Determining the orientation of radio antennas at the South Pole using Galactic noise measurements

Paras Koundal; Valeria Torres-Gomez (for the IceCube Collaboration)

REVIEW 1 major objections 1 minor 12 references

Determining the orientation of radio antennas at the South Pole using Galactic noise measurements

T0 review · 1 major / 1 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The orientation of South Pole radio antennas can be recovered from the sinusoidal variation of Galactic noise with the Galactic Center's azimuth.

desk verdict Clever and plausible use of the sidereal Galactic rotation at the South Pole to get antenna orientation; proof-of-principle works against GPS, but the unquantified point-source assumption and missing systematics keep it from being an operational tool yet. read the letter →

arxiv 2507.07224 v1 pith:3R3UEJQ3 submitted 2025-07-09 astro-ph.HE astro-ph.IM

Paras Koundal , Valeria Torres-Gomez (for the IceCube Collaboration) This is my paper

classification astro-ph.HEastro-ph.IM

keywords radioantennaorientationGalacticnoiseCenterIceCubeSouthPoleair-showerdetectioncalibrationsiderealmodulation

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 claims that the orientation of each radio antenna in IceCube's prototype surface station can be determined from periodic variations in the strength of Galactic background noise. Because the Galactic Center circles the South Pole once per sidereal day at nearly constant zenith angle, the antenna's changing response to this bright radio source imprints a sinusoid on the measured noise amplitude. Fitting that sinusoid versus the azimuth of the Galactic Center yields the antenna arm direction, after adding or subtracting 90 degrees. The yearly data-driven orientations agree with GPS survey measurements within two standard deviations, so the method offers a way to monitor antenna alignment without field campaigns.

What carries the argument

The central object is the sidereal-day modulation of the root-mean-square (RMS) amplitude of cleaned radio waveforms, plotted against the azimuth of the Galactic Center. A fit line of the form $A\sin(2\pi f x + \varphi) + \mu$ with $f = 2/360$ per degree, i.e., a 180-degree-period sinusoid, extracts the phase $\varphi$, which gives the Galactic Center azimuth of maximum response. The underlying hypothesis is that the antenna arm is perpendicular to the direction of the Galactic Center when the noise is maximal. A preprocessing chain (spike removal, an inverted spectrum filter, a band-pass filter selecting 150-170 MHz, and a moving-window RMS with 2D outlier cuts) isolates the Galactic signal before the fit.

What would settle it

Simulate the expected RMS variation versus Galactic Center azimuth using a realistic all-sky Galactic emission map and the antenna beam pattern; if the phase of a 180-degree sinusoid fitted to the simulation differs from the point-source expectation by more than the roughly 6-degree fit uncertainty, the reported orientations carry a bias. A direct test would be to rotate an antenna arm by a known angle and check whether the phase of the measured sinusoid tracks that rotation within the fit uncertainty.

Watch

Extended reading notes

Core claim

The central claim is that the phase of a sinusoidal fit to the RMS Galactic noise as a function of the Galactic Center azimuth encodes the antenna's orientation. For each antenna polarization, monthly noise measurements are binned by the Galactic Center azimuth and fitted with a sinusoid of fixed period 180 degrees; the azimuth at which the noise peaks is interpreted as the direction perpendicular to the antenna arm, so the arm orientation is that azimuth shifted by 90 degrees. Yearly orientation estimates obtained this way for the three prototype antennas agree with the December 2024 GPS survey within two standard deviations, and the two polarizations of each antenna come out perpendicular as expected. The paper presents this as a data-driven complement to GPS-based alignment that can be repeated on a yearly basis.

Load-bearing premise

The analysis assumes the Galactic noise peaks when the antenna arm is perpendicular to the direction of the Galactic Center and that a single 180-degree-period sinusoid describes the response, so any extended Galactic plane emission that breaks this symmetry would shift the inferred orientation.

Editorial extensions

If this is right

Yearly antenna orientations from data match GPS surveys within two standard deviations, so the method can serve as an independent cross-check of GPS alignment.
Antenna orientation can be monitored on a yearly basis using only radio data, removing the need for specialized GPS field campaigns.
The same pipeline can be applied to future surface-array stations as they are deployed, since the Galactic Center is always visible from the South Pole.
Monthly orientation estimates also flag months or polarizations where anthropogenic noise corrupts the signal, because their fitted directions deviate from the perpendicular expectation.

