| Title | Blind Identification of Channel Codes under AWGN and Fading Conditions via Deep Learning |
| Publication Type | Conference Paper |
| Year of Publication | 2022 |
| Authors | Peng, Haifeng, Cao, Chunjie, Sun, Yang, Li, Haoran, Wen, Xiuhua |
| Conference Name | 2022 International Conference on Networking and Network Applications (NaNA) |
| Keywords | AWGN channels, blind identification, channel codes, channel coding, codes, composability, Deep Learning, fading channels, feature extraction, Metrics, physical layer security, pubcrawl, resilience, Resiliency, Robustness, Wireless communication |
| Abstract | Blind identification of channel codes is crucial in intelligent communication and non-cooperative signal processing, and it plays a significant role in wireless physical layer security, information interception, and information confrontation. Previous researches show a high computation complexity by manual feature extractions, in addition, problems of indisposed accuracy and poor robustness are to be resolved in a low signal-to-noise ratio (SNR). For solving these difficulties, based on deep residual shrinkage network (DRSN), this paper proposes a novel recognizer by deep learning technologies to blindly distinguish the type and the parameter of channel codes without any prior knowledge or channel state, furthermore, feature extractions by the neural network from codewords can avoid intricate calculations. We evaluated the performance of this recognizer in AWGN, single-path fading, and multi-path fading channels, the results of the experiments showed that the method we proposed worked well. It could achieve over 85 % of recognition accuracy for channel codes in AWGN channels when SNR is not lower than 4dB, and provide an improvement of more than 5% over the previous research in recognition accuracy, which proves the validation of the proposed method. |
| DOI | 10.1109/NaNA56854.2022.00020 |
| Citation Key | peng_blind_2022 |
Blind Identification of Channel Codes under AWGN and Fading Conditions via Deep Learning