{"id":"cff4d5d1-f74a-426f-aa9a-dcd6008d433f","arxiv_id":"2608.09657","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Correlating Wigner functions of the received fields yields a coherence pattern with roughly twice the spatial frequency content of the standard visibility, pointing to a possible two-fold angular resolution gain.","lead":"Radio interferometry usually correlates the electric field at pairs of telescopes. This paper instead correlates the Wigner time-frequency representations of those fields and finds the resulting pattern is about twice as narrow, suggesting a possible factor-of-two angular resolution gain.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The factor-of-two pattern rests on an unproven equivalence between the band regularization in Eqs. (18)-(20) and the finite-time Wigner correlation that an experiment would actually measure; without that equivalence the central claim is not tied to an observable.","rationale":"The paper makes a concrete, falsifiable claim and provides analytic solutions plus open-source simulations, so the central idea deserves serious consideration. The most load-bearing weakness is not the internal algebra but the physical status of the regularized Wigner correlation defined by Eqs. (18)-(20). The original integral in Eq. (4) diverges for stationary fields because the surviving fourth-order pairings depend on xi+eta, leaving an unbounded integration over eta-xi. The paper chooses one particular way to cut off and normalize that divergence, but it does not show that this choice reproduces the finite-time, time-averaged quantity that a real measurement, or even the paper's own Sec. 7.3 estimator, would produce. Since the entire factor-of-two appears through the phase in Eq. (69) after this regularization, an unjustified normalization would make the headline claim an artifact rather than a property of a physical observable. The downstream delta approximation in Eq. (60) is also uncontrolled, but it is secondary: if the regularization is not physically meaningful, the delta approximation is moot, and if the regularization is validated, the delta approximation can be tested separately. This is exactly the weakest assumption identified by the reader, so my read does not change the conditional verdict. A finite-time derivation and comparison would settle the concern: if the regularized J' equals the long-time limit of the actual Wigner-correlation estimator, the factor-of-two stands; if not, the central resolution claim is unverified.","tokens_in":20668,"tokens_out":15230,"duration_ms":155445,"concrete_test":"Derive the exact finite-time Wigner correlation for a two-point source: define W_T(t,omega) with lags restricted to |xi|,|eta|<=T and form the time average (1/T) integral W_1(t)W_2(t) dt. Take the T->infinity limit after the normalization used in Sec. 7.3 and compare the resulting baseline-dependent function with J' from Eq. (20). If the finite-time limit differs in its p,q dependence, or if the limit depends on the shape of the lag window, then the factor-of-two in Eq. (69) is a regularization artifact and the central claim is not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Equation (4) defines the Wigner correlation as an ensemble average of products of two infinite-lag Wigner integrals. For a stationary Gaussian source, the surviving Wick pairings, such as the product in Eq. (28), depend on s=xi+eta, so the double integral contains an unconstrained integration over d=eta-xi and diverges. Equations (18)-(20) replace this divergence with a band-limited, width-normalized limit J'. That replacement is a convention, not a consequence of Eq. (4). The paper does not prove that this J' equals the finite-time correlation actually estimated in Sec. 7.3, where discrete Wigner distributions are computed and their products are time-averaged. Since the doubled spatial frequency enters through the phase 2k(p chi + q psi) in Eq. (69) only after this regularization and after the subsequent delta approximation K(a) approx K'(a) delta(alpha+beta) in Eq. (60), an unphysical or window-dependent regularization would make the headline factor-of-two an artifact. The simulation agreement in Fig. 3 is supportive, but the simulation estimator is itself a finite-window regularization whose equivalence to Eqs. (18)-(20) is not demonstrated.