Human Activities, Gestures and Fall detection using Channel State Information (CSI) of IEEE-802.11n Devices

This repository is to make the dataset collected for Human activities using the Channel State information (CSI) of IEEE 802.11n devices. As of this time, we are waiting for our papers to be published.

UPDATE (2021-09-19): If you are interested in our dataset please contact me at [email protected]

(i) Our first paper has been published using preliminary results: WiHACS: Leveraging WiFi for human activity classification using OFDM subcarriers' correlation https://ieeexplore.ieee.org/document/8308660

(ii) The thesis outlining details can be found here: Using Wi-Fi channel state information (CSI) for human activity recognition and fall detection https://open.library.ubc.ca/cIRcle/collections/ubctheses/24/items/1.0365967

We are currently waiting for our other papers to be published. Once published we will upload all our data, and also the source codes for our algorithms. (Sorry for the delay-- the uploading delay is due to some technical problems). I will update this repo from time to time.

All the data will be presented in MAT or CSV files as time-series. The signal processing matlab files will also be uploaded. Including the python files for machine and deep learning algorithms used.

Any questions about this repo or the papers/thesis, please email at [email protected]

If you use the provided matlab codes and/or dataset in this rep, please consider citing the following reference:

@phdthesis{Chowdhury_2018, series={Electronic Theses and Dissertations (ETDs) 2008+}, title={Using Wi-Fi channel state information (CSI) for human activity recognition and fall detection}, url={https://open.library.ubc.ca/collections/ubctheses/24/items/1.0365967}, DOI={http://dx.doi.org/10.14288/1.0365967}, school={University of British Columbia}, author={Chowdhury, Tahmid Z.}, year={2018}, collection={Electronic Theses and Dissertations (ETDs) 2008+}}