Research Article
Forecasting Parameters of Satellite Navigation Signal through Artificial Neural Networks for the Purpose of Civil Aviation
Table 1
Comparison of network structure in individual models.
| | Model ANN 1. Accuracy | Model ANN 2. Continuity | Model ANN 3. Availability | Model ANN 4. Integrity |
| Number of neurons for input | 5 | 6 | 6 | 6 | Number of neurons in the output layer | 1 | 1 | 1 | 1 | Number of hidden layers | 1 | 1 | 1 | 1 | Number of neurons in the hidden layer | 1 | 8 | 2 | 2 | Division of the training set | 70/15/15 | 80/10/10 | 70/15/15 | 70/15/15 | Training algorithm | Levenberg-Marquardt | Levenberg-Marquardt | Levenberg-Marquardt | Levenberg-Marquardt |
|
|
Source: own study based on structures of artificial neural networks in MALTAB Neural Network Toolbox.
|