Research on Hierarchical Recognition Method for Long Channel Coding Types Based on Deep Learning

In the process of signal transmission, the receiver obtains the data without channel decoding after a series of processing such as synchronization and demodulation of the received signal. In the case of non-cooperation, the encoded data cannot be directly decoded because the coding type of the data is unknown, and the coding type needs to be recognized first. In this paper, a hierarchical channel coding recognition scheme based on deep learning is proposed, which employs CNNBiLSTM and CNN-LSTM models for multi-stage recognition. The system sequentially recognizes different coding types using raw encoded data, extracted code features, and IP matrices. Experimental results show that the proposed method achieves a recognition accuracy of 99.76% at bit error rate of 10<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−3</sup>, delivering robust recognition performance.