Underwater Communication Signal Recognition Using Sequence Convolutional Network

Automatic modulation recognition (AMR) is one of the essential parts in intelligent communication system. In underwater acoustic communication, it a challenging work that promptly and easily recognizes signal schemes by conventional methods. The deep neural network method good solution to problem, which creates better effect. packets data are fed familiar constant. However, on course consistently change, seriously reflects veracity. A novel learning with sequence convolutional this paper proposed, composed one-dimensional convolution residual modules variable kernel range. By extracting time-domain characteristics, affection various can be mitigated. experiments, employed not only has more concentrated veracity, but also owns lower parameter quantity shorter training time, indicates ideal results environment. Moreover, valuable real