Reading between the lines

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

A testable extension would be to replace the point-source assumption with an all-sky Galactic emission map and check whether the fitted phase shifts by more than the roughly 6-degree fit uncertainty reported in the paper; if it does, the inferred orientations carry a systematic bias.
The technique could be exported to any site where a bright, compact radio source is circumpolar, not just to the South Pole, since the required geometry is a source that circles the antenna once per sidereal day.
The monthly scatter in estimated orientations may encode information about radio-frequency interference or snow effects, so correlating that scatter with local conditions could turn the method into an environmental diagnostic.
Because the fit assumes a pure 180-degree-period response, fitting additional harmonics would test whether higher-order structure in the antenna beam pattern biases the recovered orientation.

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.

Desk Editor's Note

read the letter

Colleague,

The thing to know up front: this is a legitimate, simple idea, and the data actually show the expected modulation. The paper extracts antenna arm orientation from the phase of the 180-degree sinusoidal modulation of RMS Galactic noise as the Galactic Center circles the South Pole, and the yearly estimates agree with GPS within 2 sigma for the three prototype antennas. That is a genuinely new application — prior work used Galactic noise for gain calibration, not orientation. The physical argument is sensible, the pipeline (spike removal, band-pass to 150-170 MHz, moving-window RMS, 2D cut) is described in enough detail to follow, and the external GPS check is an honest comparison. Self-citations are only supporting references; no circularity problem.

The soft spots are real, and mostly acknowledged. The load-bearing assumption is that the RMS maximum occurs when the antenna arm is perpendicular to a point-like Galactic Center. At 150-170 MHz the Galactic plane is extended and the SKALA-v2 beam is broad, so the first harmonic of the integrated sky signal can be phase-shifted from the true GC azimuth. Because both polarizations share the same sky, the perpendicularity check cannot see a common offset. The paper calls this a "key hypothesis" but gives no estimate of the resulting systematic error. That is the main gap. The 2D RMS cut is visually selected, with no stability test; the GPS survey was done in December 2024 against 2023 data, so small seasonal ground shifts are possible but minor; and six polarization-year comparisons at 2 sigma is not strong evidence of zero bias. No code or data are shipped, so reproducibility is limited, but this is a conference proceedings.

All that said, this is a fair conference paper. The central idea is sound, and the validation path is clear: quantify the phase bias from an extended sky model and the real beam, and the method becomes a genuinely self-contained calibration capability for IceCube-Gen2 and similar arrays. It deserves a serious referee rather than a desk reject; the referee can ask for exactly that quantifying step. I would not cite it in my own work yet, but I would happily bring it to a group meeting to discuss the bias question.

Recommendation: engage with it as a promising proof-of-principle, and push for a systematic bias estimate in the next iteration.

Referee Report

1 major / 1 minor

Summary. The paper presents a data-driven method to determine the rotational orientation of the SKALA-v2 radio antennas of the IceCube surface-array enhancement prototype station by analyzing the sidereal modulation of Galactic background noise. Waveforms from fixed-rate triggers are cleaned of artifacts and radio-frequency interference, an RMS amplitude per waveform is computed in a frequency band where Galactic emission is strong (150-170 MHz), and the monthly RMS versus Galactic-Center azimuth is fit with a sinusoid of fixed period 180 degrees. The phase of this sinusoid is converted into an antenna-arm orientation under the assumption that the RMS maximum occurs when the antenna arm is perpendicular to the Galactic Center direction. Yearly orientation estimates for three antennas and two polarizations are compared with a differential GPS survey from December 2024, and the paper claims agreement within two standard deviations. The method is proposed as a potential alternative or complement to GPS-based surveys for future IceCube-Gen2 surface-array antennas.

