Physics-informed neural networks for pathloss prediction
This incremental approach addresses pathloss prediction for wireless communication applications, potentially aiding tasks like localization.
The paper tackles pathloss prediction by integrating physical dependencies and measured data into neural network training, resulting in improved generalization and prediction quality with fewer layers and parameters, enabling fast inference and reduced data requirements.
This paper introduces a physics-informed machine learning approach for pathloss prediction. This is achieved by including in the training phase simultaneously (i) physical dependencies between spatial loss field and (ii) measured pathloss values in the field. It is shown that the solution to a proposed learning problem improves generalization and prediction quality with a small number of neural network layers and parameters. The latter leads to fast inference times which are favorable for downstream tasks such as localization. Moreover, the physics-informed formulation allows training and prediction with a small amount of training data which makes it appealing for a wide range of practical pathloss prediction scenarios.