Pith. sign in

REVIEW 4 major objections 6 minor 27 references

Short note on phase retrievable weaving fusion frames

T0 review · 4 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read Two interwoven subspace frames are phase retrievable exactly when every weaving phase-lift operator avoids rank-two symmetric operators.

desk verdict A natural definition and a plausible kernel condition, but the main proof relies on a false lemma and the application section is vacuous; not refereeable as written. read the letter →

arxiv 2506.08478 v3 pith:57HZ472N submitted 2025-06-10 math.FA

Avinash Bhardwaj , Animesh Bhandari This is my paper · ORCID
classification math.FA MSC 42C1546A3247A05
keywords phaseretrievalweavingfusionframesliftoperatorrank-twosymmetricoperatorsprobabilisticerasureframetheorycomplementproperty
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

Two collections of closed subspaces of an $n$-dimensional Hilbert space are called weaving fusion frames when every way of interleaving them still produces a frame. The paper asks when such a pair is phase retrievable: when the lengths of the projections of a vector onto the chosen subspaces determine the vector up to a global unimodular factor, and do so for every interleaving. The central result, Theorem 3.4, says this happens exactly when the weaving phase lift operator $F_\sigma$ has trivial intersection with $S_{1,1}(H_n)$, the space of symmetric operators of rank at most two, for every subset $\sigma$. That converts a nonlinear injectivity question into a linear-algebraic kernel condition. The paper also derives a quantitative stability lower bound, proves invariance under a common unitary, and applies the setup to probabilistic erasure of measurements.

What carries the argument

The central object is the weaving phase lift operator $F_\sigma:\operatorname{Sym}(H_n)\to\mathbb R^m$, defined by $(F_\sigma T)_i=v_i^2\operatorname{tr}(P_{V_i}T)$ for $i\in\sigma$ and $w_i^2\operatorname{tr}(P_{W_i}T)$ for $i\notin\sigma$. It linearizes magnitude measurements because $\operatorname{tr}(P_V(f\otimes f))=\|P_V f\|^2$. The companion space $S_{1,1}(H_n)=\{T\in\operatorname{Sym}(H_n):\dim R(T)\le 2\}$ is exactly the collection of differences of two rank-one operators $f\otimes f-g\otimes g$; by the cited decomposition theorem [14], every such operator is of the form $\frac12(f\otimes g+g\otimes f)$. The equivalence holds because two vectors produce identical magnitude data precisely when their rank-one difference lies in the kernel of $F_\sigma$, so a trivial kernel intersection with $S_{1,1}(H_n)$ is exactly injectivity of $\gamma_\sigma$ on the quotient $H_n/\!\sim$.

What would settle it

Compute $F_\sigma(x\otimes x-y\otimes y)$ for the pair in Example 3.3 with $\sigma=\{1,2\}$, where $x=(1,i)$ and $y=(i,1)$ are non-parallel yet share all magnitude measurements; the theorem predicts the resulting nonzero rank-two operator lies in $\operatorname{Ker} F_\sigma$. A direct search over small $n$ and random subspace pairs that finds any nonzero rank-two operator in $\operatorname{Ker} F_\sigma$ for a genuinely phase-retrievable pair would disprove the iff, and exhaustively checking all $\sigma$ would either confirm or reject the characterization computationally.

Watch

Extended reading notes

Core claim

The paper's central assertion is Theorem 3.4: for two weaving fusion frames $\{(V_i,v_i)\}_{i=1}^m$ and $\{(W_i,w_i)\}_{i=1}^m$ in $H_n$, the pair is phase retrievable for every interleaving if and only if $\operatorname{Ker} F_\sigma \cap S_{1,1}(H_n)=\{0\}$ for every $\sigma\subseteq\{1,\dots,m\}$. Here $F_\sigma$ is the weaving phase lift operator and $S_{1,1}(H_n)$ is the set of symmetric operators whose range has dimension at most two. In the author's framing, the nonlinear problem of deciding whether magnitude measurements distinguish vectors up to phase collapses to a linear problem: checking whether any rank-two symmetric operator is annihilated by all selected measurement traces. The paper proves this equivalence by representing every rank-two symmetric operator as a difference of two rank-one operators, $T=f\otimes f-g\otimes g$, and showing that equality of the magnitude data $\gamma_\sigma(f)=\gamma_\sigma(g)$ is exactly the condition $F_\sigma(f\otimes f-g\otimes g)=0$.