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces \"Wigner interferometry,\" a proposed measurement technique in which one correlates the Wigner time-frequency distributions of the fields received at two points rather than correlating the fields themselves. For a stationary, zero-mean Gaussian source, the author uses the Isserlis theorem to decompose the fourth-order Wigner correlation into cross-product and self-product contributions. A divergent double integral is regularized by a width-normalized band limit, and the cross-product kernel is then approximated by a delta function localized at symmetric source points. The resulting integral, Eq. (69), contains the phase exp(i 2k(pχ+qψ)) instead of the classical exp(ik(pχ+qψ)), implying that the Wigner coherence pattern samples twice the spatial frequency and has about half the scale of the classical van Cittert-Zernike pattern. Exact analytical solutions for rectangular and circular sources and conceptual numerical simulations are presented, and the author concludes that Wigner interferometry could improve angular resolution by a factor of two.","tokens_in":20951,"tokens_out":16522,"duration_ms":151358,"significance":"If the central claim is correct, the paper opens a conceptually new route to increased angular resolution in interferometry without increasing baselines, which would be of substantial interest to radio and optical interferometry communities. The manuscript is clearly organized and has several strengths: the starting point is a standard Gaussian description of radiation, the factor-two effect emerges from a clean Fourier integral, an openly available simulation code supports the analytical result, and the comparison against classical coherence and its square is a useful sanity check. However, the significance is conditional on the validity of two nonstandard steps: the regularization of the divergent Wigner-correlation integral and the delta-function approximation in Eq. (60). Both steps are load-bearing for the factor-two claim and are not currently justified with controlled errors or a direct link to the simulation estimator.","major_comments":[{"comment":"The regularization step is internally inconsistent and its physical counterpart is not defined. Equation (17) places the Wigner phase as e^{-iωη'} after the change of variables, whereas the correct transformation of e^{-iωξ}e^{-iωη} gives e^{-iωξ'}; Eq. (18) then drops the phase factor altogether. In addition, the normalization in Eq. (20) does not follow from Eq. (19) with the stated change of variables: the Jacobian of (ξ,η) to (ξ',η') is 1/2, so the normalized band limit in Eq. (20) should contain an extra factor of 1/4 (or 1/2, depending on the intended band width). The phase e^{-iωξ'} is not a harmless detail: it reappears in the definition of K(a) in Eq. (57) and is responsible for the factor e^{-i2ωα} that ultimately produces the factor 2 in Eq. (69). The paper must carry the Wigner phase through the regularization consistently and must state precisely which finite-time estimator corresponds to the normalized limit J'.","section":"Section 3, Eqs. (17)-(20)"},{"comment":"The delta-function approximation K(a) ≈ K'(a) δ(α+β) is introduced without an error estimate. The exact K(a) in Eq. (58) is a principal-value integral with a first-order pole at α+β=0 and oscillatory structure away from the pole; replacing it by a delta function eliminates all non-symmetric source-point pairs. The derivation of L(ω) in Eq. (69) and, hence, the factor-two claim depend directly on this substitution. The author should either bound the correction from the non-delta part of K(a), or evaluate the source integral with the exact K(a) (at least numerically) to demonstrate that the symmetric-point contribution indeed dominates to the required accuracy.","section":"Section 5.1, Eq. (60)"},{"comment":"The abstract and conclusions state that the Wigner correlation pattern is about twice smaller in scale than the classical pattern, but the explicit factor-two result is demonstrated only for the cross-product contribution W_ab. The self-product contribution W_aa (Eqs. 79-105) is governed by integrals m_n(a,b,c,d) that may have a different spatial-frequency content. The paper should show analytically or numerically that the total W_ab + W_aa indeed has a main lobe about twice narrower than the classical coherence pattern, and should quantify the relative amplitude of W_aa so that the total pattern is not dominated by a broader component.","section":"Sections 5.2 and 6"},{"comment":"The numerical estimator is described only as selecting the temporal variations of the Wigner distributions at a given frequency, multiplying them for different baselines, and summing the products. Since the analytical theory rests on a particular regularization of a divergent integral, the paper must specify the discrete Wigner window, the normalization, and the exact operation performed in the time average, and show that this estimator converges to the regularized J'_ab and J'_aa used in the theory. Without this link, the agreement in Fig. 3, while suggestive, cannot be used to validate the two nonstandard analytical approximations.","section":"Section 7.3"}],"minor_comments":[{"comment":"The expansion of Eq. (10) contains sixteen terms, not twelve as stated; the first group of four plus the remaining three groups of four gives sixteen.","section":"Section 3, text before