Causality Sum Rules in Conventional Scattering Matrices

Hongsheng Chen; Mingzhu Li; Ning Han; Rui Zhao; Shuxing Yang; Yihao Yang

REVIEW 3 major objections 3 minor 25 references

Causality Sum Rules in Conventional Scattering Matrices

T0 review · 3 major / 3 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Two new causality sum rules act directly on measured scattering matrices

desk verdict The domain-delay idea is right, but Eq. (2) is off by a factor of two—the one-pole Schur check gives πτ/2, not πτ, so the Rozanov recovery is actually 2πd/c rather than πd/c. read the letter →

arxiv 2608.10427 v1 pith:DESXGYBD submitted 2026-08-11 physics.optics cs.AI

Ning Han , Rui Zhao , Shuxing Yang , Mingzhu Li , Hongsheng Chen , Yihao Yang This is my paper · ORCID

classification physics.opticscs.AI

keywords causalitysumrulesscatteringmatrixpassivitySchurfunctionsHerglotzrepresentationRozanovboundmultichannelattenuationdelay-bandwidthlimit

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

The paper shows that the standard scattering matrix $S(\omega)$ of a passive electromagnetic device already encodes causality limits, once the time advance introduced by the finite reference domain is removed. Multiplying by the unitary factor $e^{i\omega T_a}$, where $T_a=\mathrm{diag}(\tau_\alpha)$ collects each channel's earliest-arrival delay, yields a domain-delayed matrix $\widetilde S(\omega)$ that keeps the real-frequency passivity of $S$ and, under stated assumptions, becomes an operator Schur function. From the Schur property the paper derives a projected sum rule that bounds the integrated logarithmic suppression of a coherent channel superposition by its causal delay, and a determinant sum rule that bounds aggregate multichannel attenuation by $\pi\,\mathrm{tr}(T_a)$. These rules recover Rozanov's absorber bound and spherical-multipole sum rules and extend causality limits to insertion loss, singular-value suppression, and conditional lossless delay-bandwidth trade-offs.

What carries the argument

The carrying object is the domain-delayed scattering matrix $\widetilde S(\omega)=D_a(\omega)S(\omega)$ with $D_a(\omega)=e^{i\omega T_a}$ and $T_a=\mathrm{diag}(\tau_\alpha)$ the earliest-arrival delay per output channel. The unitary factor removes the reference-domain time advance without altering real-frequency power balance, and the product is claimed to be an operator Schur function: analytic in the upper half-plane and contractive there. The matrix Cayley transform $W=i(I+\widetilde S)(I-\widetilde S)^{-1}$ converts contractivity into an operator Herglotz function with $\mathrm{Im}\,W(z)\succeq 0$, and the Herglotz spectral representation turns analyticity and passivity into the logarithmic moment identities that produce the two sum rules.

What would settle it

Measure the full passive scattering matrix of a finite object over a sufficiently broad band, form $\int_0^\infty -\ln|\det S(\omega)|\,d\omega/\omega^2$, and compare it with $\pi\,\mathrm{tr}(T_a)$ determined from the known earliest-arrival delays; any passive object exceeding the bound would invalidate the determinant sum rule.

Watch

Extended reading notes

Core claim

The central claim is that causality sum rules can be written directly in the conventional incoming-outgoing scattering matrix, with no change of variables, provided one first corrects the apparent time advance imposed by the finite reference domain. With $T_a=\mathrm{diag}(\tau_\alpha)$ the channel-dependent earliest-arrival delays, the domain-delayed matrix $\widetilde S(\omega)=e^{i\omega T_a}S(\omega)$ preserves real-frequency passivity because $e^{i\omega T_a}$ is unitary, and, under the paper's analyticity, transparency, and regularity assumptions, is an operator Schur function: analytic in the upper half-plane and contractive there. The Cayley transform then supplies the Herglotz representation, whose low-frequency expansion is controlled by the delay operator. The paper's two results are the projected inequality $\int_0^\infty -\ln|\langle v,S(\omega)v\rangle|\,d\omega/\omega^2 \le \pi\langle v,T_a v\rangle$ for unit superpositions in a common-delay eigenspace, and the determinant inequality $\int_0^\infty -\ln|\det S(\omega)|\,d\omega/\omega^2 \le \pi\,\mathrm{tr}(T_a)$. These recover the Rozanov absorber thickness-bandwidth bound and the spherical-multipole bounds as scalar limits, and the multichannel forms constrain quantities no scalar bound can see.