Load-bearing premise

The equivalence assumes that every failure of phase retrieval is visible as a rank-two symmetric operator $f\otimes f-g\otimes g$ lying in the kernel of $F_\sigma$, so checking that the kernel meets $S_{1,1}(H_n)$ only at zero is sufficient to rule out all failures.

Editorial extensions

If this is right

  • Phase retrievability of a weaving fusion frame pair can be certified by finite linear algebra: for each of the $2^m$ subsets $\sigma$, build the matrix of $F_\sigma$ and test whether its kernel contains any nonzero rank-two symmetric operator.
  • Every obstruction has a concrete witness: a nonzero $T\in\operatorname{Ker}F_\sigma\cap S_{1,1}(H_n)$ yields, through the decomposition $T=f\otimes f-g\otimes g$, two vectors that are not unimodular multiples yet have identical magnitude data.
  • When the pair is phase retrievable, Theorem 3.6 provides a uniform lower bound $\sum_{i\in\sigma}\|v_i^2\operatorname{tr}(P_{V_i}[f,g])\|^2+\sum_{i\notin\sigma}\|w_i^2\operatorname{tr}(P_{W_i}[f,g])\|^2\ge\alpha\|[f,g]\|_1$, giving a stability margin $\alpha>0$ that does not depend on $f,g$.
  • Applying a common unitary $Q$ to all subspaces preserves phase retrievability and the weaving bounds, so the property is unchanged under a change of orthonormal basis (Proposition 3.7).
  • If the uniform-tightness hypothesis of Theorem 4.1 were satisfiable, random erasure of weaving measurements would give expected reconstruction error controlled by $\epsilon$ whenever $\epsilon^2\ge \frac{n}{m}\log n$.

Reading between the lines

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

  • The kernel condition is the weaving analogue of the complement property for ordinary frames: instead of asking that one side of every subset spans, it asks that no rank-two symmetric operator is invisible to all interleaved measurements, a weaker and more linear condition.
  • Theorem 4.1's hypothesis $\|P_{V_n}f\|=\sqrt n=\|P_{W_n}f\|$ for every $f\in\mathbb R^n$ cannot be met by proper orthogonal projections when $n>1$, since a projection's norm varies with $f$ unless it is $0$ or $I$; the probabilistic erasure theorem as stated is therefore vacuous, although this does not affect Theorems 3.4 and 3.6.
  • The converse direction of Theorem 3.6 is not fully written out; the proof says it follows 'using a similar approach,' so the claimed equivalence of phase retrievability with the lower-bound stability condition rests partly on a sketch.
  • Because $F_\sigma$ acts on traces of symmetric operators, the criterion transfers to quantum-state discrimination: the same traces are intensities measured in von Neumann measurements, so Theorem 3.4 gives a condition for distinguishing pure states from intensity data across interleaved measurement configurations.
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

4 major / 6 minor

Summary. The manuscript defines phase retrievable weaving fusion frames as pairs of fusion frames for which every interleaved measurement map γ_σ is injective on the projective quotient. Its main theoretical result, Theorem 3.4, claims that this property is equivalent to a kernel condition on the associated weaving phase lift operator F_σ intersected with the space S_{1,1}(H_n) of symmetric operators of range dimension at most two. Theorem 3.6 claims an equivalent lower-bound inequality involving the symmetrized rank-one operators [f,g]; Proposition 3.7 asserts unitary invariance; and Section 4 presents a probabilistic erasure estimate. The paper also includes two examples, a Python complement-property check, and several remarks connecting the results to quantum information and distributed processing.

Significance. If Theorem 3.4 were correct, it would provide a compact algebraic characterization of phase retrievability for weaving fusion frames and would be a natural extension of Balan's phase-lift framework. The erasure result, if valid, would give a quantitative robustness statement for weaving fusion frames. However, the soundness defects described below affect the central characterization, the stated equivalent inequality, and the application result. The manuscript has not established its main claims, so its significance as a contribution is currently not realized. The paper does contain some useful definitions and a plausible example in R^2, and the idea of studying phase retrieval in the weaving fusion frame setting is reasonable.

