Pith. sign in

REVIEW 1 major objections 6 minor 36 references

Optimal filtering for CMB lensing reconstruction

T0 review · 1 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

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

Pith's one-line read For CMB lensing reconstruction from maps with anisotropic noise, optimally filtering the input CMB maps and then Wiener-filtering the reconstructed convergence map cuts lensing power-spectrum variance by a factor of 2–5, with a further…

desk verdict Solid methods paper: the new kappa-filtering step is real and well tested, but the headline factor-2-5 variance gain is a full-pipeline comparison that mostly reflects extra sky area, not the filter alone. read the letter →

arxiv 1909.02653 v1 pith:NPYRKLAC submitted 2019-09-05 astro-ph.CO

Mark Mirmelstein , Julien Carron , Antony Lewis This is my paper · ORCID
classification astro-ph.CO
keywords CMBlensingreconstructionanisotropicnoiseinverse-variancefilteringquadraticestimatorpowerspectrumWienerfilterpatchapproximationconjugategradient
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

This paper is about how to extract the cosmic-microwave-background lensing signal from maps with uneven, cut-sky noise. The authors claim that quadratic-estimator lensing reconstruction should use two filtering steps: an inverse-variance filter on the input CMB maps, solved by conjugate gradients, and then a Wiener-like filter applied to the reconstructed convergence map. Compared with the simple isotropic filters used in earlier analyses, the first step lowers the reconstructed lensing spectrum variance by a factor of 2–5 for Simons Observatory- and CMB Stage-4-like noise, and the second step reduces it further by up to about 30% on large scales. An analytic patch model for the anisotropic response keeps the pipeline unbiased at the percent level and fast enough to run on hundreds of simulations.

What carries the argument

The load-bearing machinery is the combination of the inverse-variance CMB map filter and the independent-patch approximation. The filter is the solution of $(\mathbf{b}C_{\mathrm{fid}}\mathbf{b}^{\top}+\mathbf{N})^{-1}X$, computed iteratively with a conjugate-gradient solver, which automatically down-weights masked and high-noise pixels. The patch approximation splits the sky into regions of nearly constant noise, computes the local quadratic-estimator response $R^p_L$ and reconstruction noise $N^p_{0,L}$ analytically in each patch, and assembles them into the normalization factor $f_{A,L}$ and the effective white noise map $N^{\kappa}_{0,\mathrm{eff}}(x)$. This patch model motivates and calibrates the second-stage Wiener filter $C^{\kappa\kappa}_{\mathrm{fid}}(C^{\kappa\kappa}_{\mathrm{fid}}+N^{\kappa}_{0,\mathrm{eff}})^{-1}$, which converts an inverse-noise-weighted map that is optimal on small scales into one that also correctly weights large-scale signal-dominated modes.

What would settle it

Run the identical pipeline on a simulated scan whose hit-count map varies sharply, for example narrow stripes or many small disconnected patches, and compare the analytic patch normalization $f_{A,L}$ with the full Monte Carlo normalization: a deviation larger than a few percent in the multipole range $100\lesssim L\lesssim 1500$, or the disappearance of the factor 2–5 variance gain relative to isotropic filtering, would show that the quasi-locality assumption is not satisfied for that noise pattern.

Watch

Extended reading notes

Core claim

The central discovery is that a near-optimal, still-quadratic lensing pipeline for anisotropic noise is a two-stage filter. Stage one applies the optimal inverse-signal-plus-noise filter to the observed CMB maps, using a preconditioned conjugate-gradient inversion rather than an isotropic harmonic-space filter. Stage two takes the resulting quadratic-estimator convergence map, locally normalizes it through an analytic patch response, and applies the Wiener filter $C^{\kappa\kappa}_{\mathrm{fid}}(C^{\kappa\kappa}_{\mathrm{fid}}+N^{\kappa}_{0,\mathrm{eff}})^{-1}$ using a slowly varying effective reconstruction-noise map. On SO- and S4-like simulations the paper finds a factor 2–5 variance reduction from stage one relative to isotropic masked or weighted filtering, and up to roughly 30% further variance reduction from stage two on large scales, with the analytic response model matching the Monte Carlo normalization to within a few percent.