Load-bearing premise

The load-bearing premise is that the domain-delayed matrix is actually a Schur function, analytic and contractive across the whole upper half-plane under realistic high-frequency behavior, so the logarithmic Herglotz moment identities hold; for the lossless delay result, an additional modal-count hypothesis on the phase accumulation is also load-bearing.

Editorial extensions

If this is right

A coherent superposition of spherical-wave channels obeys a size-limited suppression: with $T_a=(2a/c)I$, the integrated logarithmic return is at most $2\pi a/c$, so deep broadband suppression of any multiport combination costs electrical size.
The determinant bound is basis independent and grows linearly with the truncated channel number $N$, so the geometric-mean singular-value depth has a ceiling that does not grow with $N$.
At least $m$ singular channels cannot all stay below a threshold $\rho_\sigma$ over a fractional bandwidth $\beta$ unless $m\le N\pi k_0 a/(\beta\ln(1/\rho_\sigma))$, giving a countable maximum number of suppressed channels.
For lossless multiport networks, under the modal-count hypothesis, the band-averaged delay satisfies a bound proportional to $N$ and inversely to fractional bandwidth, quantifying how much effective propagation length is needed for a target delay-bandwidth product.
Because both rules operate on the measured S-matrix itself, experimental $S$-parameter data can be checked against fundamental limits without reconstructing polarizability, Green-function, or volume-operator quantities.

Reading between the lines

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

The same construction suggests a testable extension to non-spherical reference surfaces with channel-dependent delays: the projected rule should hold separately within each common-delay eigenspace, which could be checked in waveguide arrays or metasurface measurements.
Since the projected rule binds any superposition in a delay eigenspace, it gives a design pre-check for coherent perfect absorption or destructive-interference suppression: required attenuation-bandwidth product must be paid for by delay available inside that eigenspace.
A practical falsification route is to compute the determinant integral from measured broad-band S-parameters of a passive device; any excess over $\pi\,\mathrm{tr}(T_a)$ would show which transparency or regularity condition fails in a real scatterer.
The lossless delay-bandwidth estimate is conditional on phase winding accumulating in a modal-count way; for slow-light or high-Q resonant structures the paper's own caveat suggests the true bound may be larger by an additive $2\pi k_{\mathrm{extra}}$ term, so this part is better read as a spectral-counting estimate than a universal limit.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Good idea, wrong constant. The domain-delay construction—removing the reference-domain time advance by multiplying with D_a = e^{iωT_a} and then treating S̃ as a Schur function—is the right way to connect causal bounds to measured S-matrices, and the projected/determinant formulation is a plausible operator-level extension. But the central inequalities are off by a factor of two.

Check the scalar limit yourself. For s(ω) = −1/(1+iωτ), a perfectly good passive one-pole Schur function, −ln|s| = ½ ln(1+ω²τ²), so ∫_0^∞ −ln|s(ω)| dω/ω² = πτ/2. Equation (2) claims the bound is π⟨v,T_a v⟩, i.e. πτ for this scalar. That is a factor of two too large. The paper's Rozanov recovery inserts τ_v = 2d/c and gets 2πd/c, while the actual Rozanov constant is πd/c. So the "direct recovery" is not a recovery; the bound is loose by exactly two. The same factor scales the determinant rule and the finite-band consequences in Figs. 3 and 4.

What is genuinely good: the operator framing is new, the distinction between projected and determinant budgets is useful, and the suppressed-channel count is a nice application that does not exist in the cited literature. The authors also state their conditions (transparency, analyticity, angular-derivative) explicitly and are honest that the lossless delay-bandwidth part is conditional on a modal-count hypothesis. That is responsible.

