Air-to-Air Path Loss Prediction Based on Machine Learning Methods in Urban Environments

Algorithm 1

Random Forest algorithm for path loss prediction in the AA scenario.

Input:

Training set with responses , where , .

Number of ensemble members .

Training Process:

For to :

Take a bootstrap sample of size from .

Use as the training data to train the th ensemble member by using binary recursive partitioning.

Repeat the following steps recursively for each unsplit node until the stopping criterion is met:

Select features randomly from the available features ( = 5 in this study).

Calculate the square error for each possible splitting point of each feature, and find the best binary split

among all binary splits on the features.

Split the node into two descendant nodes using the best split.

Prediction:

Given a new , the predicted path loss value is obtained by , where

is the prediction of the th ensemble member.