major comments (4)
  1. [§2, Theorem 2.6(2)] Theorem 2.6(2) is false over the real field, and the proof of Theorem 3.4 depends on it. For H=R^2, the operator T=diag(1,2) belongs to S_{1,1}(H) by the paper's definition, since dim R(T)=2. But det(f⊗f−g⊗g) = −det([f g])^2 ≤ 0 for every f,g∈R^2, so this positive-definite T cannot be written as f⊗f−g⊗g. Thus the step in the proof of Theorem 3.4, implication (1⇒2), in which an arbitrary kernel element T is represented as f⊗f−g⊗g via Theorem 2.6, is unjustified. The authors need either to prove that positive-semidefinite rank-two operators cannot lie in Ker F_σ under phase retrievability, or to replace Theorem 2.6 by a decomposition valid on the relevant subset of S_{1,1}(H_n).
  2. [§3, Theorem 3.6] Theorem 3.6 is false as stated because of scaling. Replacing f,g by t f,t g with t>0 scales the left side of the claimed inequality by t^4, while ||[f,g]||_1 scales by t^2; hence no fixed constant α>0 can satisfy the inequality for all f,g. The proof itself has the same problem: it defines α as a minimum over the unit trace sphere, but then applies the resulting bound to [f,g]/||[f,g]||_1 and multiplies by ||[f,g]||_1^2, while the numerator already contains squared terms. Moreover, implication (2⇒1) is not proved; the text only says it follows by a similar approach. The statement and proof need a consistent normalization, for example a bound of the form ||F_σ([f,g])||^2 ≥ α ||[f,g]||_1^2 with α depending on the frames, not on f,g.
  3. [§4, Theorem 4.1] The hypothesis of Theorem 4.1 is unsatisfiable for n>1. It assumes that for every f∈R^n and every n, ||P_{V_n}f||=√n=||P_{W_n}f||. For an orthogonal projection P, the function f↦||P f|| is constant on all of R^n only when P is the zero operator or the identity, neither of which gives the value √n for every f. Therefore the theorem's assumption can never hold, and the probabilistic erasure application is vacuous. The definitions of δ_n and tilde δ_n/p_n in the proof are also not spelled out, and the displayed expression for \hat f switches between operators and vectors, so the derivation cannot be checked as written.
  4. [§3, Example 3.5] The verification of Example 3.5 is not complete. The text asserts that the frames are phase retrievable and that Ker F_σ(T) ∩ S_{1,1}(R^3)={0}, but it only computes the kernel explicitly for two subsets σ, and even for those cases the conclusion is summarized as matrices that are 'either invertible or zero' rather than as a full rank/range-dimensionality argument. Since the example is offered as validation of Theorem 3.4, all subsets σ should be checked or an algorithmic verification should be provided.
minor comments (6)
  1. [Title and abstract] The title contains spacing errors ('RETRIEV ABLE', 'WEA VING') and the abstract is generic; it would help to state the main theorems explicitly.
  2. [§2, Theorem 2.5] The theorem is labeled 'Comlementary Property'; this should be 'Complementary Property'.
  3. [§3, Example 3.2] The injectivity verification for σ=∅ says that six of eight sign combinations are 'not well defined' but does not justify this claim. All mixed-sign cases should either be listed or handled by a uniform argument.
  4. [§3, Example 3.3] The example works over C^2 while most of the paper's linear-algebra statements are phrased for real symmetric operators; the authors should state explicitly which results are intended over C and which over R.
  5. [§4] The notation δ_n is used for both a 'standard dirac delta function' and a Bernoulli-type random variable; the tilde quantities are undefined. The displayed derivation also confuses the operator E with the vectors PV_n f⊗S^{-1}PV_n f.
  6. [Listing 1] The Python code is difficult to read because of line breaks and missing whitespace; it should be reformatted as a proper appendix or supplementary listing.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular reduction; the central characterization rests on an external theorem of Balan, and the only self-citations are background pointers.

full rationale

The paper's central derivation, Theorem 3.4, is not circular. Definition 3.1 defines phase retrievability by injectivity of the map gamma_sigma, and the phase lift operator F_sigma is then shown to satisfy F_sigma(f otimes f - g otimes g) = 0 exactly when gamma_sigma(f) = gamma_sigma(g) coordinatewise. The equivalence with the kernel condition Ker F_sigma cap S_{1,1}(H_n) = {0} is a genuine linearization step; its nontrivial input is the external representation theorem Theorem 2.6, attributed to Balan [14], not to the authors' own work. The only self-citations are background references: [12] is cited for general discussion of weaving fusion frames, and [20] is cited in the introduction for K-fusion frames; neither is load-bearing for the main theorem. Theorem 4.1 is indeed vacuous because the hypothesis ||P_{V_n} f|| = sqrt(n) = ||P_{W_n} f|| for every f is unsatisfiable for nontrivial projections in dimension n > 1. There is also a serious correctness gap in Theorem 3.4: Theorem 2.6 is false over R, since T = diag(1,2) lies in S_{1,1}(R^2) by the paper's definition but has positive determinant and therefore cannot equal f otimes f - g otimes g. These are correctness problems, not circular reductions: no parameter is fitted, no known result is renamed as new, and no conclusion is assumed in its own proof. Accordingly, no circular step is present; the score 2 reflects only the presence of minor, non-load-bearing self-citations.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The main characterization depends on standard external theorems (Balan, complement property) and finite-dimensionality. The erasure application adds an impossible existence assumption on the frames, which is the most fragile part of the paper.

