ICASSP 2023accepted0 citations

PMNet: Large-Scale Channel Prediction System for ICASSP 2023 First Pathloss Radio Map Prediction Challenge

Ju-Hyung Lee, Joohan Lee, Seon-Ho Lee, Andreas F. Molisch

Abstract

This paper describes our pathloss prediction system submitted to the ICASSP 2023 First Pathloss Radio Map Prediction Challenge. We describe the architecture of PMNet, a neural network we specifically designed for pathloss prediction. Moreover, to enhance the prediction performance, we apply several machine learning techniques, including data augmentation, fine-tuning, and optimization of the network architecture. Our system achieves an RMSE of 0.02569 on the provided RadioMap3Dseer dataset, and 0.0383 on the challenge test set, placing it in the 1st rank of the challenge.

BibTeX
@inproceedings{icassp2023_pmnetlargescalec,
  title = {PMNet: Large-Scale Channel Prediction System for ICASSP 2023 First Pathloss Radio Map Prediction Challenge},
  author = {Ju-Hyung Lee and Joohan Lee and Seon-Ho Lee and Andreas F. Molisch},
  booktitle = {ICASSP 2023},
  year = {2023}
}