Linear map-making with LuSEE-Night
Hugo Camacho, Kaja M. Rotermund, Anže Slosar, Stuart D. Bale, David W. Barker, Jack Burns, Christian H. Bye, Johnny Dorigo Jones, Adam Fahs, Keith Goetz, Sven Herrmann, Joshua J. Hibbard, Oliver Jeong, Marc Klein-Wolt, Léon V.E. Koopmans, Joel Krajewski, Zack Li, Corentin Louis, Milan Maksimović, Ry · 2025
Read Linear map-making with LuSEE-Night by Hugo Camacho, Kaja M. Rotermund, Anže Slosar, Stuart D. Bale, David W. Barker, Jack Burns, Christian H. Bye, Johnny Dorigo Jones, Adam Fahs, Keith Goetz, Sven Herrmann, Joshua J. Hibbard, Oliver Jeong, Marc Klein-Wolt, Léon V.E. Koopmans, Joel Krajewski, Zack Li, Corentin Louis, Milan Maksimović, Ry (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-16773.txt.
Introduction
Hugo Camacho, Kaja Rotermund, and a large team describe map-making for a radio telescope bound for the far side of the Moon in a 2025 astronomy study. The instrument, a pathfinder called LuSEE-Night, will listen to faint cosmic radio signals shielded from Earth's noise. Turning its raw measurements into a sky map needs care because the telescope's response mixes directions and frequencies.
Linear map-making writes that response as a matrix problem and solves it directly. The authors lay out the method for this lunar instrument and its data pipeline. Clean maps of the radio sky at these frequencies could reveal conditions in the early universe.
Catalog record
| CREATOR | Hugo Camacho, Kaja M. Rotermund, Anže Slosar, Stuart D. Bale, David W. Barker, Jack Burns, Christian H. Bye, Johnny Dorigo Jones, Adam Fahs, Keith Goetz, Sven Herrmann, Joshua J. Hibbard, Oliver Jeong, Marc Klein-Wolt, Léon V.E. Koopmans, Joel Krajewski, Zack Li, Corentin Louis, Milan Maksimović, Ry |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 56 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-16773
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Linear map-making with LuSEE-Night Preprint: arXiv:xxxx.xxxx Hugo Camacho, Kaja M. Rotermund, Anže Slosar, Stuart D. Bale, David W. Barker, Jack Burns, Christian H. Bye, Johnny Dorigo Jones, Adam Fahs, Keith Goetz, Sven Herrmann, Joshua J. Hibbard, Oliver Jeong, Marc Klein-Wolt, Léon V.E. Koopmans, Joel Krajewski, Zack Li, Corentin Louis, Milan Maksimović, Ryan McLean, Raul A. Monsalve, Paul O’Connor, Aaron Parsons, Michel Piat, Marc Pulupa, Rugved Pund, David Rapetti, Benjamin Saliwanchik, Graham Speedie, Nikolai Stefanov, David Sundkvist, Aritoki Suzuki, Harish K. Vedantham, Fatima Yousuf, Philippe Zarka, LuSEE-Night Science Collaboration Abstract LuSEE-Night is a pathfinder radio telescope on the lunar far side employing four 3-m monopole antennas arranged as two horizontal cross pseudo-dipoles on a rotational stage and sensitive to the radio sky in the 1-50 MHz frequency band. LuSEE-Night measures the corresponding 16 correlation products as a function of frequency. While each antenna combination measures radiation coming from a large area of the sky, their aggregate information as a function of phase in the lunar cycle and rotational stage position can be deconvolved into a low-resolution map of the sky. We study this deconvolution using linear map-making based on the Wiener filter algorithm. We illustrate how systematic effects can be effectively marginalised over as contributions to the noise covariance and demonstrate this technique on beam knowledge uncertainty and gain fluctuations. With reasonable assumptions about instrument performance, we show that LuSEE-Night should be able to map the sub-50 MHz sky at a ∼ \sim 5-degree resolution. I Introduction The lunar far side has been touted as the ultimate radio observatory for low-frequency ( <30 MHz) observations of the sky. At these low frequencies, observations from Earth are limited by both the Earth’s ionosphere, which refracts and reflects radiation from the sky, and emissions from anthropogenic and meteorological sources. The lunar far side, permanently shielded from Earth’s signals and, during the lunar night, also shielded from the Sun, has long been considered an ideal location for low-frequency radio receivers. Indeed, there are several proposals for large futuristic telescopes, such as FarSide Burns et al. 2021a , ALO-DEX Brinkerink et al. 2025 , LCRT Bandyopadhyay et al. 2021 , as well as several recent operating pathfinders, such as the receiver on Chang’e-4 Su et al. 2019 , ROLSES Burns et al. 2021b ; Hibbard et al. 2025 on the Intuitive Machines 1 mission, and the upcoming LuSEE-Night on the Firefly’s Blue Ghost 2 mission. LuSEE-Night is a pathfinder radio telescope slated for delivery to the lunar far side in September 2026. A preliminary instrument description can be found in ( Bale et al. 2023 ) , with a more complete instrument design paper in preparation. Since the monopole antennas are electrically short at the lower end of the observed frequency range and have inherently low directivity, the instrument is more akin to a radio receiver than a traditional telescope. Given the available 16 correlation products, coupled with lunar rotation and rotation of the antenna turntable, however, the dataset will contain a large number of distinct measurements. Although each one of these measurements