Load-bearing premise

The analytic response model and the second filter rest on the assumption that lensing convergence estimators are quasi-local, so that the sky can be treated as a patchwork of independent regions of nearly constant noise; if the noise varies on scales comparable to the estimator's real-space support, the analytic normalization and the reported gains would have to be revalidated by simulation.

Editorial extensions

If this is right

  • For SO- and S4-like noise and the scan pattern tested, replacing isotropic filtered maps with optimally inverse-variance filtered maps lowers the reconstructed lensing spectrum variance by a factor of 2–5 over most of the analysed multipole range.
  • Adding the approximate convergence ($\kappa$) filtering stage reduces large-scale lensing power variance by up to about 30%, in line with the analytic patch prediction, at modest extra numerical cost.
  • The analytic patch response model leaves only a small percent-level Monte Carlo correction, so the optimized quadratic-estimator pipeline stays fast enough to run over many simulated skies.
  • Polarization-only reconstructions benefit most from the second filtering stage because their reconstruction noise is more anisotropic; temperature reconstructions, being more signal-dominated, benefit less.
  • The reconstructed convergence maps after the second filter are more correlated with the true lensing field on large scales than maps from optimal CMB filtering alone, which matters for applications that use the maps rather than just their spectra.

Reading between the lines

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

  • The analytic variance expressions could be used before observing to rank candidate scan strategies: compute the expected variance reduction from the hit-count map alone and choose the scan whose noise pattern gives the largest gain.
  • Because the second stage improves signal-dominated large-scale modes, the filtered convergence maps should also be more effective for CMB delensing, an application the paper mentions but does not quantify.
  • The reported gains are lower bounds relative to a fully optimal iterative maximum-likelihood estimator, which the paper notes could do still better in the high-signal-to-noise regime; the practical value of the present pipeline is that it stays quadratic and fast.
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

1 major / 6 minor

Summary. This paper develops and tests an optimized quadratic-estimator pipeline for CMB lensing reconstruction under anisotropic noise. The first filtering stage applies inverse-signal-plus-noise (optimal anisotropic) filtering to the input CMB maps via conjugate-gradient inversion; the second stage approximately normalizes the reconstructed convergence map with a local effective response and applies a Wiener-like filter using a pixel-dependent effective reconstruction noise. Bias subtraction uses paired mean-field estimates and 500 Monte Carlo simulations for the N0 and N1 contractions, and the analytic 'patch' model is used to compute the response normalization and to predict variance improvements. On flat-sky SO- and S4-like simulations, the paper reports a factor of 2-5 variance reduction relative to isotropic masked/weighted baselines and an additional reduction of up to about 30% from the κ-filtering step on large scales.

Significance. If the reported gains are robust, the paper is practically important for upcoming ground-based lensing analyses. Its strengths are substantial: the simulation pipeline is unusually complete (500 lensed simulations, independent mean-field pairs, Monte Carlo N0 and N1 debiasing, and documented convergence-tolerance tests in Appendix A); the analytic patch normalization is explicitly compared with Monte Carlo results rather than tuned; and the κ-filtering improvement is predicted analytically and then verified in independent simulations. The main caveat is that the headline factor-2-5 comparison is not matched in sky coverage, so the paper's central quantitative claim mixes the effect of optimal filtering with the effect of using a larger effective sky area. This is fixable either with a matched-area control or by reframing the claim as a full-pipeline comparison.

