Path Loss Characterization Using Machine Learning Models for GS-to-UAV-Enabled Communication in Smart Farming Scenarios

Algorithm 1

SVR.

	Hyperparameter setting:
(1)	Let the be the RSSI training data sets and be the path loss training data set. The data set was separated as 80% for the training set and 20% for the test set.
(2)	Let the function and be the basis function for RSSI and path loss of the hyperparameter setting, respectively. The hyperplane in the feature space is and , where and denote the slope and intercept of the regression model.
	Model training:
(3)	The predicted value is and for RSSI and path loss, respectively, where the value controls the iteration between the test and training sets, and represent the average mean of RSSI and path loss test set, and is the Euclidean norm.