pith. sign in

arxiv: 2605.17150 · v1 · pith:R5LPUWM6new · submitted 2026-05-16 · 📡 eess.SP · astro-ph.EP· astro-ph.IM

A reversed solar illumination dependence of unintended emission from Starlink Direct-to-Cell satellites at 72-234 MHz with the EDA2

Haofan Dong , Houtianfu Wang , Hanlin Cai , Ozgur B. Akan This is my paper

Pith reviewed 2026-05-20 14:21 UTC · model grok-4.3

classification 📡 eess.SP astro-ph.EPastro-ph.IM
keywords StarlinkDirect-to-Cellunintended electromagnetic radiationradio frequency interferencesatellite emissionssolar illumination dependenceEDA2
0
0 comments X

The pith

Starlink Direct-to-Cell satellites emit stronger unintended radio waves in Earth's shadow than in sunlight

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

The paper reanalyzes over 112,000 radio detections of Starlink satellites at 72-234 MHz to characterize emissions from the new Direct-to-Cell payload. It separates 175 DTC satellites from 1,623 standard Ku-only satellites using an external catalogue and compares their range-corrected flux densities. DTC satellites overall produce 1.45 times higher flux than the comparison group, with a markedly higher detection rate at 230.469 MHz. The central result is a reversal in solar dependence: DTC emissions are stronger in eclipse than in sunlight (illuminated/eclipsed ratio of 0.47), while Ku-only satellites show the opposite pattern (ratio of 1.18). This holds after matching on altitude, latitude, frequency, and launch epoch and points to an active on-board source whose duty cycle rises at lower temperatures rather than scaling directly with solar photocurrent.

Core claim

Direct-to-Cell Starlink satellites produce unintended emissions whose population-level flux density is 1.45 times that of Ku-only v2-Mini satellites. Within the DTC group the illuminated-to-eclipsed flux density ratio is 0.47, reversing the 1.18 ratio seen in the Ku-only comparison; the reversal survives matching on altitude, sub-satellite latitude, frequency, and launch epoch. Excess emission at 230.469 MHz collapses to a single narrow ~24 kHz bin near 230.627 MHz that is absent in five control channels and is not explained by clock harmonics or uniform thermal scaling across the fleet.

What carries the argument

Catalogue-based separation of DTC and Ku-only satellite populations followed by direct comparison of their flux densities under eclipse versus sunlight conditions.

If this is right

  • An active on-board emitter is operating whose effective duty cycle is larger when the satellite equilibrium temperature is lower.
  • The narrow 24 kHz feature near 230.627 MHz accounts for most of the DTC excess at that coarse channel.
  • Polarisation anomalies at 11 of 21 frequencies indicate the emission carries structured properties beyond simple thermal radiation.
  • The heterogeneous expression across the fleet rules out explanations that rely on a few permanently bright units or uniform scaling with temperature.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • If the active-source picture is correct, changes in satellite thermal management could reduce low-frequency leakage.
  • Radio-quiet site planning may need to treat DTC satellites as more interfering during eclipse periods than during sunlit passes.
  • The same eclipse-versus-sunlight test could be applied to other large constellations to distinguish passive from active emission mechanisms.

Load-bearing premise

The catalogue correctly and completely identifies which satellites carry the Direct-to-Cell payload and the illumination state of each detection is assigned without major misclassification.

What would settle it

A fresh set of observations in which the DTC population no longer shows higher flux density in eclipse than in sunlight after the same altitude, latitude, and epoch matching.

Figures

Figures reproduced from arXiv: 2605.17150 by Hanlin Cai, Haofan Dong, Houtianfu Wang, Ozgur B. Akan.