Soft spots beyond the coefficient: the crucial lemma (S̃ is Schur) and both sum-rule proofs are deferred to Supplementary Notes 2–4, which are not in the reviewed text. Referees cannot check the main theorem from the manuscript. The finite-band conversion uses a narrowband approximation, which is fine if labeled; the lossless extension has a free parameter λ and an explicit extra-phase caveat, so it is not a universal limit. None of these is fatal; the factor of two is.

My recommendation: send it to peer review, but the referee must be asked to verify Eq. (2) with a one-pole example. If the coefficient is corrected to π/2 (or T_a redefined so that the product matches), the framework likely survives and the applications need rescaling. As written, the central quantitative claim is wrong, and the authors should not publish it without fixing that.

Referee Report

3 major / 3 minor

Summary. The manuscript proposes a formulation of causality sum rules directly in the conventional multichannel scattering matrix. The central construction is the domain-delayed matrix S̃(ω) = D_a(ω)S(ω), where D_a(ω) = exp(iωT_a) and T_a is the earliest-arrival delay operator determined by the reference surfaces. The paper asserts that, under stated analyticity, transparency, and regularity assumptions, S̃ is an operator Schur function, and that a Cayley-Herglotz construction yields a projected sum rule for coherent channel superpositions and a determinant sum rule for aggregate multichannel attenuation. These rules are claimed to recover Rozanov's absorber bound and spherical-multipole sum rules, and to yield finite-band depth-bandwidth constraints, aggregate and geometric-mean attenuation bounds, suppressed-channel-count bounds, and a conditional lossless delay-bandwidth estimate. The manuscript also describes the use of an AI system, Qiushi Engine, in the initial exploration, with the authors taking responsibility for verification.

Significance. If the central Schur-function lemma and the logarithmic moment identities are correct, the paper offers a valuable connection between experimentally accessible scattering data and fundamental causality limits. The construction is attractive because T_a is a geometric input rather than a fitted parameter, and the scalar limit checks are explicit and recover known results. The paper also clearly states several limitations, such as the need for common-delay eigenspaces for the projected rule and the requirement of channel completeness for an absorption interpretation. The significance is, however, conditional: the proofs of the central claims are deferred to a missing supplement, and the precise regularity conditions are not stated in the main text, so the substantive results cannot currently be verified from the submitted material.

major comments (3)

[Domain-delayed Schur construction; Lemma S2.1, Supplementary Notes 3–4] The central step—that S̃(ω)=D_a(ω)S(ω) is an operator Schur function and that the logarithmic Herglotz moment identities for ⟨v,S̃(ω)v⟩ and det S̃(ω) hold—is asserted in the main text, but its proof is deferred entirely to Lemma S2.1 and Supplementary Notes 3 and 4, which are not included in the reviewed preprint. Since both sum rules (2) and (3) rest on these identities, the manuscript is not self-contained. Please provide the supplementary notes, or include a complete proof and a precise statement of the required analyticity, high-frequency transparency, and regularity assumptions in the main text.
[Projected sum rule, Eq. (2)] The 'angular-derivative condition' is not defined in the main text. The text states only that it 'ensures that the non-geometric terms do not increase the low-frequency coefficient.' Without a precise definition, a referee cannot check whether the low-frequency coefficient of the Herglotz function is indeed proportional to ⟨v,T_a v⟩, nor whether the logarithmic integral converges. In particular, if ⟨v,S(ω)v⟩ or det S(ω) has a zero on the real axis, the integral ∫_0^∞ −ln|·| dω/ω² diverges, and the stated bound cannot hold as written; the conditions must exclude this case or provide a limiting interpretation.
[Conditional phase-delay extension, Eq. (6)] The lossless delay-bandwidth estimate (6) depends on an 'effective propagation length λ' and on the modal-count hypothesis B_H, neither of which is defined in the main text. As written, λ is a free parameter, so Eq. (6) is not a closed-form causal bound. The paper acknowledges that the result is conditional, but the hypothesis B_H should be stated precisely so that a reader can determine which systems satisfy it and how the 'effective propagation length' is related to the scattering geometry.

minor comments (3)