major comments (1)
  1. [Sec. IV, Fig. 9; Sec. V; footnote after Eq. (3.9)] The factor-2-5 variance reduction is presented in the Abstract and Conclusions as a gain from optimal CMB map filtering, but the comparison is not isolated in sky coverage. The optimal anisotropic runs use all non-zero hit pixels (fsky≈0.13), while the isotropic-masked baseline has effective area fraction fA,L=0.13 of the flat-sky map (fsky≈0.04) and the weighted baseline has fA,L=0.03 (fsky≈0.01), as stated in the footnote after Eq. (3.9). Since lensing power-spectrum variance scales approximately inversely with effective sky fraction for both cosmic-variance- and noise-dominated modes, a substantial fraction of the reported factor can be attributable to area alone, especially for the weighted baseline. I request either a matched-area comparison (for example, an isotropic filter applied over the same scanned region, or an anisotropic-filter run restricted to the masked baseline region) or a rewording of the Abstract and Sec. V so that the factor-2-5 claim is explicitly described as a full-pipeline (area plus filter) gain rather than a filter-only improvement.
minor comments (6)
  1. [Sec. V] The phrase "pre-trained neutral network" should read "pre-trained neural network."
  2. [Eq. (2.8)] Please define the notation C^{κκ}_fid explicitly as a diagonal matrix in harmonic space; as written, the symbol could be confused with the lensing power spectrum C^{κκ}_L.
  3. [Sec. II A 2] The sentence "mask all pixels with (temperature map) noise over ~9.6 µK arcmin" should read "noise above ~9.6 µK arcmin" for clarity.
  4. [Fig. 5 and footnote after Eq. (3.9)] The effective area fractions fA,L are reported in one place as fractions of the flat-sky map and in another as full-sky fractions (fsky≈0.04 and 0.01). Please state the convention at each occurrence, since this is directly relevant to interpreting the variance improvements.
  5. [Fig. 6] Please state more explicitly in the caption or text whether the plotted quantity is C^{φφ}_{L,fid} - <Ĉ^{φφ}_L>, and note that this is an additive correction applied after the fA,L normalization; the current caption is easy to misread as a bias subtraction akin to N0/N1.
  6. [Sec. III, Eq. (3.13)] The variance expression neglects N1 variance, which the paper notes can matter on small scales. Since Fig. 7 quotes the theoretical variance improvement over the full L-range, it would be helpful to indicate in the figure which scales are most affected by this approximation rather than only in the text.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: variance predictions are checked with independent Monte Carlo simulations, not fitted, and the area-mismatch in the headline comparison is a statistical attribution issue, not a derivation-equals-inputs loop.

full rationale

The central claims of the paper are not obtained by fitting the analytic model to the final variance curves. The patch approximation is constructed independently: Eq. (3.8) defines the analytic normalization fA,L from local responses (Eq. (3.10)), and the variance formulae (3.15)-(3.17) are closed-form predictions. Section IV then compares these predictions to 500-simulation Monte Carlo results, with the Fig. 7 predictions in good agreement with the measured Fig. 9 differences. The effective reconstruction noise N^kappa_0,eff is a modeling choice stated in Sec. II D, but the paper explicitly compares it with the ideal N_0,L filter (Fig. 7) and does not tune it to the target variance curves. The mean-field, N0, and N1 biases are MC-estimated rather than assumed from the model being tested. The paper cites prior work including some by the authors (Refs. [9], [15], [16], [27]), but these support standard likelihood and quadratic-estimator background results that are also available in independent references ([8], [10]-[12]); no uniqueness or optimality claim rests solely on a self-citation. The footnote to Eq. (3.9) discloses the different effective sky areas (f_sky = 0.04 and 0.01 for the masked and weighted baselines versus f_sky = 0.13 for the anisotropic runs), so the factor-2-5 gain is partly an area effect; that is a potential over-attribution in the Abstract, but it is a statistical comparison issue, not a circular reduction of the prediction to its inputs. Accordingly, no circular step satisfying the quoted-evidence standard is present.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central variance-improvement result rests on standard CMB lensing assumptions and a local patch model; the free parameters are analysis choices rather than new physics. No invented physical entities are introduced.

