TY - GEN
T1 - Deep Learning based Channel Code Recognition using TextCNN
AU - Qin, Xiongfei
AU - Peng, Shengliang
AU - Yang, Xi
AU - Yao, Yu Dong
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/11
Y1 - 2019/11
N2 - The recognition of channel code of primary user signal is a important task for the full awareness of wireless environment in cognitive radio. Previous solutions to this problem usually suffer from high computational complexity that is not suitable for real-time applications and manual feature extraction that requires experience and expertise. This paper proposes a deep learning based channel code recognition algorithm that extracts features automatically and avoids complicated calculation. Three convolutional codes are considered as the candidate codes. To recognize which channel code has been adopted by the primary user, the received sequence is regarded as a text sentence and then understood by TextCNN. Experimental results show that the proposed algorithm works well and outperforms the max-log-MAP decoding algorithm in recognition accuracy.
AB - The recognition of channel code of primary user signal is a important task for the full awareness of wireless environment in cognitive radio. Previous solutions to this problem usually suffer from high computational complexity that is not suitable for real-time applications and manual feature extraction that requires experience and expertise. This paper proposes a deep learning based channel code recognition algorithm that extracts features automatically and avoids complicated calculation. Three convolutional codes are considered as the candidate codes. To recognize which channel code has been adopted by the primary user, the received sequence is regarded as a text sentence and then understood by TextCNN. Experimental results show that the proposed algorithm works well and outperforms the max-log-MAP decoding algorithm in recognition accuracy.
UR - http://www.scopus.com/inward/record.url?scp=85077984461&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85077984461&partnerID=8YFLogxK
U2 - 10.1109/DySPAN.2019.8935805
DO - 10.1109/DySPAN.2019.8935805
M3 - Conference contribution
AN - SCOPUS:85077984461
T3 - 2019 IEEE International Symposium on Dynamic Spectrum Access Networks, DySPAN 2019
BT - 2019 IEEE International Symposium on Dynamic Spectrum Access Networks, DySPAN 2019
T2 - 2019 IEEE International Symposium on Dynamic Spectrum Access Networks, DySPAN 2019
Y2 - 11 November 2019 through 14 November 2019
ER -