Eq. (11)"},{"comment":"The title reads \"Generalization to a continues source\"; \"continues\" should be \"continuous\".","section":"Section 4 title"},{"comment":"Beyond the substantive issue above, the displayed formulas contain typographical inconsistencies that should be corrected: the phase variable in Eq. (17) appears as η' and should be ξ', and the normalization factors in Eqs. (19) and (20) disagree with the stated Jacobian.","section":"Eqs. (17)-(20)"},{"comment":"The carrier frequency is denoted ω in Eqs. (31)-(32) but ω0 in the band-limited spectrum definition (38)-(39); unify the notation to avoid confusion.","section":"Eqs. (31), (38), (39)"},{"comment":"The simulation points are plotted without error bars or any description of the number of realizations, the source grid, the baseline sampling, and the time-series length; please add these details so the reader can assess the scatter and the independence of the simulation.","section":"Fig. 3"},{"comment":"The reference to a similar effect for the Unterberger time-frequency distribution is made without a citation; a reference and a brief explanation would be helpful.","section":"Section 8, footnote 3"},{"comment":"The statement that the simulation code is openly available is good practice; please cite the repository version and provide the exact commands or configuration used to generate Fig. 3.","section":"Section 7"}],"recommendation":"major_revision","confidential_remarks":"This manuscript presents an intriguing idea with plausible numerics, but the central derivation currently rests on two unproven approximations: the link between the band regularization and the actually measured finite-time Wigner correlation, and the delta-function replacement of the cross-product kernel. I would encourage the editor to request a revision in which these two steps are made rigorous or at least quantitatively controlled, and in which the total pattern (W_ab + W_aa) is shown explicitly to have the claimed factor-two scale. If the author can supply these ingredients, the paper would be a valuable methods contribution to interferometry; in its present form the theoretical demonstration is not yet compelling."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This is a genuinely new idea, and I think the reader's conditional verdict is about right. Correlating Wigner functions instead of fields is a fresh observable, and the factor-of-two in spatial frequency is derived from first principles with no free parameters. The exact analytic solutions for rectangular and circular sources and the open-source simulations are real assets. The paper also deals honestly with its own scope: it admits the missing image reconstruction routine and stops short of noise or sensitivity analysis.\n\nThe soft spots are exactly where the reader puts them. The regularization in Eqs. (17)-(20) is a convention, not a consequence of Eq. (4). That is not fatal by itself—some normalization of an infinite-lag product is needed—but the paper never proves that this band-width-normalized limit equals the finite-window estimator used in Sec. 7.3. The simulation agreement in Fig. 3 therefore supports the theory only conditionally. The delta-function approximation K(a) ≈ K'(a) δ(α+β) in Eq. (60) is more concerning. It confines the cross-product contribution to symmetric source points and is the step that produces the 2k-phase in L(ω). There is no error bound, and without it the double-source integral could behave differently. A skeptical referee could reasonably demand a controlled asymptotic argument or numerical convergence check for this approximation.\n\nOn the citation pattern: nothing suspicious. The prior work cited covers standard interferometry, intensity interferometry, and time-frequency analysis; the claim of novelty relative to that literature holds up.\n\nWho gets value from this? Interferometrists (radio and optical) and time-frequency specialists will find the idea provocative and worth engaging. The paper is not ready to be taken as proof of the factor-of-two, but it is far from a crank document. The thinking is clear and the limitations are stated rather than hidden.\n\nMy recommendation: a serious editor should send this to peer review, not desk-reject it. A good referee would ask for a rigorous derivation of the regularization's equivalence to a measurable time average, and an error-controlled treatment of the delta approximation (or a numerical convergence study). With those additions, the work could be publishable; in current form it is a promising but unproven proposal.","headline":"Wigner interferometry proposes a clever new observable that may double interferometric resolution, but the factor-of-two rests on an unchecked regularization and a delta-function approximation; a serious referee could fix it, and the idea deserves that