Wi-Fi signal strength measurements from smartphone for various hand gestures

Citation Author(s):
Mohamed Haseeb, Ramviyas Parasuraman
Submitted by:
Ramviyas Parasuraman
Last updated:
Thu, 11/08/2018 - 10:34
Data Format:
0 ratings - Please login to submit your rating.


The dataset is an extensive collection of labeled high-frequency Wi-Fi Radio Signal Strength (RSS) measurements corresponding to multiple hand gestures made near a smartphone under different spatial and data traffic scenarios. We open source the software code and an Android app (Winiff) to create this dataset, which is available at Github (https://github.com/mohaseeb/wisture). The dataset is created using an artificial traffic induction (between the phone and the access point) approach to enable useful and meaningful RSS values. We invite researchers, engineers, and the App development community to test and improve the Winiff application and enable crowd-sourced smartphone Wi-Fi RSS datasets available for various machine learning related researches. The Winiff app includes options to enable or disable traffic induction and supports both positive and negative RSSI values reported by the Wi-Fi device drivers on the smartphone.



This is a dataset of Wi-Fi Received Signal Strength Indicator (RSSI) values collected in an Android phone (Samsung Galaxy S7) while performing a set of over-the-air hand gestures (Push, Pull, and Swipe) around the phone. It was collected as part of the work reported in the paper: "Wisture: RNN-based Learning of Wireless Signals for Gesture Recognition in Unmodified Smartphones".


Each file contains a series of RSSI values belonging to a single collection session, recorded while performing the same hand gesture repeatedly. A collection session proceeded as below:

* The AP and the smartphone are configured according to specific spacial setup and traffic scenario.

* The subject performing the experiment sits in a chair facing the smartphone.

* The smartphone is connected to the AP.

* The RSSI collection application is started.

* At a specific point in time (start time), the subject starts performing the gestures. Consecutive gestures are separated by a ten seconds gap (gap time).

* Both the start and gap times are noted down and used later to extract the gesture windows (portions of the collected RSSI stream that correspond to hand gestures).

* The collected RSSI stream is stored by the phone in a text file with a name specifying the performed gesture. The file is then exported to a PC.



File names format:





<part1>: Performed gesture name (swipe|push|pull)

<part2>: Time in seconds that marks the start of the first gesture

<part3>: The time gap in seconds between each two consecutive gestures

<part4>: Absoluate time when the collection started

<part5>: Absoluate time when the collection ended

<part6>: Spatial placement of the smartphone and AP; (same-room: when both are placed in the same room | two-rooms: if they are placed in different rooms)*

<part7>: (induction: if artificial traffic is being induced between the smartphone and the AP | no-induction: if no artificial traffic is induced)*

<part8>: (internet: if Internet is enabled in the smarthpone via the AP | no-internet: if Internet is disabled)


* Refer to the paper "Wisture: RNN-based Learning of Wireless Signals for Gesture Recognition in Unmodified Smartphones" for details on the collection setup and procedure.


File format:



(All fields are separated by a 'tab' character)

first row: columns labels, i.e. (time          rssi)

remaining rows: <c1>    <c2>


<c1>: nanoseconds since some fixed arbitrary time. These values do not correspond to wall clock time but give a precise time measure for RSSI recoding times, relative to each other.

<c2>: RSSI value