[Eq. (2) and Eq. (3)] The displayed equations are typeset ambiguously; the right-hand side coefficient appears as π/2 in the text, and the Rozanov recovery with τ_v=2d/c confirms that the intended coefficient is π/2. Please typeset the coefficient explicitly as π/2 in both Eq. (2) and Eq. (3) to avoid misreading.
[Consequence 1] The inline inequality '∫ −ln|s_v| dω/ω² ≤ π τ_v / 2' is formatted unclearly; the division by 2 is easily lost. Please rewrite this in displayed form.
[Notation] The Herglotz function h_v mentioned in the projected sum rule is introduced without a definition in the main text; please define it explicitly or remove the reference to it.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: T_a is a geometric input, the sum rules follow from deferred Schur-Herglotz lemmas, and the Rozanov/spherical recoveries are independent consistency checks.

full rationale

The derivation chain is self-contained with respect to circularity. Eq. (1) defines S̃ = D_a S with D_a = exp(iωT_a), and T_a = diag(τ_α) with τ_α the earliest arrival time (2a/c for spherical waves). No parameter is fitted to the scattering data being bounded; the right-hand sides of Eqs. (2) and (3) are determined by geometry, not by the measured response. The claim that S̃ is an operator Schur function is supported by Lemma S2.1 in the supplement under explicit analyticity, transparency, and regularity assumptions, rather than by citing the authors' prior work. The 'recoveries' of Rozanov's bound and the Bernland-Gustafsson spherical-multipole bounds are presented as consistency checks of the same construction and are not used as inputs to derive Eqs. (2)-(3). The only self-citation (Ref. 25, Qiushi Engine) concerns the discovery workflow and is not load-bearing for the physics. The reviewer's factor-of-two mismatch between Eq. (2) and the one-pole Schur check (πτ/2 versus πτ) is a potential correctness issue in the coefficient, not an instance of circularity; no equation in the paper reduces to its own input by construction.

Assumptions & free parameters 1 free parameters · 5 assumptions · 1 invented entities

The main sum rules depend on standard Schur/Herglotz theory, the physical assumptions of passivity and causality, the unproven-in-main-text regularity and transparency conditions, and an ad hoc angular-derivative condition. The lossless delay-bandwidth result depends on an additional modal-count hypothesis. No fitted parameters enter the central bounds; the delay operator is geometric.

free parameters (1)

Effective propagation length λ (lossless delay-bandwidth bound)
Enters the modal-count hypothesis B_H for the conditional delay-bandwidth estimate (Eq. 6). It is a modeling parameter, not fitted to data, but it is ad hoc to the paper.

assumptions (5)

domain assumption The physical system is passive and causal, so the domain-delayed scattering matrix S̃ = D_a S is an operator Schur function (analytic and contractive in the upper half-plane).
Stated in the Main Text after Eq. (1). The proof is delegated to Lemma S2.1 in Supplementary Note 2, not available in the reviewed text.
domain assumption Appropriate high-frequency transparency and regularity conditions hold, ensuring the logarithmic Schur/Herglotz representation and convergence of the integrals.
Stated in Main Text: 'The construction assumes causal response, appropriate high-frequency transparency, and the regularity conditions required for the logarithmic Schur representation.'
ad hoc to paper The angular-derivative condition holds at zero frequency, preventing non-geometric terms from increasing the low-frequency coefficient.
Mentioned in the projected sum rule section: 'the angular-derivative condition, which ensures that the non-geometric terms do not increase the low-frequency coefficient'. Not defined in the main text.
ad hoc to paper For the lossless delay-bandwidth bound, the modal-count hypothesis B_H holds, meaning the delay-corrected determinant has no singular inner factor and its Blaschke phase accumulation matches the modal count.
Stated in the main text: 'If the delay-corrected determinant contains no singular inner factor and its Blaschke phase accumulation satisfies the modal-count hypothesis BH (see details in Supplementary Note 8)...' This is an additional assumption for the conditional extension.
standard math Standard Schur-function and Herglotz-representation theory, including the Cayley transform and logarithmic moment identities.
Inherited background used across the paper.

invented entities (1)

Domain-delay operator D_a(ω) = exp(iωT_a)
purpose: Unitary phase factor that removes the reference-domain time advance from the scattering matrix to restore causal analytic structure.
A mathematical construction, not a physical entity. Its validity is internal to the derivation; it has no falsifiable handle outside the paper.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Causality Sum Rules in Conventional Scattering Matrices." pith.science (2026). https://pith.science/paper/DESXGYBD