Significance. If the method is validated, it would provide a low-cost, remote, and repeatable way to monitor antenna orientations at the South Pole, where GPS field campaigns are rare and expensive. The physical principle is sound in the idealized limit: a linearly polarized antenna's response to a compact radio source modulates with a 180-degree period as the source circles the pole, and the phase of that modulation encodes the arm orientation. The manuscript is commendably transparent about its pipeline, and the use of an external GPS comparison is the right validation strategy; the claimed agreement, if confirmed with proper uncertainties, would be a useful result for the radio-array community. The main limitation is that the validation rests on three antennas, one year of data, and a comparison that currently quotes only statistical errors, with no quantitative treatment of the extended-emission and beam-shape effects that could bias the fitted phase.

major comments (1)

[Section 3, Fig. 8] The sinusoid fit in Figure 7 fixes f = 2/360 and fits only amplitude, phase, and offset. Please justify this fixed period explicitly or test it by allowing the period to be free, or by including a first-harmonic term. Extended Galactic emission would generically introduce additional harmonic content, and the phase of the fitted fundamental could shift if the model is misspecified.

minor comments (1)

[Section 3, Fig. 8] The description 'The rectangular panels on the left in Figure 8' is confusing because Figure 8 appears to contain both a circular illustration and rectangular subpanels; please rephrase to refer to the subpanels unambiguously.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the antenna orientation is read out from the free phase of a sinusoidal fit and checked against independent GPS surveys, so the central result is not forced by its inputs.

full rationale

The derivation chain is self-contained in the relevant sense. The RMS Galactic-noise values are measured from cleaned waveforms (Section 2), binned as a function of the computed Galactic Center azimuth, and fit with a sinusoid A sin(2π f x + φ) + μ with f fixed at 2/360 but amplitude, phase, and offset free (Figure 7). The antenna orientation is then obtained from the fitted phase by adding or subtracting 90 degrees under the explicitly stated 'key hypothesis' that the RMS maximum occurs when the antenna arm is perpendicular to the direction toward the Galactic Center (Section 1). Nothing in the fit constrains the phase to the GPS-measured orientation; the GPS December 2024 survey is introduced only afterward as an external comparison (Section 3), and the claimed agreement is an empirical check rather than an input. The Cane model is used only to select a frequency band (Section 2, item 2), not to define the orientation. Self-references to IceCube proceedings and the Turcotte-Tardif thesis concern station deployment and artifact-cleaning details and are not load-bearing for the central phase-to-orientation mapping. The point-like Galactic Center assumption could bias the phase if extended emission dominates, but that is a modeling and systematic uncertainty, not a case of the prediction being equivalent to the input by construction.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper introduces no new physical entities. It relies on standard antenna response, the Cane model, and an idealized point-source Galactic Center assumption.

free parameters (2)

Sinusoid fit parameters A, phi, mu per antenna polarization per month = e.g., Jan 2023 Ant-1 Pol-0: A=0.167±0.020 µV, phi=274.301±6.406°, mu=0.007±0.014 µV
The orientation estimate is derived from the fitted phase; amplitude and offset are nuisance parameters. These are fit to the RMS-vs-azimuth data (Figure 7).
2D RMS cut boundaries = not specified numerically
Visually selected using the cleanest month (October for Ant-1 Pol-0) and kept constant; the cut affects the event sample and hence the fitted phase, but no stability check is provided.

assumptions (4)

domain assumption A dipole-like antenna has maximum response when the signal arrives perpendicular to the antenna arm (key hypothesis).
Stated in Section 1; used to convert the phase of the fitted sinusoid into an orientation by adding/subtracting 90 degrees.
domain assumption The Galactic noise in the 150-170 MHz band varies with GC azimuth as a period-180-degree sinusoid; the GC dominates the anisotropy.
The fit fixes f=2/360 (Figure 7); no test of higher harmonics or contribution from the extended Galactic plane is shown.
standard math The Cane model approximates the expected Galactic spectrum in the selected band.
Used to select the 150-170 MHz band (Figure 4, reference [10]).
domain assumption The pre-processing pipeline (spike removal, spectral filters, 2D cut) removes noise without biasing the phase.
Section 2 assumes the cleaned waveforms represent the Galactic signal; no bias study is presented.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Determining the orientation of radio antennas at the South Pole using Galactic noise measurements." pith.science (2026). https://pith.science/paper/3R3UEJQ3

@misc{pith2026250707224,
  author       = {Pith},
  title        = {Pith review of: Determining the orientation of radio antennas at the South Pole using Galactic noise measurements},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/3R3UEJQ3}},
  note         = {Machine review of arXiv:2507.07224}
}

read the original abstract

The IceCube Neutrino Observatory is a multi-messenger observatory at the South Pole. As preparation for an enhancement of its surface array, IceTop, a prototype station consisting of elevated scintillation panels and radio antennas has been installed and is operating since 2020. The radio antennas detect emissions from cosmic-ray-induced air showers, and their precise orientation is essential for an accurate reconstruction of the air-shower properties. This work presents a novel method to determine the orientation by analyzing periodic variations of the Galactic background noise recorded by the antennas. In particular, we examine noise level variations correlated with the Earth's rotation and the apparent position of the Galactic Center. The method can provide a potential alternative or augment GPS-based measurements of the alignment of radio antennas at the South Pole.

