEICE Transactions on Information, E97-D(9):2514–2517, Sept. 2014. doi:10.1587/transinf.2014EDL8079

    2014 doi

  14. [22]

    Pai, R.-S

    K.-J. Pai, R.-S. Chang, R.-Y. Wu, and J.-M. Chang. A two-stages tree-searching algorithm for finding three completely independent spanning trees.Theoretical Computer Science, 784: 65–74, Sept. 2019. doi:10.1016/j.tcs.2019.03.035. 18

    2019 doi

  15. [23]

    Pai, R.-S

    K.-J. Pai, R.-S. Chang, and J.-M. Chang. A protection routing with secure mechanism in M¨ obius cubes.Journal of Parallel and Distributed Computing, 140:1–12, June 2020. doi:10.1016/j.jpdc.2020.02.007

    2020 doi

  16. [24]

    P´ eterfalvi

    F. P´ eterfalvi. Two counterexamples on completely independent spanning trees.Discrete Mathematics, 312(4):808–810, Feb. 2012. doi:10.1016/j.disc.2011.11.015

    2012 doi

  17. [25]

    Qin and R.-X

    X.-W. Qin and R.-X. Hao. Reliability analysis based on the dual-CIST in shuffle-cubes.Ap- plied Mathematics and Computation, 397:125900, May 2021. doi:10.1016/j.amc.2020.125900

    2021

  18. [26]

    B. R. Shaw. Completely independent spanning trees in the hypercube, Dec. 2024

    2024

  19. [27]

    Shih, H.-C

    Y.-K. Shih, H.-C. Chuang, S.-S. Kao, and J. J. M. Tan. Mutually independent Hamil- tonian cycles in dual-cubes.The Journal of Supercomputing, 54(2):239–251, Nov. 2010. doi:10.1007/s11227-009-0317-2

    2010 doi

  20. [28]

    Wu and H

    C. Wu and H. Wu. On self-similarity and hamiltonicity of dual-cubes. InProceedings International Parallel and Distributed Processing Symposium, pages 7–pp. IEEE, 2003

    2003

  21. [29]

    Fault-tolerant routing and disjoint paths in dual-cube: a new interconnection network

    Yamin Li and Shietung Peng. Fault-tolerant routing and disjoint paths in dual-cube: a new interconnection network. InProceedings. Eighth International Conference on Par- allel and Distributed Systems. ICPADS 2001, pages 315–322. IEEE Comput. Soc, 2001. doi:10.1109/ICPADS.2001.934835

    2001

  22. [30]

    Yang, K.-J

    Y.-X. Yang, K.-J. Pai, R.-S. Chang, and J.-M. Chang. Constructing Two Completely Inde- pendent Spanning Trees in Balanced Hypercubes.IEICE Transactions on Information and Systems, E102.D(12):2409–2412, Dec. 2019. doi:10.1587/transinf.2019PAL0001. Appendices A Linear modeling L...

    2019 doi

  23. [2021]

    doi:10.1016/j.tcs.2020.12.023

    2020 doi

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.