@proceedings {621, title = {Ionosphere Plasma Density Estimation by Ray Tracing Optimization}, year = {2022}, month = {03/2022}, publisher = {HamSCI}, address = {Huntsville, AL}, abstract = {

In recent years, several studies have tried to estimate volumetric electron density by methods of refraction tomography on an HF network. These methods involve a dynamic optimization problem where the ray tracing equations have to be solved in every optimization step. Furthermore, to improve the estimates, data from incoherent scatter radars and GPS can also be assimilated. However, the computational complexity involved in these estimates is considerable. Even though some efforts have been implemented to reduce this complexity, it is clear that new methods have to be explored. Furthermore, to our knowledge, the possibility of using the frequency sweep capability of ionosondes has not been considered. This work simplifies the dispersion relation to an unmagnetized collisionless plasma to focus our efforts on the inverse process. Instead of using sensitivity analysis, we propose a direct collocation approach, where the points on the transmitter and receiver can be fixed, therefore, eliminating the chances of the extreme misfire.

}, author = {David de la Torre and Enrique Rojas} } @proceedings {610, title = {Modeling ionograms with Deep Neural Networks: Applications to Nowcasting}, year = {2022}, month = {03/2022}, publisher = {HamSCI}, address = {Huntsville, AL}, abstract = {

The state parameters of the ionosphere are of fundamental importance not only for space weather studies but also for technological applications such as satellite radio communications. As with many geophysical phenomena, the ionosphere dynamics are governed by nonlinear processes that make ionospheric forecasting a challenging endeavor. However, we have enormous datasets and ubiquitous experimental sources that can help us find the complex regularities in these phenomena. We forecasted ionograms for different solar activity times and database sizes using regression deep neural networks. Due to the neural network{\textquoteright}s extrapolation of virtual heights for all frequencies given to the model, we estimated foF2 using two embedded different models to identify the last frequency of each ionogram. Furthermore, we made hyperparameter tuning for each training set applying the k-fold cross-validation method. The predictions were compared to measurements collected with the Digisonde system at the Jicamarca Radio Observatory, a persistence model, IRI, and the SAMI2 model estimations. Finally, we will present preliminary results on a new virtual heights model that predicts the difference between consecutive ionograms and preliminary results from a model to estimate electron densities.

}, author = {Jhassmin Aricoch{\'e} and Enrique Rojas and Marco Milla and Reynaldo Rojas} }