shot.","tokens_in":21373,"tokens_out":3580,"would_cite":false,"duration_ms":32555,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Correlating the Wigner functions of the fields received at two points yields a coherence pattern that samples spatial frequencies twice as high as the standard field correlation, opening a route to double the angular resolution of…","keywords":["Wigner interferometry","time-frequency distributions","radio interferometry","angular resolution","coherence function","van Cittert-Zernike theorem","intensity interferometry","Gaussian random process"],"falsifier":"Measure the Wigner correlation from two independent noise sources at two detectors as a function of baseline and compare the spacing of its zeros (or the half-width of its central lobe) with the classical field-coherence pattern obtained from the same time series at the same frequency; if the Wigner spacing is not about half the classical spacing, the central claim fails. For a uniform rectangular source the paper's Eq. (107) predicts zeros at $2kp\\chi_0=m\\pi$, exactly half the baseline spacing of the classical $\\sin(kp\\chi_0)/(kp)$ pattern.","tokens_in":20477,"feed_emoji":"🔭","tokens_out":13961,"duration_ms":124493,"temperature":0.7,"pith_summary":"The paper introduces Wigner interferometry: rather than correlating the fields at two antennas, it constructs the Wigner time-frequency distribution of each recorded field and correlates those distributions. Under the assumptions of stationary, zero-mean, Gaussian, band-limited noise from an incoherent source, the paper shows that the Wigner correlation is dominated by a term that is the Fourier transform of $I(\\chi,\\psi)I(-\\chi,-\\psi)$ at twice the spatial frequency, $e^{i2k(p\\chi+q\\psi)}$ instead of $e^{ik(p\\chi+q\\psi)}$. The resulting coherence pattern is roughly half the scale of the classical van Cittert-Zernike pattern, which is what the analytical solutions for uniform square and circular sources and the numerical simulations show. For a fixed baseline, this means an interferometer could in principle resolve angular scales two times smaller than the classical diffraction limit would allow.","feed_headline":"Wigner correlation doubles interferometric angular resolution","feed_subtitle":"Correlating Wigner time-frequency maps samples twice the spatial frequency of the classic coherence pattern.","key_machinery":"The central object is the Wigner correlation function $W(P_1,P_2)=\\int\\!\\int\\langle V_1(t+\\xi/2)V_1(t-\\xi/2)V_2(t+\\eta/2)V_2(t-\\eta/2)\\rangle e^{-i\\omega\\xi}e^{-i\\omega\\eta}\\,d\\xi d\\eta$, i.e. the time average of the product of the Wigner time-frequency distributions (joint time-frequency representations) of the fields at the two antennas. The argument is carried by four ingredients: Isserlis's theorem factorises the Gaussian fourth-order moment into pairwise correlations; source incoherence kills all terms except those pairing each source point with itself or with its symmetric counterpart; the surviving divergent double integrals are regularised by normalising to the width of the integration band and taking the limit; and the approximation $K(a)\\approx K'(a)\\delta(\\alpha+\\beta)$ selects only axially symmetric source points in the cross-product channel. The factor of two appears because the geometric phase $R_{a1}-R_{a2}\\approx\\zeta-(p\\chi+q\\psi)$ enters the exponent twice, once from each of the two correlation functions in the product.","core_discovery":"The central discovery is that the Wigner correlation function carries twice the spatial-frequency content of the ordinary field coherence function. Specifically, for a source with intensity distribution $I(\\chi,\\psi)$, the cross-product contribution reduces, after the divergent double integral is regularised by the band-width normalisation of Eqs. (17)-(20) and with the delta-function approximation $K(a)\\approx K'(a)\\delta(\\alpha+\\beta)$, to $L(\\omega)=\\int_\\sigma I(\\chi,\\psi)I(-\\chi,-\\psi)e^{i2k(p\\chi+q\\psi)}\\,d\\chi d\\psi$. The factor of two in the exponent means that a given baseline samples twice the spatial frequency sampled by the van Cittert-Zernike integral $\\int I(\\chi,\\psi)e^{ik(p\\chi+q\\psi)}\\,d\\chi d\\psi$. The self-product contribution, integrated with the kernel $m_n(a,b,c,d)$, adds a second term of the same physical scale. The paper obtains closed-form Wigner patterns for uniform rectangular and circular sources, reproduces them numerically, and interprets the factor-of-two as a doubling of the angular resolution available to interferometric imaging.","pith_inferences":["A bench-top experiment with two phase-uncorrelated noise sources would provide a decisive test: record the two time series, compute Wigner functions, and check whether the measured correlation's zero spacing is half that of the ordinary field correlation.","The cross-product term's coupling of only symmetric source points suggests that