Underwater acoustic FBMC communication signal detection method based on deep learning

The invention discloses an underwater acoustic FBMC communication signal detection method based on deep learning. A trained deep neural network model (DNN) is used for replacing a channel estimation module, a channel equalization module and the like in a receiving end of a traditional underwater acoustic FBMC communication system, modular limitation of the system is broken through, underwater acoustic channel state information is learned in a self-adaptive mode, the inherent imaginary part interference influence of an original system is avoided, and the bit error rate performance of the systemis improved. The underwater acoustic FBMC communication system has the beneficial effects that the DNN with complete training is used for replacing the original processes of channel estimation, equalization and the like at the receiving end of the traditional underwater acoustic FBMC communication system. Underwater acoustic channel state information is obtained by using a DNN training stage, anddemodulation recovery of