Systems and methods for predicting ionospheric electron content

1. A method for predicting total electron content (TEC) in an ionosphere, comprising:

obtaining a dataset;

preprocessing the obtained dataset by cropping and converting images in the dataset to red, green and blue (RGB) arrays;

inputting the dataset into a machine learning (ML) model, wherein the ML model is a combination of a long short-term memory (LSTM) neural network and a generative adversarial network (GAN) neural network;

training the LSTM-GAN ML model for at least 10 epochs, wherein each epoch includes a cycle of the LSTM-GAN ML model with assigned weights, and wherein the assigned weights are adjusted between successive epochs;

predicting, for a predetermined number of days, the TEC using the LSTM-GAN ML model; and

observing a performance improvement in loss after the at least 10 epochs over the obtained dataset based on the prediction,

wherein the prediction is made for a region having a number of ground transmitters satisfying a sparseness criterion.

2. The method of claim 1 , wherein no interpolation of the prediction is performed by the ML model.

3. The method of claim 1 , further comprising:

determining a delay of a signal based on the prediction.

4. The method of claim 3 , further comprising:

determining, for geolocating a transmitter, a location of the signal exiting the ionosphere based on the prediction.

5. The method of claim 1 , wherein the predetermined number of days is 2 to 5 days for (i) an amount of data in the obtained dataset input into the ML model and (ii) an amount of days of the prediction output from the ML model.

6. The method of claim 1 , wherein the obtained dataset has at least one of a data lapse, an amount of noise, and an outlier.

7. The method of claim 1 , wherein the preprocessing further includes (i) performing an extraction from the images and (ii) scaling-down or compressing the cropped images.

8. The method of claim 1 , wherein the preprocessing further includes removing an outline of a map comprised in the images.

9. The method of claim 1 , further comprising:

separately training the LSTM and GAN neural networks,

wherein the combination is performed after the separate trainings of the LSTM and GAN neural networks.

10. The method of claim 1 , wherein the performance improvement of the prediction comprises a root mean square (RMS) error satisfying a smallness criterion, the RMS error being based on a comparison of color intensity of each pixel of RGB arrays between the obtained dataset and the prediction.

11. The method of claim 5 , wherein the predetermined number of days is 3, and

wherein an amount of days of the dataset inputted into the ML model is the same as an amount of days in the prediction such that training of the ML model causes more accurate prediction in deployment.

12. The method of claim 1 , further comprising:

displaying, via a user interface, an image of the predicted TEC.

13. The method of claim 12 , further comprising:

displaying, via the user interface, a representation of a vertical TEC dataset in comparison with the predicted TEC.

14. A method for predicting TEC in an ionosphere, comprising:

obtaining a dataset;

preprocessing the obtained dataset by cropping and converting images in the dataset to red, green and blue (RGB) arrays;

inputting the dataset into a LSTM-GAN ML model;

training the LSTM-GAN ML model for at least 10 epochs, wherein each epoch includes a cycle of the LSTM-GAN model with assigned weights, and wherein the assigned weights are adjusted between successive epochs;

predicting, for a predetermined number of days, the TEC using the LSTM-GAN ML model; and

determining a delay of a signal based on the prediction.

15. The method of claim 14 , wherein the prediction is made for a region having a number of ground transmitters satisfying a sparseness criterion.

16. The method of claim 14 , wherein the predetermined number of days is 2 to 5 days for (i) an amount of data in the obtained dataset input into the ML model and (ii) an amount of days of the prediction output from the ML model.

17. The method of claim 14 , wherein the preprocessing further includes (i) performing an extraction from the images and (ii) scaling-down or compressing the cropped images.

18. The method of claim 14 , further comprising:

displaying, via a user interface, an image of the predicted TEC.

19. A method for predicting TEC in an ionosphere, comprising:

obtaining a dataset;

preprocessing the obtained dataset by cropping and converting images in the dataset to red, green and blue (RGB) arrays;

inputting the dataset into a LSTM-GAN ML model;

training the LSTM-GAN ML model for at least 10 epochs, wherein each epoch includes a cycle of the LSTM-GAN ML model with assigned weights, and wherein the assigned weights are adjusted between successive epochs;

predicting, for a predetermined number of days, the TEC using the LSTM-GAN ML model; and

determining, for geolocating a transmitter, a location of a signal exiting the ionosphere based on the prediction.

20. The method of claim 19 ,

wherein the preprocessing further includes:

(i) performing an extraction from the images,

(ii) scaling-down or compressing the cropped images, and

(iii) removing an outline of a map comprised in the images.