Human Silhouette Visualization System Using CSI Data

This work aims to understand how a person can influence the propagation of electromagnetic waves in a wireless transmission medium in order to determine his location and visualization. The parameters analyzed in this work were the values of the phases and amplitudes of the signal for 9 pairs of antennas and 56 suborthogonal subcarriers. The article proposes an algorithm that converts the original data into an image with a human silhouette. The images obtained after the algorithm can be used as an auxiliary tool for security alarms. This algorithm does not violate the fact of confidentiality, which is an advantage compared to conventional video surveillance. This algorithm can visualize a person's silhouette with an accuracy of more than 88% on a small amount of training data.