Action recognition method based on wavelet transform and neural network in wireless network

With the popularization of wireless networks in life, wireless network-based action recognition methods developed better prospects in the fields of human-computer interaction, security and Internet of Things. In order to perform action recognition efficiently and stably, this paper considers combining Butterworth filtering and wavelet transform for denoising, and extracting features through self-organizing competitive neural network, and finally using Softmax regression function for action classification and recognition. Experimental results show that this method has a higher recognition rate and a strong stability.