@misc{pith2026260810427,
  author       = {Pith},
  title        = {Pith review of: Causality Sum Rules in Conventional Scattering Matrices},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DESXGYBD}},
  note         = {Machine review of arXiv:2608.10427}
}

read the original abstract

Scattering matrices are the standard experimental and computational description of photonic and electromagnetic devices. Passivity is explicit in the conventional incoming-outgoing matrix, whereas causality sum rules are usually formulated only after transforming the response into auxiliary variables. Here we show that these rules can be written directly in the conventional scattering matrix by removing the time advance introduced by the reference domain. Using the earliest-arrival delay of each channel, we define a domain-delayed matrix that preserves real-frequency passivity while restoring the causal time origin. Under explicit analyticity, transparency, and regularity assumptions, this matrix becomes a Schur function, enabling a Cayley-Herglotz construction. The resulting projected and determinant bounds constrain coherent channel superpositions and aggregate multichannel loss. The framework recovers Rozanov's absorber limit and spherical-multipole sum rules, while extending causality bounds to measurable quantities including insertion loss, suppressed singular-value channels, and conditional lossless delay-bandwidth trade-offs. Our work directly connects fundamental causality theory with experimentally accessible scattering data. The initial theoretical route is autonomously explored by Qiushi Engine, an AI research system for open-ended scientific discovery, and subsequently verified, refined, and developed by the authors, demonstrating a hybrid AI-human discovery workflow.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

25 extracted references · 23 canonical work pages

[1]

Kramers, H. A. La diffusion de la lumière par les atomes. Atti Congr. Internazionale Dei Fis. 2, 545-557 (1927)

work page 1927
[2]

Kronig, R. de L. On the theory of dispersion of X -rays. J. Opt. Soc. Am. 12, 547- 557 (1926)

work page 1926
[3]

Nussenzveig, H. M. Causality and Dispersion Relations . (Academic Press, New York, 1972)

work page 1972
[4]

Foster, R. M. A reactance theorem. Bell Syst. Tech. J. 3, 259-267 (1924)

work page 1924
[5]

Bode, H. W. Network Analysis and Feedback Amplifier Design. (Van Nostrand, New York, 1945)

work page 1945
[6]

Fano, R. M. Theoretical limitations on the broadband matching of arbitrary impedances. J. Frankl. Inst. 249, 57-83 (1950)

work page 1950
[7]

C., Castriota, L

Youla, D. C., Castriota, L. J. & Carlin, H. J. Bounded real scattering matrices and the foundations of linear passive network theory. IRE Trans. Circuit Theory 6, 102- 124 (1959)

work page 1959
[8]

Rozanov, K. N. Ultimate thickness to bandwidth ratio of radar absorbers. IEEE Trans. Antennas Propag. 48, 1230-1234 (2000)

work page 2000

Show all 25 references

[9]

& Nordebo, S

Bernland, A., Gustafsson, M. & Nordebo, S. Physical limitations on the scattering of electromagnetic vector spherical waves. J. Phys. Math. Theor. 44, 145401 (2011)

2011
[10]

Bandwidth limitations for scattering of higher order electromagnetic spherical waves with implications for the antenna scattering matrix

Bernland, A. Bandwidth limitations for scattering of higher order electromagnetic spherical waves with implications for the antenna scattering matrix. IEEE Trans. Antennas Propag. 60, 4345-4353 (2012)

2012
[11]

& Gustafsson, M

Bernland, A., Luger, A. & Gustafsson, M. Sum rules and constraints on passive systems. J. Phys. Math. Theor. 44, 145205 (2011)

2011
[12]

& Kristensson, G

Sohl, C., Gustafsson, M. & Kristensson, G. Physical limitations on broadband scattering by heterogeneous obstacles. J. Phys. Math. Theor. 40, 11165- 11182 (2007)

2007
[13]

& Kristensson, G

Sohl, C., Larsson, C., Gustafsson, M. & Kristensson, G. A scattering and absorption identity for metamaterials: experimental results and comparison with theory. J. Appl. Phys. 103, 054906 (2008)

