The IceTop array, located at the surface of the IceCube Neutrino Observatory at the South Pole, is currently used as a veto for the in-ice neutrino detection as well as a cosmic-ray detector. Over the years, additional snow has accumulated on the IceTop detectors leading to a reduction in sensitivity and resolution.
In order to mitigate this issue as well as further increase the accuracy of cosmic-ray measurements, a detector enhancement is planned in the next few years. The enhanced array will consist of 32 stations, each comprising 8 scintillation detectors and 3 radio antennas, and will span an area of approximately 1$\,$km$^2$. Specifically, upgrading IceTop with radio antennas will provide precise $X_\mathrm{max}$ measurements, a variable widely used to reconstruct the composition of cosmic rays.

In January 2020, a complete prototype station
was deployed at the South Pole. Following the measurement of cosmic-ray events with the antennas,
we developed the tools necessary to use a template-matching method for energy and $X_\mathrm{max}$ reconstruction.
This template method uses Monte-Carlo simulations and compares them to recorded data. Thus, a set of simulated air showers is created using air shower parameters reconstructed by IceTop as input to the CORSIKA/CoREAS simulation software for each of the measured events.
In this work, the method is applied to measured events, and we will show that it works for a third of the events in the sample. Subsequently, the technique is modified into a log-likelihood minimization, and the results obtained with a simulations-only study of different parameters are discussed. This work concludes with a confirmation that the template method works for air showers recorded with a frequency band of 80 to 300$\,$MHz,
and shows that the log-likelihood method can potentially increase the accuracy of $X_\mathrm{max}$ reconstruction for the complete planned array.