ahmarey/plgan

Predicting point-wise path loss through satellite images and height maps using deep learning.

PL-GAN Models

IEEE | BibTeX

PL-GAN: Path Loss Prediction using Generative Adversarial Networks
Ahmed Marey, Mustafa Bal, Hasan Ates, Bahadir Gunturk

This is a repository for the Path Loss GAN (PL-GAN) project. PL-GAN infers excess path loss of an area from a satellite or height map image.

Running the evaluation code infer.py will generate the path loss images and the statistical comparison between height map input and satellite image input.

The test set can be downloaded from here. Place it in the test folder and run infer.py.

The figure below shows input images and results for some sample regions.

BibTeX

@ARTICLE{9866771,
  author={Marey, Ahmed and Bal, Mustafa and Ates, Hasan F. and Gunturk, Bahadir K.},
  journal={IEEE Access}, 
  title={PL-GAN: Path Loss Prediction Using Generative Adversarial Networks}, 
  year={2022},
  volume={10},
  number={},
  pages={90474-90480},
  doi={10.1109/ACCESS.2022.3201643}}

plgan's Issues

Help source code

Hello，Recently, I found one of your articles, titled PL-GAN: Path Loss Prediction Using Generative Adversarial Networks in IEEE Access. This article has deeply inspired me.This would make a really positive contribution to my work.I am wondering if you could kindly send me the source code for GAN model.This part was not shared on Github. I promise they will be used only for research purposed.
Thank you very much for your kind consideration and I am looking forward to your early reply.

Height map cannot be retrieved