Figure 1
Figure 1. Figure 1: Overview of the 1 798 v2-Mini Starlink satellites in the Grigg et al. (2025) EDA2 catalogue after the GCAT classification and quality cuts of Sect. 2. Left: number of stacked detections per satellite as a function of NORAD catalogue identifier, for the two populations compared in this work. Right: per-satellite median range-corrected flux density (in Jy at 1000 km), shown as a violin plot with interquartil… view at source ↗
Figure 2
Figure 2. Figure 2: Pass geometry of the analysed v2-Mini detections. Left: sub-satellite longitude and latitude for all v2-Mini Ku-only (purple) and DTC (orange) detections; the star marks the EDA2 location at the Murchison Radio-astronomy Observatory. Right: azimuth and zenith-angle polar projection of all detections, colour-coded by log10 range-corrected flux density. rejects the null hypothesis of identical distributions … view at source ↗
Figure 3
Figure 3. Figure 3: Dynamic spectra of the brightest pass at the 230.469 MHz coarse channel for (a) a representative v2-Mini DTC satellite (NORAD 60041, pass duration 128 s) and (b) a representative Ku-only v2-Mini satellite (NORAD 60347, pass duration 48 s), constructed from the 31 fine channels within the coarse band. In each panel the brightest pass is the pass with the highest integrated Snorm at this coarse channel for t… view at source ↗
Figure 4
Figure 4. Figure 4: Observed flux density as a function of satellite range, colour-coded by elevation, for (a) v2-Mini Ku-only, (b) v2-Mini DTC, and (c) v1.x Starlink satellites. The dashed line in each panel shows the inverse-square scaling ∝ r −2 anchored at the median range-corrected flux density of the Ku-only comparison population, drawn as a guide to the eye; the dotted line shows ten times this baseline. Points offset … view at source ↗
Figure 5
Figure 5. Figure 5: Cumulative distribution functions of the flux density for the DTC (orange) and Ku-only (purple) populations under four equivalent reductions: (a) raw per￾detection; (b) range-corrected per-detection; (c) raw per-satellite median; (d) range-corrected per-satellite median. The ratio of medians is quoted in each panel title; the 95% bootstrap confidence interval is shown for panel (d). test used here addresse… view at source ↗
Figure 6
Figure 6. Figure 6: Per-frequency ratio of DTC to Ku-only range-corrected median flux den￾sity, with 95% bootstrap confidence intervals. Squares filled in dark orange lie above unity (DTC-brighter channels); squares filled in lighter salmon lie below unity. The most extreme below-unity channel is at 130.469 MHz, where the DTC/Ku ratio falls to ≈ 0.16. The shaded band indicates the Radio Astronomy Service band protected by the… view at source ↗
Figure 7
Figure 7. Figure 7: Deviation of the XX polarisation fraction from the instrumental baseline fXX,inst in each coarse channel. Bars coloured in dark red survive the Benjamini– Hochberg false discovery rate control at q = 0.05; grey bars do not. The annotation at 230.469 MHz marks the channel with the largest deviation. DTC/Ku ratio plotted in [PITH_FULL_IMAGE:figures/full_fig_p008_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Characterisation of the 230.469 MHz XX fraction anomaly. (a) Per-satellite XX fraction distribution for satellites with at least five detections in this channel. (b) The same, sorted by XX fraction and colour-coded by mean elevation. (c) XX fraction by population, with the number of detections quoted. (d) Range-corrected flux density distributions in XX and YY, showing that the XX excess is not accompanied… view at source ↗
Figure 9
Figure 9. Figure 9: Per-fine-channel statistics within the 230.469 MHz coarse band (N = 80 894 v2-Mini fine-bin observations from the Grigg et al. (2025) catalogue, geometry￾filtered). Top: mean flux, split DTC vs Ku-only. Middle: median and 95th-percentile flux, pooled. Bottom: XX-feed fraction with binomial error bars; the pooled-sample value pˆ = 0.810 is marked. Fine channel index 22 (≈ 230.627 MHz) is highlighted as the … view at source ↗
Figure 10
Figure 10. Figure 10: Cross-coarse-channel control for the fine-channel 22 feature. The z￾score is the deviation of the fine-channel-22 mean flux from the mean over the other 30 fine bins, in units of the inter-bin standard deviation. The 230.469 MHz channel (red star) shows z = +11.0, whereas five high-occupancy control chan￾nels (150.78, 153.12, 161.72, 170.31, and 200.00 MHz; blue circles) all lie within ±1σ, ruling out an … view at source ↗
Figure 11
Figure 11. Figure 11: Three falsifiable mechanism-discrimination tests for the 230.627 MHz fine-channel feature reported in Sect. 3.4. (a) T1, harmonic-coincidence test: predicted frequencies N × f0 of all 14 fundamentals reported by Bassa et al. (2024) for v2-Mini Starlink satellites (top row) and a superset of 9 plausible crystal-oscillator candidates (bottom row), evaluated at the integer N minimising the residual relative … view at source ↗
Figure 12
Figure 12. Figure 12: Robustness of the DTC eclipse-state reversal. (a) Distribution of per-satellite illuminated/eclipsed ratios for DTC satellites with at least five detections in each state; the solid vertical line marks the population ratio of medians (the same statistic adopted in the primary test in the text), and the dashed line marks unity. (b) Illuminated/eclipsed ratio as a function of satellite altitude, for DTC (or… view at source ↗
Figure 13
Figure 13. Figure 13: Summary of the effect sizes reported in Sect. 3. Intervals show 95% bootstrap confidence intervals for median ratios and 95% Wilson score intervals for binomial proportions; the plotted interval for the Benjamini–Hochberg FDR-significant channel fraction is a visual guide and not a formal interval. This panel combines ratios and proportions on a common numerical axis as a compact visual summary, not as co… view at source ↗
read the original abstract

