This project is done as a part of IDP 2020 : Theme is Communication Systems !!
The projext proposal is : HERE
A detailed report of the project is provided in : HERE
YOUTUBE LINKS FOR BETTER EXPLANATION : They are two parts made as a part of mid term and end term evaluation.
MID TERM VIDEO : PART 1
END TERM VIDEO : PART 2
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
Brief Summary of the project :
Primary Aim :
➔ Since WiFi signals are omnipresent these days. We can easily find a WiFi access point in a modern household. So we plan to build a ubiquitous Wi-Fi-based Gesture Recognition System.
➔ We are planning to build a system that can detect simple gestures like swiping or pushing. Thereafter we plan to use this system to control IoT devices through gestures.
➔ We collect different training data samples for a given gesture and feed it to an ML model.
➔ Metric to be used in the data set will either be Received signal strength indicator(RSSI) (or) Channel state information(CSI).
➔ We will be using a Machine learning model to predict the event performed.
➔ The Machine learning algorithm will be a supervised learning classification algorithm and we have not decided on the specific algorithm.
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><> A small detour from the prime aim :
We gathered some data from a company , the data consists of .dat files which has the csi values for collected between receivers when a man moves from one point to the other over a time span of 30 secs !
And it consists of 30 sub carriers with a sampling rate of 1Khz !
We are trying to analyse the data and thus want to predice whether there is a movement or not !!!!
WORK IN PROGRESS !!!
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
As of 10 Aug. 2020, the project has been completed