free parameters (3)
  • Effective reconstruction noise band (40<=L<=90) for κ-filter = Not a scalar fit; average of isotropic N0,L over L=40..90 mapped to pixel patch noise
    Used in Eq. (2.8) for the approximate Wiener filter of the reconstructed convergence map. The band choice is a modeling decision; the paper shows it is close to ideal N0,L filtering in Fig. 7.
  • Isotropic filter effective noise level in comparison pipelines = Chosen to minimize reconstructed lensing potential variance over multipoles; no single scalar quoted
    Sec. II A 2 states the assumed isotropic noise level in the filter is chosen to minimize the variance of the reconstructed lensing potential. This calibrates the baseline, so the factor 2-5 improvement is relative to an optimized baseline.
  • Mask threshold and apodization for isotropic masked filtering = SO: exclude pixels with temperature noise above ~9.6 uK arcmin; S4: ~1.8 uK arcmin; 30 arcmin Gaussian apodization
    These thresholds set the effective sky area for the masked-baseline comparison. The paper states the masked area was not fully optimized, so the variance comparison depends on this choice.
assumptions (5)
  • domain assumption Unlensed CMB and instrumental noise are Gaussian; lensing is a deterministic deflection.
    Used to write the likelihood in Eq. (2.1) and the QE derivation in Sec. II. Standard for CMB lensing, but foregrounds and non-Gaussian noise are ignored.
  • domain assumption Instrumental noise is uncorrelated between pixels and between T/Q/U, with infinite variance outside the scanned region.
    Eq. (2.1) and the scan model from the hit count map in Eq. (2.5). Real ground-based noise is correlated, as the authors note in Sec. V.
  • domain assumption Lensing reconstruction estimators are quasi-local and the noise varies slowly enough that the independent-patch decomposition is valid.
    Sec. III, first paragraph: convergence QEs are quasi-local and patches are taken as uncorrelated. This is the core analytical assumption behind fA,L and the κ-filter response model.
  • domain assumption Flat-sky approximation with a 116-degree square map is adequate for the comparisons.
    All numerical results use flat-sky LensIt simulations; the authors state full-sky extension is future work in Sec. V.
  • domain assumption The quadratic estimator is a sufficiently good approximation to the MAP lensing estimator over the tested noise levels.
    Sec. II B and the conclusions note QEs are not fully optimal in high signal-to-noise regimes, but they are the practical standard used here.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimal filtering for CMB lensing reconstruction." pith.science (2026). https://pith.science/paper/NPYRKLAC

@misc{pith2026190902653,
  author       = {Pith},
  title        = {Pith review of: Optimal filtering for CMB lensing reconstruction},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NPYRKLAC}},
  note         = {Machine review of arXiv:1909.02653}
}
read the original abstract

Upcoming ground-based cosmic microwave background experiments will provide CMB maps with high sensitivity and resolution that can be used for high fidelity lensing reconstruction. However, the sky coverage will be incomplete and the noise highly anisotropic, so optimized estimators are required to extract the most information from the maps. We focus on quadratic-estimator based lensing reconstruction methods that are fast to implement, and compare new more-optimally filtered estimators with various estimators that have previously been used in the literature. Input CMB maps can be optimally inverse-signal-plus-noise filtered using conjugate gradient (or other) techniques to account for the noise anisotropy. However, lensing reconstructions from these filtered input maps have an anisotropic response to the lensing signal and are difficult to interpret directly. We describe a second-stage filtering of the lensing maps and analytic response model that can be used to construct lensing power spectrum estimates that account for the anisotropic response and noise inhomogeneity in an approximately optimal way while remaining fast to compute. We compare results for simulations of upcoming Simons Observatory and CMB Stage-4 experiments to show the robustness of the more optimal lensing reconstruction pipeline and quantify the improvement compared to less optimal estimators. We find a substantial improvement in reconstructed lensing power variance between optimal anisotropic and isotropic filtering of CMB maps, and up to 30% improvement in variance by using the additional filtering step on the reconstruction potential map. Our approximate analytic response model is unbiased to within a small percent-level additional Monte Carlo correction.

Figures

Figures reproduced from arXiv: 1909.02653 by the authors.

Figure 1
Figure 1. FIG. 1. The structure of the optimized QE lensing reconstruction [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2. The normalized hit count map used to simulate anisotropic [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3. The normalized inverse approximate lensing reconstruction [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: FIG. 4. Mean field power spectrum [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5. Comparison between the approximate analytic [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6. The additive MC correction required after applying analytic [PITH_FULL_IMAGE:figures/full_fig_p010_6.png]
Figure 7
Figure 7. Figure 7: FIG. 7. Fractional improvement in the theoretical binned variance [PITH_FULL_IMAGE:figures/full_fig_p011_7.png]
Figure 8
Figure 8. Figure 8: FIG. 8. Simulated lensing potential reconstruction power spectrum [PITH_FULL_IMAGE:figures/full_fig_p011_8.png]
Figure 9
Figure 9. Figure 9: FIG. 9. Reconstructed lensing power spectrum variances from the various filtering methods for an SO-like experiment and an S4-like [PITH_FULL_IMAGE:figures/full_fig_p012_9.png]
Figure 10
Figure 10. Figure 10: FIG. 10 [PITH_FULL_IMAGE:figures/full_fig_p013_10.png]
Figure 11
Figure 11. Figure 11: FIG. 11. Cross-correlation coefficients (Eq. ( [PITH_FULL_IMAGE:figures/full_fig_p014_11.png]
Figure 12
Figure 12. Figure 12: FIG. 12. Lensing power spectra differences built from partially conjugate-gradient-converged inverse-variance filtered CMB maps, in units of [PITH_FULL_IMAGE:figures/full_fig_p016_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 8 canonical work pages

