arXiv subjects

M. Koller

Publications and source records attributed to M. Koller.

6 recordsLinked to original sources

arXiv2025

Exploring Spatially-Resolved Metallicities, Dynamics and Outflows in Low-Mass Galaxies at $z \sim 7.6$

A majority of JWST/NIRSpec/IFU studies at high redshifts to date have focused on UV-bright or massive objects, while our understanding of low-mass galaxies at early cosmic times remains limited. In this work, we present NIRSpec/IFS high-resolution observations of two low-mass ($M_* < 10^9 \ M_\odot$), low-metallicity ($[12 + \log(\text{O/H})] < 8$) galaxies at $z \sim 7.66$, one of which we identify as hosting a Type-II AGN. We measure flat strong-line metallicity gradients, suggestive of ISM redistribution by outflows or past merging, but also identify tension with the direct-$T_\text{e}$ metallicity gradient in one galaxy. We measure $v_\text{rot}/\sigma < 1$ in both galaxies, consistent with observations of lower rotational support at early cosmic times. We identify broad kinematical components decoupled from galactic rotation with velocities of $\sim 250 - 500 \ \text{km} \ \text{s}^{-1}$ and argue these components trace outflows, for which we infer outflow rates of $\sim 8 - 14 \ M_\odot \ \text{yr}^{-1}$ with $v_\text{out}/v_\text{esc} \sim 1$. We compare our findings to results from the new large-volume AESOPICA simulations, which fully incorporate different models of black hole growth and AGN feedback. We find that our observational results of $v_\text{out}/v_\text{esc}$ are consistent with the simulated dwarf AGN population, hinting AGN-driven feedback may contribute to quenching both in our systems and in a wider population of low-mass galaxies in the early Universe. This novel study illustrates the necessity of deep IFU observations to decompose the complex kinematics and morphology of high-$z$ galaxies, trace outflows, and constrain the effect of feedback in the early Universe.

astro-ph.GA↗

arXiv2024

The MAGPI survey: The interdependence of the mass, star formation rate, and metallicity in galaxies at z~0.3

Star formation rates (SFRs), gas-phase metallicities, and stellar masses are crucial for studying galaxy evolution. The different relations resulting from these properties give insights into the complex interplay of gas inside galaxies and their evolutionary trajectory and current characteristics. We aim to characterize these relations at $z\sim 0.3$, corresponding to a 3-4 Gyr lookback time. We utilized optical integral field spectroscopy of 65 emission-line galaxies from the MAGPI survey at a redshift of $0.28<z<0.35$ and spanning a total stellar mass range of $8.2<\log(M_{*}/M_{\odot}) < 11.4$. We derived the resolved star formation main sequence (rSFMS), resolved mass metallicity relation (rMZR), and resolved fundamental metallicity relation (rFMR) at $z\sim 0.3$. We find a relatively shallow rSFMS slope of $\sim 0.425 \pm 0.014$ compared to the expected slope at this redshift for an ordinary least square (OLS) fitting routine. For an orthogonal distance regression (ODR) routine, a much steeper slope of $\sim 1.162 \pm 0.022$ is measured. We confirm the existence of an rMZR at $z\sim 0.3$ with an average metallicity located $\sim 0.03$ dex above the local Universe's metallicity. Via partial correlation coefficients, evidence is found that the local metallicity is predominantly determined by the stellar mass surface density and has a weak secondary (inverse) dependence on the SFR surface density $\Sigma_{SFR}$. Additionally, a significant dependence of the local metallicity on the total stellar mass $M_{*}$ is found. Furthermore, we find that the stellar mass surface density $\Sigma_{*}$ and $M_{*}$ have a significant influence in determining the strength with which $\Sigma_{SFR}$ correlates with the local metallicity. We observe that at lower stellar masses, there is a tighter correlation between $\Sigma_{SFR}$ and the gas-phase metallicity, resulting in a more pronounced rFMR.

astro-ph.GA↗

arXiv2022

Electric-field-controlled cold dipolar collisions between trapped CH$_3$F molecules

Reaching high densities is a key step towards cold-collision experiments with polyatomic molecules. We use a cryofuge to load up to 2$\times10^7$ CH$_3$F molecules into a box-like electric trap, achieving densities up to 10$^7$/cm$^3$ at temperatures around 350 mK where the elastic dipolar cross-section exceeds 7$\times$10$^{-12}$cm$^2$. We measure inelastic rate constants below 4$\times$10$^{-8}$cm$^3$/s and control these by tuning a homogeneous electric field that covers a large fraction of the trap volume. Comparison to ab-initio calculations gives excellent agreement with dipolar relaxation. Our techniques and findings are generic and immediately relevant for other cold-molecule collision experiments.

physics.atom-ph↗

arXiv2021

Centralized Learning of the Distributed Downlink Channel Estimators in FDD Systems using Uplink Data

In this work, we propose a convolutional neural network (CNN) based low-complexity approach for downlink (DL) channel estimation (CE) in frequency division duplex (FDD) systems. In contrast to existing work, we use training data which solely stems from the uplink (UL) domain. This allows to learn the CNN centralized at the base station (BS). After training, the network parameters are offloaded to mobile terminals (MTs) within the coverage area of the BS. The MTs can then obtain channel state information (CSI) of the MIMO channels with the low-complexity CNN estimator. This circumvents the necessity of an infeasible amount of feedback, i.e., acquisition of training data at the user, and the offline training phase at each MT. Numerical results show that the CNN which is trained solely based on UL data performs equally well as the network trained based on DL data. Furthermore, the approach is able to outperform state-of-the-art CE algorithms.

cs.IT↗

arXiv2021

A Low-Complexity MIMO Channel Estimator with Implicit Structure of a Convolutional Neural Network

A low-complexity convolutional neural network estimator which learns the minimum mean squared error channel estimator for single-antenna users was recently proposed. We generalize the architecture to the estimation of MIMO channels with multiple-antenna users and incorporate complexity-reducing assumptions based on the channel model. Learning is used in this context to combat the mismatch between the assumptions and real scenarios where the assumptions may not hold. We derive a high-level description of the estimator for arbitrary choices of the pilot sequence. It turns out that the proposed estimator has the implicit structure of a two-layered convolutional neural network, where the derived quantities can be relaxed to learnable parameters. We show that by using discrete Fourier transform based pilots the number of learnable network parameters decreases significantly and the online run time of the estimator is reduced considerably, where we can achieve linearithmic order of complexity in the number of antennas. Numerical results demonstrate performance gains compared to state-of-the-art algorithms from the field of compressive sensing or covariance estimation of the same or even higher computational complexity. The simulation code is available online.

cs.IT↗

arXiv2015

Non-linear two-photon resonance fluorescence on a single artificial atom

We report two-photon resonance fluorescence of an individual semiconductor artificial atom. By non-linearly driving a single quantum dot via a two-photon transition, we probe the linewidth of the two-photon processes and show that, similar to their single-photon counterparts, they are close to being Fourier limited at low temperatures. The evolution of the population of excitonic states with the Rabi frequency exhibits a clear s-shaped behavior, indicative of the non-linear response via the two-photon excitation process. We model the non-linear response using a 4-level atomic system representing the manifold of excitonic and biexcitonic states of the quantum dot and show that quantitative agreement is obtained only by including the interaction with LA-phonons in the solid state environment. Finally, we demonstrate the formation of dressed states emerging from a two-photon interaction between the artificial atom and the excitation field. The non-linear optical dressing induces a mixing of all four excitonic states that facilitates the tuning of the polarization selection rules of the artificial atom.

cond-mat.mes-hall↗