2008
[14]

& Sjöberg, D

Gustafsson, M., Sohl, C., Larsson, C. & Sjöberg, D. Physical bounds on the all - spectrum transmission through periodic arrays. EPL 87, 34002 (2009)

2009
[15]

& Greffet, J.-J

Biehs, S.-A., Rousseau, E. & Greffet, J.-J. Mesoscopic description of radiative heat transfer at the nanoscale. Phys. Rev. Lett. 105, 234301 (2010)

2010
[16]

Gustafsson, M., Vakili, I., Keskin, S. E. B., Sjöberg, D. & Larsson, C. Optical theorem and forward scattering sum rule for periodic structures. IEEE Trans. Antennas Propag. 60, 3818-3826 (2012)

2012
[17]

D., Johnson, S

Miller, O. D., Johnson, S. G. & Rodriguez, A. W. Shape-independent limits to near- field radiative heat transfer. Phys. Rev. Lett. 115, 204302 (2015)

2015
[18]

Shim, H., Fan, L., Johnson, S. G. & Miller, O. D. Fundamental limits to near-field optical response over any bandwidth. Phys. Rev. X 9, 011043 (2019)

2019
[19]

Molesky, S., Jin, W., Venkataram, P. S. & Rodriguez, A. W. T-operator bounds on angle-integrated absorption and thermal radiation for arbitrary objects. Phys. Rev. Lett. 123, 257401 (2019)

2019
[20]

& Miller, O

Kuang, Z. & Miller, O. D. Computational bounds to light –matter interactions via local conservation laws. Phys. Rev. Lett. 125, 263607 (2020)

2020
[21]

& Rodriguez, A

Molesky, S., Chao, P., Jin, W. & Rodriguez, A. W. Global T operator bounds on electromagnetic scattering: upper bounds on far-field cross sections. Phys. Rev. Res. 2, 033172 (2020)

2020
[22]

& Rodriguez, A

Molesky, S., Chao, P., Mohajan, J., Reinhart, W., Chi, H. & Rodriguez, A. W. T - operator limits on optical communication: metaoptics, computation, and input - output transformations. Phys. Rev. Res. 4, 013020 (2022)

2022
[23]

& Miller, O

Zhang, L., Monticone, F. & Miller, O. D. All electromagnetic scattering bodies are matrix-valued oscillators. Nat. Commun. 14, 7724 (2023)

2023
[24]

O. D. Miller and F. Monticone, Fundamental limits in photonics and electromagnetics: a tutorial, arXiv:2605.24738, 2026

2026 arXiv
[25]

End- to-end autonomous scientific discovery on a real optical platform, arXiv:2604.27092 (2026)

Yang, S., et al. End- to-end autonomous scientific discovery on a real optical platform, arXiv:2604.27092 (2026)

2026 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Kramers, H. A. La diffusion de la lumière par les atomes. Atti Congr. Internazionale Dei Fis. 2, 545-557 (1927)

work page 1927

[2] [2]

Kronig, R. de L. On the theory of dispersion of X -rays. J. Opt. Soc. Am. 12, 547- 557 (1926)

work page 1926

[3] [3]

Nussenzveig, H. M. Causality and Dispersion Relations . (Academic Press, New York, 1972)

work page 1972

[4] [4]

Foster, R. M. A reactance theorem. Bell Syst. Tech. J. 3, 259-267 (1924)

work page 1924

[5] [5]

Bode, H. W. Network Analysis and Feedback Amplifier Design. (Van Nostrand, New York, 1945)

work page 1945

[6] [6]

Fano, R. M. Theoretical limitations on the broadband matching of arbitrary impedances. J. Frankl. Inst. 249, 57-83 (1950)

work page 1950

[7] [7]

C., Castriota, L

Youla, D. C., Castriota, L. J. & Carlin, H. J. Bounded real scattering matrices and the foundations of linear passive network theory. IRE Trans. Circuit Theory 6, 102- 124 (1959)

work page 1959

[8] [8]

Rozanov, K. N. Ultimate thickness to bandwidth ratio of radar absorbers. IEEE Trans. Antennas Propag. 48, 1230-1234 (2000)

work page 2000