recovering a general brightness distribution will require combining the cross-product and self-product terms; the paper does not spell out such an inversion algorithm.","Because the derivation starts from generic spherical waves rather than radio-specific assumptions, the same factor-of-two Wigner effect may transfer to optical intensity-interferometry setups or even quantum fields, though the paper leaves that extension undeveloped."],"forward_implications":["A fixed maximum baseline would deliver twice the spatial-frequency content, so existing interferometers could image at half their current diffraction-limited angular scale.","Because the new observable is a fourth-order field correlation rather than the square of the second-order coherence, it does not reduce to intensity interferometry and carries information the classical pattern does not.","The factor-of-two scaling holds for the two exactly solved source shapes (uniform rectangle and uniform disk), and the numerical simulations reproduce the analytical patterns.","Practical use will require an inversion method that turns sampled Wigner correlations into sky images, and the paper argues there is no fundamental obstacle to building one.","The band-limited white-spectrum assumption can be achieved observationally by whitening the received radiation, so the method is tied to a realisable signal-processing step rather than an idealised spectrum."],"supporting_citations":[{"why":"It supplies the standard field-coherence relation that defines the diffraction limit and serves as the baseline against which the Wigner correlation is compared.","marker":"Thompson et al. 2017"},{"why":"It gives the van Cittert-Zernike theorem and the classical diffraction integrals for rectangular and circular apertures that the new pattern is compared with.","marker":"Born & Wolf 2019"},{"why":"It provides the intensity-interferometry result used to show that the intensity correlation carries no extra spatial frequency.","marker":"Hanbury Brown 1974"},{"why":"It supplies the Gaussian fourth-moment factorisation used to expand the Wigner fourth-order correlation into pairwise products.","marker":"Isserlis 1918"},{"why":"It provides the autocorrelation $R(\\tau)=\\sin(B\\tau)/(B\\tau)\\,\\cos(\\omega_0\\tau)$ for band-limited white noise used throughout the analytic solutions.","marker":"Bendat & Piersol 2010"},{"why":"It gives the complex-exponential treatment of averaged cosine products used in deriving the pairwise correlation forms.","marker":"Landau & Lifshitz 1975"},{"why":"It supplies the Bessel and Struve function identities used to close the circular-source integral $m_n$ analytically.","marker":"Prudnikov et al. 1998b"},{"why":"It supports the assumption that synchrotron radio emission behaves as a Gaussian random process, on which the Isserlis expansion rests.","marker":"Bénard & Rousseau 1974"}],"fun_headline_variants":["Wigner correlation doubles spatial-frequency coverage","Factor-of-two resolution gain from Wigner interferometry","Wigner-function correlation samples 2x finer detail","Double the resolution: Wigner-based interferometry","Wigner interferometry: same baseline, twice the detail"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the divergent double integral defining the Wigner correlation can be regularised by dividing by the width of the integration band and taking the limit, and that this regularised quantity is what a physical time-averaged Wigner-correlation measurement actually records.","fun_headline_variants_meta":{"raw":{"variants":["Wigner correlation doubles spatial-frequency coverage","Factor-of-two resolution gain from Wigner interferometry","Wigner-function correlation samples 2x finer detail","Double the resolution: Wigner-based interferometry","Wigner interferometry: same baseline, twice the detail"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00051,"raw_usage":{"total_tokens":2434,"prompt_tokens":852,"completion_tokens":1582,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":468,"completion_tokens_details":{"reasoning_tokens":1509}},"tokens_in":468,"tokens_out":1582,"duration_ms":11967,"temperature":1.0,"reasoning_tokens":1509,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T13:17:36.772235+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Measure the Wigner correlation from two independent noise sources at two detectors as a function of baseline and compare the spacing of its zeros (or the half-width of its central lobe) with the classical field-coherence pattern obtained from the same time series at the same frequency; if the Wigner spacing is not about half the classical spacing, the central claim fails. For a uniform rectangular source the paper's Eq. (107) predicts zeros at $2kp\\chi_0=m\\pi$, exactly half the baseline spacing of the classical $\\sin(kp\\chi_0)/(kp)$ pattern.","supporting_citations":[],"review_version":1}