Figures

Figures reproduced from arXiv: 2507.07224 by the authors.

**Figure 1.** 3D model of SKALA-v2 radioantenna used in the prototype station at IceCube. This work utilizes waveforms measured with the radio antennas of the SAE prototype station in 2023. A three-dimensional model of the antenna used for the measurements is shown in [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

**Figure 2.** Flowchart illustrating the pre-processing pipeline applied to waveforms prior to antenna orientation estimation. The steps include fixed-rate trigger event selection, artifact/spike removal (targeting anthropogenic and electronic noise), and spectral cleaning via an Inverted Spectrum Filter followed by a band-pass filter to suppress narrowband interference and isolate a specific galactic emission band. This is foll… view at source ↗

**Figure 3.** Two examples waveforms showing affect of artifact removal. For details check item 1 (section 2). spike location is within five mean absolute deviation of the waveform bin amplitudes, then the amplitude at the spike location is assigned the bit-flipped value. Otherwise, the amplitude is assigned as the mean of the neighbors (one on both sides). If the spike is at the end (last-bin) of the waveform, then the mean is t… view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: Frequency Spectra for Antenna 1 - Polarization 0 for January, 2023. For details check item 2 (section 2). also, offset for easier visibility). This filter can also be applied twice. The galactic emission strength expectation is labeled as "Cane Model" [10]. The 150-17…

**Figure 5.** Figure 5: The 2D distribution of RMS values and their spread used to identify and remove noisy events that persist even after initial cleaning. The shaded regions represent the applied cuts, optimized to retain clean events while minimizing loss in months with the fewest outlier…

**Figure 7.** Figure 7: RMS variation as a function of galactic center azimuth for a full month, revealing a sinusoidal pattern. The phase of this sinusoid indicates the azimuthal orientation of galactic center maxima, allowing determination of antenna arm alignment. Gaps occur because meas…

**Figure 8.** Figure 8: Estimated orientation directions using the methodology outlined section 2. Each rectangular subplot corresponds to a unique antenna and polarization. Monthly orientation estimates are shown with error bars, which are subsequently fit to obtain a yearly estimate. GPS-de…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

12 extracted references · 9 canonical work pages

[1]

write newline

" write newline "" before.all 'output.state := FUNCTION blank.sep after.quote 'output.state := FUNCTION fin.entry output.state after.quoted.block = 'skip 'add.period if write newline FUNCTION new.block output.state before.all = 'skip output.state after.quote = after.quoted.block 'output.state := after.block 'output.state := if if FUNCTION new.sentence out...
[2]

IceCube Collaboration , http://dx.doi.org/10.22323/1.444.0377 `` Accounting for changing snow over 10 years of IceTop, and its impact on the all-particle cosmic ray spectrum ,'' ICRC2023, p. 377. July, 2023

work page doi:10.22323/1.444.0377 2023
[3]

Haungs, http://dx.doi.org/10.1051/epjconf/201921006009 EPJ Web of Conferences 210 (2019) 06009

A. Haungs, http://dx.doi.org/10.1051/epjconf/201921006009 EPJ Web of Conferences 210 (2019) 06009

arXiv 2019
[4]

Shefali, PoS, ICRC2025 (these proceedings) 394

IceCube Collaboration, S. Shefali, PoS, ICRC2025 (these proceedings) 394

work page
[5]

Venugopal, PoS, ICRC2025 (these proceedings) 427

IceCube Collaboration, M. Venugopal, PoS, ICRC2025 (these proceedings) 427

work page
[6]

