Unbalanced protocol identification method based on residual U-Net network

The invention discloses an unbalanced protocol identification method based on a residual U-Net network. The method comprises the following steps: acquiring a network flow to be identified; preprocessing the network flow to be identified to obtain a one-dimensional matrix with a fixed length; and inputting the one-dimensional matrix into a pre-trained unbalanced protocol identification model for feature extraction, and outputting a protocol identification result of the network flow to be identified. According to the method, the SE-Net network is introduced into the residual U-Net network to mine the importance of different channels with protocol data characteristics, and partial channels with the protocol data characteristics are selectively enhanced or inhibited. Meanwhile, a new nonlinear activation function is introduced into the model, when input is activated, the new activation function can effectively reduce information loss, and therefore the feature extraction capacity of the model is improved. In addi