assumptions (5)
  • domain assumption Finite-dimensional Hilbert space H_n with n < infinity for main theorems
    Theorems 3.4, 3.6, and 3.7 are stated for H_n; the extension to infinite dimensions is not treated.
  • standard math Balan's characterization of S_{1,1}(H_n) (Theorem 2.6)
    Used in the proof of Theorem 3.4 to represent T in S_{1,1} as f⊗f - g⊗g.
  • standard math Complement property of Balan-Casazza-Edidin (Theorem 2.5)
    Quoted as background and used in the Python code to check phase retrieval via spanning of subsets.
  • ad hoc to paper Existence of uniform tight weaving frames with ||P_{V_n} f|| = sqrt(n) for all f in R^n
    Assumed in Theorem 4.1; impossible for n > 1, making the theorem vacuous.
  • standard math Rudelson's lemma [25]
    External concentration inequality cited for the erasure bound in Theorem 4.1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Short note on phase retrievable weaving fusion frames." pith.science (2026). https://pith.science/paper/57HZ472N

@misc{pith2026250608478,
  author       = {Pith},
  title        = {Pith review of: Short note on phase retrievable weaving fusion frames},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/57HZ472N}},
  note         = {Machine review of arXiv:2506.08478}
}
read the original abstract

Fusion frames are extensively studied due to their effectiveness in recovering signals from large-scale data. They are applicable in distributed processing, wireless sensor networks, and packet encoding systems due to their robustness and redundancy. Motivated by the foundational work of Bemrose et al.\cite{Be16} and Balan\cite{Ba13}, this paper investigates the theoretical properties and characterizations of phase retrievable weaving fusion frames. These frames offer enhanced redundancy and stability in signal reconstruction. We present key results that deepen the understanding of their structure and behaviour. Lastly, an application involving probabilistic erasure is explored to demonstrate their practical utility.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

27 extracted references · 27 canonical work pages

  1. [1]

    Duffin and A

    R. Duffin and A. C. Schaeffer, A class of non harmonic Fourier series. Trans. Amer. Math. Soc., 72 (1952), 341–366

    work page 1952

  2. [2]

    Daubechies, A

    I. Daubechies, A. Grossmann and Y. Meyer, Painless non orthogonal expansions, J. Math. Phys., 27 (1986), 1271–1283

    work page 1986

  3. [3]

    Han and D

    D. Han and D. R. Larson, Frames, Bases and Group Representations, Amer. Math. Soc., 2000

    work page 2000

  4. [4]

    Gr¨ ochenig, Foundations of Time-frequency Analysis, New York(NY), Birkh¨ auser, 2001

    K. Gr¨ ochenig, Foundations of Time-frequency Analysis, New York(NY), Birkh¨ auser, 2001

    work page 2001

  5. [5]

    Gumber and N

    A. Gumber and N. K. Shukla, Uncertainty principle corresponding to an or- thonormal wavelet system, Applicable Analysis, 97 (2018), 486–498

    work page 2018

  6. [6]

    Ferreira, Mathematics for multimedia signal processing II: discrete finite frames and signal reconstruction, Signal Processing for Multimedia

    P. Ferreira, Mathematics for multimedia signal processing II: discrete finite frames and signal reconstruction, Signal Processing for Multimedia. J. S. Byrnes (Ed.), 1999, 35–54

    work page 1999

  7. [7]

    Arefijamaal and E

    A. Arefijamaal and E. Zekaee, Image processing by alternate dual Gabor frames, Bull. Iran. Math. Soc., 42 (2016), 1305–1314

    work page 2016

  8. [8]

    P. G. Casazza, G. Kutyniok, S. Li and C. J. Rozell, Modeling sensor networks with fusion frames, Proc. SPIE. 2007;6701:67011M-1–67011M-11

    work page 2007

Show all 27 references
  1. [9]