  1. [1]

    The noise can include variance due to foreground residuals, and a mask can be accounted for simply by taking the elements ofN−1 to be zero in the corresponding pixels

    Optimal filtering The inverse-variance filtered CMB maps can be written as ¯X≡ ( bCfidb⊤ +N )−1 X = ( Cfid)−1[( Cfid)−1 +b⊤N−1b ]−1 b⊤N−1X, (2.6) whereb is the transfer function (here we consider a simple isotropic Gaussian beam with full-width half-maximum θFWHM) and Cfid is a set of fiducial lensed power spectra. The noise can include variance due to foregroun...

    work page

  2. [2]

    (2.6) is trivial to invert in harmonic space if the noise (and beam) is taken as isotropic, as both the theory and noise covariance matrices are then diagonal in harmonic space

    Isotropic filtering of masked maps The covariance matrix ( bCfidb⊤ +N ) in Eq. (2.6) is trivial to invert in harmonic space if the noise (and beam) is taken as isotropic, as both the theory and noise covariance matrices are then diagonal in harmonic space. We refer to this approximation as ‘isotropic filtering’, which is substantially faster than optimal filt...

    work page

  3. [3]

    This corresponds to using the filter ( bCfidb⊤ +N )−1 WX , where W is diagonal in pixel space and N is taken to be isotropic

    Isotropic filtering of weighted maps Instead of applying isotropic filtering to apodized masked maps, one can also apply it to a more generally weighted map, for example to try to down-weight regions with higher noise. This corresponds to using the filter ( bCfidb⊤ +N )−1 WX , where W is diagonal in pixel space and N is taken to be isotropic. We consider the ...

    work page

  4. [4]

    Weak gravitational lensing of the CMB,

    Antony Lewis and Anthony Challinor, “Weak gravitational lensing of the CMB,” Phys. Rept. 429, 1–65 (2006), arXiv:astro-ph/0601594 [astro-ph]

    arXiv 2006

  5. [5]

    Planck 2018 results. VIII. Gravitational lensing,

    Planck Collaboration VIIII (Planck), “Planck 2018 results. VIII. Gravitational lensing,” (2018), arXiv:1807.06210 [astro- ph.CO]. 15

    arXiv 2018

  6. [6]

    The Simons Observatory: Science goals and forecasts,

    Peter Ade et al. (Simons Observatory), “The Simons Observatory: Science goals and forecasts,” (2018), arXiv:1808.07445 [astro-ph.CO]

    arXiv 2018

  7. [7]

    CMB-S4 Science Book, First Edition,

    Kevork N. Abazajian et al. (CMB-S4), “CMB-S4 Science Book, First Edition,” (2016), arXiv:1610.02743 [astro-ph.CO]

    arXiv 2016

  8. [8]

    Two-season Atacama Cosmology Telescope polarimeter lensing power spectrum,

    Blake D. Sherwin et al. , “Two-season Atacama Cosmology Telescope polarimeter lensing power spectrum,” Phys. Rev. D 95, 123529 (2017), arXiv:1611.09753 [astro-ph.CO]

    arXiv 2017

Show all 36 references
  1. [9]

    A Measurement of the Cosmic Microwave Background Gravitational Lensing Potential from 100 Square Degrees of SPTpol Data,

    K. T. Story et al. (SPT), “A Measurement of the Cosmic Microwave Background Gravitational Lensing Potential from 100 Square Degrees of SPTpol Data,” Astrophys. J. 810, 50 (2015), arXiv:1412.4760 [astro-ph.CO]

    2015 arXiv

  2. [10]

    A Measurement of the Cosmic Microwave Background B-Mode Polarization Power Spectrum at Sub-Degree Scales from 2 years of POLARBEAR Data,

    P. A. R. Ade et al. (POLARBEAR), “A Measurement of the Cosmic Microwave Background B-Mode Polarization Power Spectrum at Sub-Degree Scales from 2 years of POLARBEAR Data,” Astrophys. J.848, 121 (2017), arXiv:1705.02907 [astro- ph.CO]

    2017 arXiv

  3. [11]