Second-generation Starlink Direct-to-Cell (DTC) satellites carry an additional payload for direct cellular phone connectivity whose unintended electromagnetic radiation (UEMR) at sub-300 MHz frequencies has not been individually characterised. We reanalyse 112,534 detections from 1,806 Starlink satellites observed with the Engineering Development Array version 2 (EDA2) at 21 frequencies between 72.685 and 234.375 MHz (Grigg et al. 2025), separating 175 DTC and 1,623 Ku-only v2-Mini comparison satellites via the McDowell General Catalogue (McDowell 2020). DTC satellites emit a range-corrected flux density 1.45x that of the Ku-only comparison (Cliff's delta = +0.30, p = 2.6e-11). At 230.469 MHz the XX detection fraction reaches 0.811 against a 0.481 baseline (p ~ 1e-274), and 11 of 21 frequency channels show Benjamini-Hochberg-significant polarisation anomalies. The DTC population is brighter in eclipse than in sunlight (illuminated/eclipsed flux density ratio 0.47) while the Ku-only comparison shows the opposite sense (1.18); the reversal persists across altitude, sub-satellite latitude, frequency, and launch-epoch matching. The reversal strongly disfavours UEMR mechanisms that scale monotonically with instantaneous solar photocurrent and favours an active on-board source whose effective duty cycle is larger at lower equilibrium temperature. Within the 230.469 MHz coarse channel, fine-channel inspection isolates the excess to a single ~24 kHz bin near 230.627 MHz, tail-driven and absent at five control channels. Three falsifiable mechanism-discrimination tests show this feature is not coincident with the LOFAR-resolved Bassa et al. (2024) clock fundamentals, is unresolved at the EDA2 24 kHz resolution, and is heterogeneously expressed across the v2-Mini fleet rather than driven by a few permanently bright units or by uniform thermal scaling.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 2 minor

Summary. The manuscript reports an analysis of unintended electromagnetic radiation (UEMR) from Starlink Direct-to-Cell (DTC) satellites observed with the EDA2 at 21 frequencies between 72.685 and 234.375 MHz. From 112,534 detections across 1,806 satellites, it separates 175 DTC from 1,623 Ku-only v2-Mini satellites using the McDowell General Catalogue. DTC satellites show 1.45x higher range-corrected flux density (Cliff's delta +0.30), with significant polarization anomalies in 11 channels (Benjamini-Hochberg corrected). The central result is a reversed solar illumination dependence: DTC brighter in eclipse (illuminated/eclipsed flux density ratio 0.47) while Ku-only shows the opposite (1.18); the reversal survives matching on altitude, sub-satellite latitude, frequency, and launch epoch. Excess is isolated to a single ~24 kHz bin near 230.627 MHz within the 230.469 MHz channel, with three falsifiable mechanism tests provided.

