Room occupancy analysis using Wi-Fi channel state information

This research shows that using the Wi-Fi CSI for occupancy analysis purposes proves that it can accurately be used to estimate the occupied rooms after gaining training data from the environment. This is done by training an FNN model to recognize different scenarios of room occupancy by collecting channel state information from all the possible occupancy combinations. Additionally, the paper highlights and analyses the parameters that influence the accuracy of the scanning like room count, and people count, and proves that distance does not have a big impact on the efficiency of the model. Even though the approach requires extended research with the implication of more people and the usage of bigger spaces, the system proved accurate enough to be used as an alternative to current detection methods.