Research on sound separation and identification of trapped miners based on fastica algorithm

After coal mine gas explosion and other accidents, rapid rescue is an important task to reduce casualties. The use of robots through the rescue channel to quickly enter the post-disaster mine search for trapped miners is currently used as a method. In the search for trapped miners, we use voice recognition to search for trapped miners, but the post-disaster mine environment is complex. Various noise will interfere with the identification of miners' voice, and a signal separation method needs to be studied to obtain the desired signal. In this paper, we use the FastICA to study the blind separation of the sound, and compared with the common blind source separation method JADE algorithm, to verify the superiority of the FastICA algorithm in signal separation.