This repo is the official implementation of
"Multitarget Device-Free Localization via Cross-Domain Wi-Fi RSS Training Data and Attentional Prior Fusion" (AAAI, 2024)
Na Fan, Zeyue Tian, Amartansh Dubey, Samruddhi Deshmukh, Ross Murch, Qifeng Chen
Abstract: Device-free localization (DFL) using easily-obtained Wi-Fi received signal strength (RSS) has wide real-world applications for not requiring people to carry trackable devices. However, accurate multitarget DFL remains challenging due to the unknown number of targets, multipath interference (MPI), especially between nearby targets, and limited real-world data. In this study, we pioneeringly propose a transformer-based learning method with Wi-Fi RSS as input, and an attentional prior fusion module, to simultaneously locate an unknown number of people at random positions. To overcome the multitarget data collection challenges, we contribute a large-scale cross-domain real-simulation-augmentation training dataset with one and two real-world nearby non-person objects at limited positions and up to five simulated and augmented randomly distributed targets. Experimental results demonstrate our method's improved accuracy, generalization ability, and robustness with fewer Wi-Fi nodes than previous methods.
We are still working on the release of the code. Currently, you can have a try at the demo of a few sample data and check the results.
- Complete Dataset Release & Introduction
- Test Code & Evaluation
- Training Code
-
This code has been tested on Ubuntu18.04.
-
Install
CUDA v10.2
withcudnn v7
following the official installation instructions -
Create Anaconda Environment:
conda create -n RSSDFL python=3.7 ipykernel conda activate RSSDFL python -m ipykernel install --user --name RSSDFL --display-name "RSSDFL"
-
Install
PyTorch>=1.8.0
andtorchvision>=0.9.0
forCUDA v10.2
:conda install pytorch==1.8.0 torchvision==0.9.0 cudatoolkit=10.2 -c pytorch
-
Install other requirements:
pip install -r requirements.txt
Complete dataset to be released...
Please download the pretrained checkpoint demo_ckpt.pth(121.1MB) and put it into the directory ckpt/.
Run demo.ipynb
Code to be released...
Code to be released...
If you find our work useful in your research, please consider citing our paper:
@inproceedings{RSSDFL,
title = {Multitarget Device-Free Localization via Cross-Domain Wi-Fi {RSS} Training Data and Attentional Prior Fusion},
author = {Na Fan and Zeyue Tian and Amartansh Dubey and Samruddhi Deshmukh and Ross D. Murch and Qifeng Chen},
booktitle = {AAAI},
pages = {91--99},
year = {2024},
}