    Reconstruction of lensing from the cosmic microwave background polarization,

    Christopher M. Hirata and Uros Seljak, “Reconstruction of lensing from the cosmic microwave background polarization,” Phys. Rev.D68, 083002 (2003), arXiv:astro-ph/0306354 [astro- ph]

    2003 arXiv

  4. [12]

    Maximum a posteriori CMB lensing reconstruction,

    Julien Carron and Antony Lewis, “Maximum a posteriori CMB lensing reconstruction,” Phys. Rev. D96, 063510 (2017), arXiv:1704.08230 [astro-ph.CO]

    2017 arXiv

  5. [13]

    Analyzing weak lensing of the cosmic microwave background using the likelihood function,

    Christopher M. Hirata and Uros Seljak, “Analyzing weak lensing of the cosmic microwave background using the likelihood function,” Phys. Rev. D67, 043001 (2003), arXiv:astro-ph/0209489 [astro-ph]

    2003 arXiv

  6. [14]

    CMB lensing reconstruction on the full sky,

    Takemi Okamoto and Wayne Hu, “CMB lensing reconstruction on the full sky,” Phys. Rev. D67, 083002 (2003), arXiv:astro- ph/0301031 [astro-ph]

    2003

  7. [15]

    Weak lensing of the CMB,

    Duncan Hanson, Anthony Challinor, and Antony Lewis, “Weak lensing of the CMB,” Gen. Rel. Grav. 42, 2197–2218 (2010), arXiv:0911.0612 [astro-ph.CO]

    2010 arXiv

  8. [16]

    Planck 2015 results. XIII. Cosmological parameters,

    Planck Collaboration XIII (Planck), “Planck 2015 results. XIII. Cosmological parameters,” A&A 594, A13 (2016), arXiv:1502.01589 [astro-ph.CO]

    2016 arXiv

  9. [17]

    CMB temperature lensing power reconstruction,

    Duncan Hanson, Anthony Challinor, George Efstathiou, and Pawel Bielewicz, “CMB temperature lensing power reconstruction,” Phys. Rev. D 83, 043005 (2011), arXiv:1008.4403 [astro-ph.CO]

    2011 arXiv

  10. [18]

    The shape of the CMB lensing bispectrum,

    Antony Lewis, Anthony Challinor, and Duncan Hanson, “The shape of the CMB lensing bispectrum,” JCAP1103, 018 (2011), arXiv:1101.2234 [astro-ph.CO]

    2011 arXiv

  11. [19]

    On the joint analysis of CMB temperature and lensing-reconstruction power spectra,

    Marcel M. Schmittfull, Anthony Challinor, Duncan Hanson, and Antony Lewis, “On the joint analysis of CMB temperature and lensing-reconstruction power spectra,” Phys. Rev. D 88, 063012 (2013), arXiv:1308.0286 [astro-ph.CO]

    2013 arXiv

  12. [20]

    Designs for next generation CMB survey strategies from Chile,

    Jason R. Stevens et al. , “Designs for next generation CMB survey strategies from Chile,” Proc. SPIE Int. Soc. Opt. Eng. 10708, 1070841 (2018), arXiv:1808.05131 [astro-ph.IM]

    2018 arXiv

  13. [21]

    Detection of Gravitational Lensing in the Cosmic Microwave Background,

    Kendrick M. Smith, Oliver Zahn, and Olivier Dore, “Detection of Gravitational Lensing in the Cosmic Microwave Background,” Phys. Rev. D76, 043510 (2007), arXiv:0705.3980 [astro-ph]

    2007 arXiv

  14. [22]

    Planck 2013 results. XVII. Gravitational lensing by large-scale structure,

    P. A. R. Ade et al. (Planck), “Planck 2013 results. XVII. Gravitational lensing by large-scale structure,” Astron. Astrophys. 571, A17 (2014), arXiv:1303.5077 [astro-ph.CO]

    2014 arXiv

  15. [23]

    Planck 2015 results. XV . Gravitational lensing,

    P. A. R. Ade et al. (Planck), “Planck 2015 results. XV . Gravitational lensing,” Astron. Astrophys. 594, A15 (2016), arXiv:1502.01591 [astro-ph.CO]

    2016 arXiv

  16. [24]

