Comparing Machine Learning Methods for Air-to-Ground Path Loss Prediction

2021 ◽  
Author(s):  
George Vergos ◽  
Sotirios P. Sotiroudis ◽  
Georgia Athanasiadou ◽  
George V. Tsoulos ◽  
Sotirios K. Goudos
Keyword(s):  
Machine Learning ◽  
Path Loss ◽  
Learning Methods ◽  
Machine Learning Methods ◽  
Loss Prediction ◽  
Path Loss Prediction

scholarly journals Path Loss Prediction Based on Machine Learning Methods for Aircraft Cabin Environments

IEEE Access ◽  
2019 ◽  
Vol 7 ◽  
pp. 159251-159261 ◽  
Author(s):  
Jinxiao Wen ◽  
Yan Zhang ◽  
Guanshu Yang ◽  
Zunwen He ◽  
Wancheng Zhang
Keyword(s):  
Machine Learning ◽  
Path Loss ◽  
Learning Methods ◽  
Aircraft Cabin ◽  
Machine Learning Methods ◽  
Loss Prediction ◽  
Path Loss Prediction

scholarly journals Path Loss Prediction Based on Machine Learning: Principle, Method, and Data Expansion

2019 ◽  
Vol 9 (9) ◽  
pp. 1908 ◽  
Author(s):  
Yan Zhang ◽  
Jinxiao Wen ◽  
Guanshu Yang ◽  
Zunwen He ◽  
Jing Wang
Keyword(s):  
Machine Learning ◽  
Communication Systems ◽  
Classical Model ◽  
Path Loss ◽  
Measured Data ◽  
Machine Learning Algorithms ◽  
Support Vector ◽  
Loss Prediction ◽  
Path Loss Prediction ◽  
Data Expansion

Path loss prediction is of great significance for the performance optimization of wireless networks. With the development and deployment of the fifth-generation (5G) mobile communication systems, new path loss prediction methods with high accuracy and low complexity should be proposed. In this paper, the principle and procedure of machine-learning-based path loss prediction are presented. Measured data are used to evaluate the performance of different models such as artificial neural network, support vector regression, and random forest. It is shown that these machine-learning-based models outperform the log-distance model. In view of the fact that the volume of measured data sometimes cannot meet the requirements of machine learning algorithms, we propose two mechanisms to expand the training dataset. On one hand, old measured data can be reused in new scenarios or at different frequencies. On the other hand, the classical model can also be utilized to generate a number of training samples based on the prior information obtained from measured results. Measured data are employed to verify the feasibility of these data expansion mechanisms. Finally, some issues for future research are discussed.

Enhancing Machine Learning Models for Path Loss Prediction Using Image Texture Techniques

2021 ◽  
pp. 1-1
Author(s):  
Sotirios Sotiroudis ◽  
Katherine Siakavara ◽  
George Koudouridis ◽  
Panagiotis Sarigiannidis ◽  
Sotirios Goudos
Keyword(s):  
Machine Learning ◽  
Path Loss ◽  
Image Texture ◽  
Learning Models ◽  
Loss Prediction ◽  
Path Loss Prediction ◽  
Machine Learning Models

scholarly journals Comparison of Path Loss Prediction Models for UAV and IoT Air-to-Ground Communication System in Rural Precision Farming Environment

2021 ◽  
pp. 60-66
Author(s):  
Sarun Duangsuwan ◽  
◽  
Myo Myint Maw
Keyword(s):  
Machine Learning ◽  
Random Forest ◽  
Communication System ◽  
Prediction Models ◽  
Path Loss ◽  
Precision Farming ◽  
Machine Learning Algorithms ◽  
Coefficient Of Determination ◽  
Loss Prediction ◽  
Path Loss Prediction

The comparison of path loss model for the unmanned aerial vehicle (UAV) and Internet of Things (IoT) air-to-ground communication system was proposed for rural precision farming. Due to the uncertainty of propagation channel in rural precision farming environment, the comparison of path loss prediction was investigated by the conventional particle swarm optimization (PSO) algorithms: PSO (exponential or Exp), PSO (polynomial or Poly) and the machine learning algorithms: k-nearest neighbor (k-NN), and random forest, are exploited to accurate the path loss models on the basic of the measured dataset. Meanwhile, the empirical model in the rural precision farming was considered. By using the machine learning-based algorithms, the coefficient of determination (R-squared: R2) and root mean squared error (RMSE) were evaluated as highly accuracy and precision more than the conventional PSO algorithms. According to the results, the random forest method was able to perform more than other methods. It has the smallest prediction errors.

Sign in / Sign up

Export Citation Format

Share Document