Significance. If the result holds, the work supplies evidence for an active on-board source in DTC satellites whose effective duty cycle increases at lower equilibrium temperature, disfavoring mechanisms that scale monotonically with solar photocurrent. This has direct implications for RFI mitigation in satellite constellations. The manuscript earns credit for its large sample (112,534 detections), multiple statistical controls (Cliff's delta, Benjamini-Hochberg, matching), fine-channel isolation, and explicit falsifiable mechanism-discrimination tests.

major comments (1)
  1. [Illumination dependence and matching analysis] The central claim of reversed illumination dependence (illuminated/eclipsed flux density ratio 0.47 for DTC versus 1.18 for Ku-only) is load-bearing on the accuracy of per-detection solar illumination state assignment. The manuscript reports that the reversal persists after matching but does not provide an error rate, confusion matrix, or sensitivity test for the illumination classifier itself, which could be affected by TLE timing offsets or shadow geometry. This directly supports the skeptic concern and requires additional validation or robustness checks.
minor comments (2)
  1. [Abstract] The abstract cites 'Grigg et al. 2025' and 'McDowell 2020'; ensure complete bibliographic entries appear in the reference list.
  2. [Methods] Clarify the precise definition and computation of 'range-corrected flux density' with an explicit formula or reference to the relevant equation.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive feedback and positive evaluation of the manuscript's significance, sample size, and statistical controls. We address the single major comment below and will incorporate additional validation to strengthen the central illumination-dependence result.

read point-by-point responses

  1. Referee: [Illumination dependence and matching analysis] The central claim of reversed illumination dependence (illuminated/eclipsed flux density ratio 0.47 for DTC versus 1.18 for Ku-only) is load-bearing on the accuracy of per-detection solar illumination state assignment. The manuscript reports that the reversal persists after matching but does not provide an error rate, confusion matrix, or sensitivity test for the illumination classifier itself, which could be affected by TLE timing offsets or shadow geometry. This directly supports the skeptic concern and requires additional validation or robustness checks.

    Authors: We agree that explicit validation of the solar illumination state assignment strengthens the central claim. Illumination states were derived from standard TLE-based propagation and umbra/penumbra geometry using the McDowell catalogue. To address potential TLE timing offsets and shadow-geometry approximations, the revised manuscript will add a dedicated sensitivity analysis: we will shift eclipse transition times by ±5 min (a conservative envelope for Starlink TLE accuracy at LEO) and recompute the illuminated-to-eclipsed flux-density ratios for both DTC and Ku-only populations. We will also report the expected misclassification rate based on published TLE precision and include a brief discussion of how the existing multi-covariate matching (altitude, latitude, frequency, launch epoch) already mitigates residual errors. These additions will appear as a new subsection with accompanying figures. revision: yes

Circularity Check

0 steps flagged

No significant circularity: results are direct empirical ratios from classified observational data

full rationale

The paper's central claims consist of statistical comparisons (flux-density ratios, detection fractions, and illumination dependence) computed directly from the 112,534 detections after external catalogue-based separation of DTC vs. Ku-only satellites and assignment of solar illumination states. No equations, fitted parameters, or predictions are defined in terms of the target quantities themselves, and no load-bearing step reduces to a self-citation chain or ansatz that assumes the reported reversal. The analysis is self-contained against the input detections and external catalogues, with the reversal tested via explicit matching rather than derived by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The analysis rests on standard radio-astronomy detection thresholds and statistical corrections plus one key external classification; no new physical entities are postulated and only measured ratios appear.

axioms (2)
  • domain assumption The McDowell General Catalogue provides accurate and exhaustive DTC vs Ku-only labels for the observed satellites.
    Used to partition the 1806 satellites into 175 DTC and 1623 comparison objects.
  • domain assumption Solar illumination state (eclipse vs sunlight) can be assigned to each detection without material error from orbital ephemeris or timing.
    Central to the illuminated/eclipsed flux ratio comparison.

pith-pipeline@v0.9.0 · 5949 in / 1470 out tokens · 49594 ms · 2026-05-20T14:21:04.926880+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