6, (Apr., 2021) 060501

IceCube-Gen2 Collaboration , http://dx.doi.org/10.1088/1361-6471/abbd48 Journal of Physics G: Nuclear and Particle Physics 48 no. 6, (Apr., 2021) 060501

doi:10.1088/1361-6471/abbd48 2021
[7]

, Fodran, T

Büsken, M. , Fodran, T. , and Huege, T. , http://dx.doi.org/10.1051/0004-6361/202245382 Astronomy and Astrophysics 679 (2023) A50

work page doi:10.1051/0004-6361/202245382 2023
[8]

Mulrey, A

K. Mulrey, A. Bonardi, S. Buitink, A. Corstanje, et al. , http://dx.doi.org/https://doi.org/10.1016/j.astropartphys.2019.03.004 Astroparticle Physics 111 (2019) 1--11

work page doi:10.1016/j.astropartphys.2019.03.004 2019

Show all 12 references

[9]

Turcotte-Tardif, http://dx.doi.org/10.5445/IR/1000160782 Radio Measurements of Cosmic Rays at the South Pole

R. Turcotte-Tardif, http://dx.doi.org/10.5445/IR/1000160782 Radio Measurements of Cosmic Rays at the South Pole . PhD thesis, Karlsruher Institut für Technologie (KIT), 2023

2023
[10]

IceCube Collaboration , http://dx.doi.org/10.22323/1.395.0317 `` Simulation Study of the Observed Radio Emission of Air Showers by the IceTop Surface Extension ,'' ICRC2021, p. 317. July, 2021

2021 doi
[11]

H. V. Cane, http://dx.doi.org/10.1093/mnras/189.3.465 Monthly Notices of the Royal Astronomical Society 189 no. 3, (Dec., 1979) 465–478

1979 doi
[12]

Bezyazeekov, N

P. Bezyazeekov, N. Budnev, et al. , http://dx.doi.org/10.48550/ARXIV.1906.10947 `` Advanced Signal Reconstruction in Tunka-Rex with Matched Filtering and Deep Learning ,'' CEUR Workshop Proceedings, pp. 7--16. 2019

1906 doi arXiv

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

write newline

" write newline "" before.all 'output.state := FUNCTION blank.sep after.quote 'output.state := FUNCTION fin.entry output.state after.quoted.block = 'skip 'add.period if write newline FUNCTION new.block output.state before.all = 'skip output.state after.quote = after.quoted.block 'output.state := after.block 'output.state := if if FUNCTION new.sentence out...

[2] [2]

IceCube Collaboration , http://dx.doi.org/10.22323/1.444.0377 `` Accounting for changing snow over 10 years of IceTop, and its impact on the all-particle cosmic ray spectrum ,'' ICRC2023, p. 377. July, 2023

work page doi:10.22323/1.444.0377 2023

[3] [3]

Haungs, http://dx.doi.org/10.1051/epjconf/201921006009 EPJ Web of Conferences 210 (2019) 06009

A. Haungs, http://dx.doi.org/10.1051/epjconf/201921006009 EPJ Web of Conferences 210 (2019) 06009

arXiv 2019

[4] [4]

Shefali, PoS, ICRC2025 (these proceedings) 394

IceCube Collaboration, S. Shefali, PoS, ICRC2025 (these proceedings) 394

work page

[5] [5]

Venugopal, PoS, ICRC2025 (these proceedings) 427

IceCube Collaboration, M. Venugopal, PoS, ICRC2025 (these proceedings) 427

work page

[6] [6]

6, (Apr., 2021) 060501

IceCube-Gen2 Collaboration , http://dx.doi.org/10.1088/1361-6471/abbd48 Journal of Physics G: Nuclear and Particle Physics 48 no. 6, (Apr., 2021) 060501

doi:10.1088/1361-6471/abbd48 2021

[7] [7]

, Fodran, T

Büsken, M. , Fodran, T. , and Huege, T. , http://dx.doi.org/10.1051/0004-6361/202245382 Astronomy and Astrophysics 679 (2023) A50

work page doi:10.1051/0004-6361/202245382 2023

[8] [8]

Mulrey, A

K. Mulrey, A. Bonardi, S. Buitink, A. Corstanje, et al. , http://dx.doi.org/https://doi.org/10.1016/j.astropartphys.2019.03.004 Astroparticle Physics 111 (2019) 1--11

work page doi:10.1016/j.astropartphys.2019.03.004 2019