---
title: Coverage Estimation DL
emoji: 🔥
colorFrom: green
colorTo: red
sdk: gradio
sdk_version: 5.12.0
app_file: app.py
pinned: false
license: mit
short_description: This app use deep learning models to radio map estimation (R
---

# Fast Radio Propagation Prediction in WLANs Using Deep Learning

In this research, we present the Deep Learning architecture [UNet](https://arxiv.org/abs/1505.04597) for fast calculation of Radio Maps Estimation (RME) and Cells Maps Estimation (CME) in indoor scenarios. This architecture was implemented for WLAN structures consisting of 1, 2, 3, 4, and 5 access points, with the capability to perform RME and CME similar to a physical simulator, but in a fast manner.

An important reference point in the state of the art was [RadioUNet](https://github.com/RonLevie/RadioUNet), which is an application for estimating path loss propagation in outdoor scenarios.

### General Database Structure

A major initial difficulty for starting the research was the lack of data, in this case, indoor scenario floor plans, coverage maps, and coverage area maps for training the [UNet](https://arxiv.org/abs/1505.04597) architecture. Therefore, it was necessary to create an appropriate database that would facilitate the respective trainings. The coverage maps were generated using the [WiFi IEEE](https://mentor.ieee.org/802.11/dcn/03/11-03-0940-04-000n-tgn-channel-models.doc) model. This implementation was carried out in the MATLAB software [Radio-Indoor-Propagation-Software](https://github.com/johanflorez98/Radio-Indoor-Propagation-Software).

Thus, this research provides a [database](https://doi.org/10.5281/zenodo.8092621) that can be used for training multiple Deep Learning architectures and can facilitate future investigations into similar problems.

## Cite as
```
  @online{andres_j_florez_gonzalez_2023_8092850,
    author       = {Andres J. Florez-Gonzalez and
                    Carlos A. Viteri -Mera},
    title        = {{Fast Indoor Radio Propagation Prediction Using 
                     Deep-Learning App}},
    month        = jun,
    year         = 2023,
    publisher    = {Zenodo},
    version      = {V1.0},
    doi          = {10.5281/zenodo.8092850},
    url          = {https://doi.org/10.5281/zenodo.8092850}
  }
```

## Requirements

matplotlib==3.10.0
numpy==1.26.4
tensorflow==2.17.1
pillow==11.1.0
gradio=5.12.0