33 extracted references · 33 canonical work pages · 2 internal anchors

  1. [1]

    and Winkel, B

    Di Vruno, F. and Winkel, B. and Bassa, C. G. and Józsa, G. I. G. and Brentjens, M. A. and Jessner, A. and Garrington, S. , title =. Astronomy & Astrophysics , year =. 2307.02316 , archiveprefix =

    work page arXiv

  2. [2]

    and Tingay, S

    Grigg, D. and Tingay, S. J. and Sokolowski, M. and Wayth, R. B. and Indermuehle, B. and Prabu, S. , title =. Astronomy & Astrophysics , year =. 2309.15672 , archiveprefix =

    work page arXiv

  3. [3]

    Bassa, C. G. and Di Vruno, F. and Winkel, B. and Jozsa, G. I. G. and Brentjens, M. A. and Zhang, X. , title =. Astronomy & Astrophysics , year =. 2409.11767 , archiveprefix =

    work page arXiv

  4. [4]

    Zhang and P

    X. Zhang and P. Zarka and C. Viou and A. Loh and C. G. Bassa and Q. Duchene and C. Tasse and J-M. Grießmeier and J. D. Turner and O. Ulyanov and L. V. E. Koopmans and F. Mertens and V. Zakharenko and C. Briand and B. Cecconi and R. Vermeulen and O. Konovalenko and J. Girard and S. Corbel , title =. Astronomy & Astrophysics , year =. 2504.10032 , archiveprefix =

    work page arXiv

  5. [5]

    and Tingay, S

    Grigg, D. and Tingay, S. J. and Sokolowski, M. , title =. Astronomy & Astrophysics , year =. 2506.02831 , archiveprefix =

    work page arXiv

  6. [6]

    doi:10.1051/0004-6361/201220873 , author =

    Astronomy & Astrophysics , year =. doi:10.1051/0004-6361/201220873 , author =

    work page doi:10.1051/0004-6361/201220873

  7. [7]

    Tingay and Raunaq Bhushan and Tom Booler and Riccardo Chiello and David B

    Randall Wayth and Marcin Sokolowski and Jess Broderick and Steven J. Tingay and Raunaq Bhushan and Tom Booler and Riccardo Chiello and David B. Davidson and David Emrich and Budi Juswardy and David Kenney and Giulia Macario and Alessio Magro and Andrea Mattana and David Minchin and Jader Monari and Andrew McPhail and Federico Perini and Giuseppe Pupillo a...

    work page arXiv

  8. [8]

    and Girard, J

    Zarka, P. and Girard, J. N. and Tagger, M. and Denis, L. , title =. SF2A-2012: Proceedings of the Annual meeting of the French Society of Astronomy and Astrophysics , editor =. 2012 , pages =

    work page 2012

  9. [9]

    doi:10.23919/ursi-at-rasc.2018.8471648 , author =

    2018 2nd URSI Atlantic Radio Science Meeting (AT-RASC) , year =. doi:10.23919/ursi-at-rasc.2018.8471648 , author =

    work page doi:10.23919/ursi-at-rasc.2018.8471648 2018

  10. [10]

    IEEE Proceedings , year = 2009, month = aug, volume =

    Proceedings of the IEEE , year =. doi:10.1109/JPROC.2009.2021005 , author =

    work page doi:10.1109/jproc.2009.2021005 2009

  11. [11]

    2015 , pages =

    Advancing Astrophysics with the Square Kilometre Array (AASKA14) , series =. 2015 , pages =. doi:10.22323/1.215.0001 , author =

    work page doi:10.22323/1.215.0001 2015

  12. [12]

    Cosmology at Low Frequencies: The 21 cm Transition and the High-Redshift Universe

    Physics Reports , year =. doi:10.1016/j.physrep.2006.08.002 , author =. astro-ph/0608032 , archiveprefix =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/j.physrep.2006.08.002 2006

  13. [13]

    doi:10.3847/2041-8213/ab8016 , author =

    The Astrophysical Journal Letters , year =. doi:10.3847/2041-8213/ab8016 , author =

    work page doi:10.3847/2041-8213/ab8016 2041

  14. [14]

    Lawler, S. M. and Boley, A. C. and Rein, H. , title =. The Astronomical Journal , year =. 2109.04328 , archiveprefix =

    work page arXiv

  15. [15]