probes large patches of the sky, their number, and signal-to-noise will overconstrain the degrees of freedom available in the low-resolution map of the sky. The purpose of this paper is to investigate this problem quantitatively and to determine if LuSEE-Night’s data can be used to recover a map of the radio sky. To this end, we employ the Wiener filter, a well-established linear method for map reconstruction. Wiener filtering provides the maximum a posteriori (MAP) map estimate assuming a Gaussian distribution for the input map pixels and noise. It has been traditionally used in Cosmic Microwave Background (CMB) map-making Tegmark 1997 and for general data compression in cosmology (e.g., Seljak 1998 ). In this work, we utilize Wiener filtering as a deconvolution process under a minimal set of assumptions. Future work (e.g., Li et al. 2025 ) will explore more aggressive non-linear map-making algorithms that may produce higher fidelity maps, albeit at the cost of stronger, and potentially incorrect, assumptions. Conceptually, Wiener deconvolution is a linear process that optimally balances information from the data with prior knowledge. On signal-dominated scales, it trusts the data to faithfully reproduce sky features. Conversely, on noise-dominated scales, it regularizes the solution by pulling map modes (e.g., spherical harmonic coefficients or pixel values) towards their prior mean (typically zero, as in the Gaussian likelihood, zero is the most likely value in the absence of other information). In the intermediate regime, it gracefully interpolates between these two extremes. Consequently, the resulting map reconstructs modes that are sufficiently non-degenerate in the data space and possess adequate signal-to-noise. While low-frequency radiometers are often strongly signal-dominated, a significant practical challenge is the precise knowledge of the antenna beam profiles. In fact, as is typical in any deconvolution, the kernels need to be known precisely in order to correctly deconvolve the data. In this paper, we limit our discussion to unpolarized sources and a deconvolution that operates one frequency at a time. An extension to polarized sources and multiple frequencies, while more cumbersome, is not fundamentally different in its approach. II Map-making with a short-antenna radio instrument II.1 Observables Consider an array of N independently amplified antennas, indexed i=1,\ldots,N . These antennas produce time-varying signals that serve as input to a spectrometer. After spectral channelization, the signal from antenna i in a given frequency bin is described by a complex number S_{i} , encoding its amplitude and phase. Since we treat each frequency bin independently in this paper, we will omit the frequency index for brevity. For thermal radiation, all relevant information is contained in the two-point statistics of these complex signals. We can form N real-valued auto-correlation products, V_{ii}=S_{i}S_{i}^{*} , and N(N-1)/2 complex-valued cross-correlation products, V_{ij}=S_{i}S_{j}^{*} (for i\neq j ). Each complex cross-correlation V_{ij} provides two real degrees of freedom (its real and imaginary parts). Therefore, the total number of independent real-valued observables is N^{2} . The directional response of a given antenna is characterized by its complex electric field pattern, (E_{\theta}(\hat{\mathbf{n}}),E_{\phi}(\hat{\mathbf{n}})) , which describes its sensitivity to the \theta and \phi components of an incident electric field from direction \hat{\mathbf{n}}\equiv(\sin\theta\cos\phi,\sin\theta\sin\phi,\cos\theta)^{T} . For any given direction \hat{\mathbf{n}} , these two complex components correspond to four real degrees of freedom, determining the antenna’s sensitivity to the four Stokes parameters (I,Q,U,V) of radiation arriving from that direction. For a pair of antennas, with electric field responses (E_{\theta},E_{\phi}) for the first and (E^{\prime}_{\theta},E^{\prime}_{\phi}) for the second, the combined beam responses to the Stokes parameters are given by: \displaystyle B_{I} \displaystyle= \displaystyle\frac{1}{2}\left(E_{\theta}E_{\theta}^{\prime*}+E_{\phi}E_{\phi}^{\prime*}\right) (1) \displaystyle B_{Q} \displaystyle= \displaystyle\frac{1}{2}\left(E_{\theta}E_{\theta}^{\prime*}-E_{\phi}E_{\phi}^{\prime*}\right) (2) \displaystyle B_{U} \displaystyle= \displaystyle\left(E_{\theta}E_{\phi}^{\prime*}+E_{\phi}E_{\theta}^{\prime*}\right) (3) \displaystyle B_{V} \displaystyle= \displaystyle i(E_{\theta}E_{\phi}^{\prime*}-E_{\phi}E_{\theta}^{\prime*}) (4) For a radiation field described by Stokes parameters (I,Q,U,V) from direction \hat{\mathbf{n}} , the contribution to the measured correlation product V_{ij} is obtained by integrating the product of the appropriate beam response and the Stokes parameter over all solid angles. Assuming unpolarized radiation ( Q=U=V=0 ), the observed signal for antenna pair (i,j) simplifies to an integral of the sky intensity I(\hat{\mathbf{n}}) weighted by the intensity beam: V_{ij}=\int B_{ij}(\hat{\mathbf{n}})I(\hat{\mathbf{n}})d\Omega, (5) where B_{ij} denotes the intensity beam pattern (equivalent to B_{I} in Eq. 1 ) for the antenna pair (i,j) ; for brevity, we have dropped the