NaLiu613 / pmnet

A Scalable and Generalizable Pathloss Map Prediction

A Scalable and Generalizable Pathloss Map Prediction

This repo is the official implementation of "A Scalable and Generalizable Pathloss Map Prediction" as well as the follow-ups.

Introduction

PMNet (Neural network tailored for Pathloss Map Prediction (PMP)) is described in arxiv, which capably serves as a backbone for the PMP task.

PMNet achieves strong performance on the PMP task ($10^{-2}$ level RMSE on val), surpassing previous models by a large margin.

Dataset: Ray-Tracing (RT)-based Channel Measurement (Updating...)

Links for Dataset
USC Dataset
Radiomapseer Reduced
Radiomapseer Orginal

Available checkpoints for PMNet

#	Feature Size	Data-Augmentation	Fine-Tuning	RMSE	Download Link
1	16/H X 16/W	4-way flips	-	0.012599	Download
2	8/H X 8/W	4-way flips	-	0.010570	Download
3	16/H X 16/W	-	UCLA Dataset	0.031449	Download
4	16/H X 16/W	-	Boston Dataset	0.009875	Download

• #3,4 checkpoints were fine-tuned using (1) which is a pre-trained model with USC Dataset.

How to use

To evaluate above models, refer to the following commands.

python [train_eval_file] 'eval' [dataset_directory] [model_to_eval]
# e.g. python train_Boston_pmnet_V1_TL_1.py 'eval' '/Boston/' '/model_0.00010.pt'

Citation

@inproceedings{lee2023pmnet,
title={PMNet: Robust Pathloss Map Prediction via Supervised Learning},
author={Ju-Hyung Lee and Omer Gokalp Serbetci and Dheeraj Panneer Selvam and Andreas F. Molisch},
year={2023},
month={December},
booktitle={Proceedings of IEEE Global Communicaions Conference (GLOBECOM)},
}