    Bulletin of the American Astronomical Society , year =

    Walker, Constance and Hall, Jeffrey and Allen, Lori and Green, Richard and Seitzer, Patrick and Tyson, Tony and Bauer, Amanda and Krafton, Kelsie and Lowenthal, James and Parriott, Joel and Puxley, Phil and Abbott, Tim and Bakos, Gaspar and Barentine, John and Bassa, Cees and Blakeslee, John and Bradshaw, Andrew and Cooke, Jeff and Devost, Daniel and Gala...

    work page

  16. [16]

    and Di Pippo, S

    Walker, C. and Di Pippo, S. and Aub. 2020 , note =

    work page 2020

  17. [17]

    doi:10.1038/s41550-023-01904-2 , author =

    Nature Astronomy , year =. doi:10.1038/s41550-023-01904-2 , author =

    work page doi:10.1038/s41550-023-01904-2

  18. [18]

    and Otarola, A

    Tregloan-Reed, J. and Otarola, A. and Ortiz, E. and Molina, V. and Anais, J. and Gonz. Astronomy & Astrophysics , year =. 2003.07251 , archiveprefix =

    work page arXiv 2003

  19. [19]

    Anthony Tyson and Željko Ivezić and Andrew Bradshaw and Meredith L

    J. Anthony Tyson and Željko Ivezić and Andrew Bradshaw and Meredith L. Rawls and Bo Xin and Peter Yoachim and John Parejko and Jared Greene and Michael Sholl and Timothy M. C. Abbott and Daniel Polin , title =. The Astronomical Journal , year =. 2006.12417 , archiveprefix =

    work page arXiv 2006

  20. [20]

    Mann, H. B. and Whitney, D. R. , title =. The Annals of Mathematical Statistics , year =

    work page

  21. [21]

    , title =

    Cliff, N. , title =. Psychological Bulletin , year =

    work page

  22. [22]

    and Hochberg, Y

    Benjamini, Y. and Hochberg, Y. , title =. Journal of the Royal Statistical Society: Series B (Methodological) , year =

    work page

  23. [23]

    Wilson, E. B. , title =. Journal of the American Statistical Association , year =

    work page

  24. [24]

    and Tibshirani, R

    Efron, B. and Tibshirani, R. J. , title =

    work page

  25. [25]

    Feigelson, E. D. and Babu, G. J. , title =

    work page

  26. [26]

    and Hancock, P

    Prabu, S. and Hancock, P. J. and Zhang, X. and Tingay, S. J. , title =. Publications of the Astronomical Society of Australia , year =. 2002.01674 , archiveprefix =

    work page arXiv 2002

  27. [27]

    Wilensky, M. J. and Barry, N. and Morales, M. F. and Hazelton, B. J. and Byrne, R. , title =. Monthly Notices of the Royal Astronomical Society , year =. 2004.07819 , archiveprefix =

    work page arXiv 2004

  28. [28]

    Paul, C. R. , title =

    work page

  29. [29]

    Harris and K

    Nature , year =. doi:10.1038/s41586-020-2649-2 , author =

    work page doi:10.1038/s41586-020-2649-2

  30. [30]

    2020 , keywords =

    Nature Methods , year =. doi:10.1038/s41592-019-0686-2 , author =

    work page doi:10.1038/s41592-019-0686-2

  31. [31]

    2010 b , in P roceedings of the 9th P ython in S cience C onference, ed

    Proceedings of the 9th Python in Science Conference , editor =. 2010 , pages =. doi:10.25080/Majora-92bf1922-00a , author =

    work page doi:10.25080/majora-92bf1922-00a 2010

  32. [32]

    Hunter , title =

    Computing in Science & Engineering , year =. doi:10.1109/MCSE.2007.55 , author =

    work page doi:10.1109/mcse.2007.55 2007

  33. [33]

    The Astropy Project: Sustaining and Growing a Community-oriented Open-source Project and the Latest Major Release (v5.0) of the Core Package

    The Astrophysical Journal , year =. doi:10.3847/1538-4357/ac7c74 , author =

    work page internal anchor Pith review doi:10.3847/1538-4357/ac7c74