    P. G. Casazza and G. Kutyniok, Finite Frames: Theory and Applications, Boston: Birkh¨ auser, Applied and Numerical Harmonic Analysis, 2012

    2012

  2. [10]

    Verma, L

    G. Verma, L. K. Vashisht and M. Singh, On Excess of Retro Banach Frames, Journal of Contemporary Mathematical Analysis (Armenian Academy of Sci- ences), 54 (2019), 143–146

    2019

  3. [11]

    Deepshikha, S. Garg, L. K. Vashisht and G. Verma, On Weaving Fusion Frames for Hilbert Spaces, International Conference on Sampling Theory and Applica- tions (SAMPTA), 2017

    2017

  4. [12]

    Bhandari, A note on weaving fusion Frames, New York J

    A. Bhandari, A note on weaving fusion Frames, New York J. Mathematics, 31 (2025), 54–65

    2025

  5. [13]

    Bemrose, P

    T. Bemrose, P. G. Casazza, K. Gr¨ ochenig, M.C. Lammers and R. G. Lynch, Weaving frames, Operators and Matrices, 10 (2016), 1093–1116

    2016

  6. [14]

    Balan, Stability of Phase Retrievable Frames, SPIE Optical Engineering and Applications, International Society for Optics and Photonics, (2013), DOI: 10.1117/12.2026135

    R. Balan, Stability of Phase Retrievable Frames, SPIE Optical Engineering and Applications, International Society for Optics and Photonics, (2013), DOI: 10.1117/12.2026135

    2013 doi

  7. [15]

    P. G. Casazza and G. Kutyniok, Frames of subspaces, Contemporary Math, AMS, 345 (2014), 87–114. SHORT NOTE ON PHASE RETRIEV ABLE WEA VING FUSION FRAMES 17

    2014

  8. [16]

    Shamsabadi, A

    M. Shamsabadi, A. Arefijamaal and P. Balazs, The invertibility of U -fusion cross Gram matrices of operators, Mediterr. J. Math., 17 (2020), 130

    2020

  9. [17]

    Sun, G-frames and G-Riesz bases, J

    W. Sun, G-frames and G-Riesz bases, J. Math. Anal. Appl., 322 (2006), 437– 452

    2006

  10. [18]

    Gˇ avrut ¸a, Frames for operators, Appl

    L. Gˇ avrut ¸a, Frames for operators, Appl. Comput. Harmon. Anal., 32 (2012), 139–144

    2012

  11. [19]

    Gˇ avrut ¸a, On the duality of fusion frames, Journal Math

    P. Gˇ avrut ¸a, On the duality of fusion frames, Journal Math. Anal. Appl., 333 (2007), 871–879

    2007

  12. [20]

    Bhandari and S

    A. Bhandari and S. Mukherjee, Atomic subspaces for operators, Indian J. Pure Appl. Math., 51 (2020), 1039–1052

    2020

  13. [21]

    Christensen, An Introduction to Frames and Riesz Bases, Boston: Birkh¨ auser, 2016

    O. Christensen, An Introduction to Frames and Riesz Bases, Boston: Birkh¨ auser, 2016

    2016

  14. [22]

    D. Han, K. Kornelson, D.R. Larson and E. Weber, Frames for Undergraduates, Student Mathematical Library, AMS Rhode Island, 2007

    2007

  15. [23]

    Kohldorfer and P

    L. Kohldorfer and P. Balazs, On the relation of the frame related operators of fusion frame systems, Sampl. Theory Signal Process. Data Anal., 21 (2023)

    2023

  16. [24]

    Parzen, Stochastic Processes, SIAM, 24 (1999)

    E. Parzen, Stochastic Processes, SIAM, 24 (1999)

    1999

  17. [25]

    Rudelson, Random Vectors in the Isotropic Position, Journal Func

    M. Rudelson, Random Vectors in the Isotropic Position, Journal Func. Anal., 164 (1999), 60–72

    1999

  18. [26]

    Balan, P

    R. Balan, P. Casazza and D. Edidin, On signal reconstruction with phase. Appl. Comput. Harmon. Anal., 20 (2006), 345–356

    2006

  19. [27]

    T. D. Juste, Frames and phase retrieval, Phd Dissertation: University of Cen- tral Florida, 2019. Department of Mathematics, SRM University,AP - Andhra Pradesh, Neeru Konda, Amaravati, Andhra Pradesh - 522240, India Email address: avinashbhardwaj320@gmail.com,bhandari.animesh@...

    2019

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.