Mutual coupling correction DOA estimation method based on deep learning (DL)

The invention relates to the technical field of signal processing, in particular to a mutual coupling correction DOA estimation method based on deep learning (DL). In the method, a DNN structure is designed, the DNN structure is composed of a coarse division and mutual coupling correction module and a subdivision mapping module, the coarse division and mutual coupling correction module and the subdivision mapping module are respectively realized by an auto-encoder and a ResNet which are fully connected, and the realization process is as follows: firstly, a signal is processed to obtain a vector containing feature information as input; secondly, the signal source is preliminarily decomposed into a plurality of angle domains through a coarse division and mutual coupling correction module, mutual coupling correction is carried out, and finally prediction results of all subdivision mapping networks are combined through the subdivision mapping network corresponding to each angle domain to determine accurate arriva