    Full-sky CMB lensing reconstruction in presence of sky-cuts,

    Aurelien Benoit-Levy, Typhaine Dechelette, Karim Benabed, Jean-Francois Cardoso, Duncan Hanson, et al., “Full-sky CMB lensing reconstruction in presence of sky-cuts,” A&A555, A37 (2013), arXiv:1301.4145 [astro-ph.CO]

    2013 arXiv

  17. [25]

    Pseudo-C𝓁 estimators which do not mix E and B modes,

    Kendrick M. Smith, “Pseudo-C𝓁 estimators which do not mix E and B modes,” Phys. Rev. D74, 083002 (2006), arXiv:astro- ph/0511629 [astro-ph]

    2006

  18. [26]

    CMB lensing reconstruction using cut sky polarization maps and pure- B modes,

    Ruth Pearson, Blake Sherwin, and Antony Lewis, “CMB lensing reconstruction using cut sky polarization maps and pure- B modes,” Phys. Rev. D90, 023539 (2014), arXiv:1403.3911 [astro-ph.CO]

    2014 arXiv

  19. [27]

    BICEP2 / Keck Array VIII: Measurement of gravitational lensing from large- scale B-mode polarization,

    P. A. R. Ade et al. (BICEP2, Keck Array), “BICEP2 / Keck Array VIII: Measurement of gravitational lensing from large- scale B-mode polarization,” Astrophys. J. 833, 228 (2016), arXiv:1606.01968 [astro-ph.CO]

    2016 arXiv

  20. [28]

    Lensing reconstruction with CMB temperature and polarization,

    Michael Kesden, Asantha Cooray, and Marc Kamionkowski, “Lensing reconstruction with CMB temperature and polarization,” Phys. Rev. D 67, 123507 (2003), astro- ph/0302536

    2003

  21. [29]

    Mass reconstruction with CMB polarization,

    Wayne Hu and Takemi Okamoto, “Mass reconstruction with CMB polarization,” Astrophys. J. 574, 566–574 (2002), arXiv:astro-ph/0111606 [astro-ph]

    2002 arXiv

  22. [30]

    CMB lensing reconstruction biases in cross-correlation with large- scale structure probes,

    Giulio Fabbian, Antony Lewis, and Dominic Beck, “CMB lensing reconstruction biases in cross-correlation with large- scale structure probes,” (2019), arXiv:1906.08760 [astro- ph.CO]

    2019 arXiv

  23. [31]

    Efficient computation of CMB anisotropies in closed FRW models,

    Antony Lewis, Anthony Challinor, and Anthony Lasenby, “Efficient computation of CMB anisotropies in closed FRW models,” Astrophys. J. 538, 473–476 (2000), arXiv:astro- ph/9911177 [astro-ph]

    2000

  24. [32]

    A hybrid approach to CMB lensing reconstruction on all- sky intensity maps,

    S. Plaszczynski, A. Lavabre, L. Perotto, and J-L Starck, “A hybrid approach to CMB lensing reconstruction on all- sky intensity maps,” Astron. Astrophys. 544, A27 (2012), arXiv:1201.5779 [astro-ph.CO]

    2012 arXiv

  25. [33]

    Lensing reconstruction from a patchwork of polarization maps,

    Toshiya Namikawa and Ryo Nagata, “Lensing reconstruction from a patchwork of polarization maps,” JCAP 1409, 009 (2014), arXiv:1405.6568 [astro-ph.CO]

    2014 arXiv

  26. [34]

    CMB-S4 Decadal Survey APC White Paper,

    Kevork Abazajian et al., “CMB-S4 Decadal Survey APC White Paper,” (2019), arXiv:1908.01062 [astro-ph.IM]

    2019 arXiv

  27. [35]

    Fast Wiener filtering of CMB maps with Neural Networks,

    Moritz M ¨unchmeyer and Kendrick M. Smith, “Fast Wiener filtering of CMB maps with Neural Networks,” (2019), arXiv:1905.05846 [astro-ph.CO]

    2019 arXiv

  28. [36]

    Healpix – a framework for high resolution discretization, and fast analysis of data distributed on the sphere,

    K. M. Gorski et al., “Healpix – a framework for high resolution discretization, and fast analysis of data distributed on the sphere,” Astrophys. J.622, 759–771 (2005), astro-ph/0409513. Appendix A: Optimal filtering performance Optimal filtering of the CMB maps (Eq. (2.6)) must ...

    2005 arXiv

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.