Research on Blind Recognition Algorithm of Channel Coding Based on One-Dimensional Convolutional Neural Network Under the Low SNR Regime
Pan Deng,Tianqi Zhang,Baoze Ma,Zeliang An
DOI: https://doi.org/10.1007/s11063-022-11104-0
IF: 2.565
2023-01-03
Neural Processing Letters
Abstract:To solve the problem of blind identification of channel coding in signal interception and intelligent communication systems, a blind recognition algorithm of channel coding based on a convolutional neural network (CNN) is proposed. To deal better with the input characteristics of one-dimensional soft decision sequences under the low SNR regime, we design and construct a low-complexity one-dimensional CNN classifier. Firstly, the dataset of five types of channel codes is generated, including linear block codes, Turbo codes, convolutional codes, LDPC codes and Polar codes. Secondly, the one-dimensional CNN classifier is initialized and imported into the database. Thirdly, the training dataset.mat file generated in matlab is imported into the convolutional neural network, and the dataset of five types of channel codes is labeled. Finally, a one-dimensional convolutional neural network model is established, and the dataset in the model is compiled and trained, the training data tags in the model are predicted and the confusion matrix is saved. The simulation results show that the proposed algorithm expands the recognition range of existing types of channel codes, and for the first time completes the blind recognition of five types of channel codes. Compared with the traditional algorithm LSTM (Long Short Term Memory), the algorithm proposed in this paper has lower complexity, the recognition rate can be kept 92% when the SNR is 0 dB, and the performance is improved by 24%.
computer science, artificial intelligence