Abstract: To improve the accuracy of modulated signal recognition in variable environments and reduce the impact of factors such as lack of prior knowledge on recognition results, researchers have gradually adopted deep learning techniques to replace traditional modulated signal processing techniques. To address the problem of low recognition accuracy of the modulated signal at low signal-to-noise ratios, we have designed a novel modulation recognition network of multi-scale analysis with deep threshold noise elimination to recognize the actually collected modulated signals under a symmetric cross-entropy function of label smoothing. The network consists of a denoising encoder with deep adaptive threshold learning and a decoder with multi-scale feature fusion. The two modules are skip-connected to work together to improve the robustness of the overall network. Experimental results show that this method has better recognition accuracy at low signal-to-noise ratios than previous methods. The network demonstrates a flexible self-learning capability for different noise thresholds and the effectiveness of the designed feature fusion module in multi-scale feature acquisition for various modulation types.

具有深度阈值噪声消除的多尺度分析调制识别网络

[1]An ZL, Zhang TQ, Shen M, et al., 2022. Series-constellation feature based blind modulation recognition for beyond 5G MIMO-OFDM systems with channel fading. IEEE Trans Cogn Commun Netw, 8(2):793-811.

[2]Chang SG, Yu B, Vetterli M, 2000. Adaptive wavelet thresholding for image denoising and compression. IEEE Trans Image Process, 9(9):1532-1546.

[3]Dahap BI, Hongshu L, 2015. Advanced algorithm for automatic modulation recognition for analogue & digital signals. Proc Int Conf on Computing, Control, Networking, Electronics and Embedded Systems Engineering, p.32-36.

[4]Doan VS, Huynh-The T, Hoang VP, et al., 2022. MoDANet: multi-task deep network for joint automatic modulation classification and direction of arrival estimation. IEEE Commun Lett, 26(2):335-339.

[5]Donoho DL, 1995. De-noising by soft-thresholding. IEEE Trans Inform Theory, 41(3):613-627.

[6]Eltaieb RA, Abouelela HAE, Saif WS, et al., 2020. Modulation format identification of optical signals: an approach based on singular value decomposition of Stokes space projections. Appl Opt, 59(20):5989-6004.

[7]Han H, Ren ZY, Li L, et al., 2021. Automatic modulation classification based on deep feature fusion for high noise level and large dynamic input. Sensors, 21(6):2117.

[8]Harris FJ, 1978. On the use of windows for harmonic analysis with the discrete Fourier transform. Proc IEEE, 66(1):51-83.

[9]Huang G, Liu Z, van der Maaten L, et al., 2017. Densely connected convolutional networks. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.2261-2269.

[10]Huang S, Yao YY, Wei ZQ, et al., 2017. Automatic modulation classification of overlapped sources using multiple cumulants. IEEE Trans Veh Technol, 66(7):6089-6101.

[11]Jia HR, Zhang XY, Bai J, 2013. A continuous differentiable wavelet threshold function for speech enhancement. J Cent South Univ, 20(8):2219-2225.

[12]Kline DM, Berardi VL, 2005. Revisiting squared-error and cross-entropy functions for training neural network classifiers. Neur Comput Appl, 14(4):310-318.

[13]Li B, Chen XF, 2014. Wavelet-based numerical analysis: a review and classification. Fin Elem Anal Des, 81:14-31.

[14]Li L, Dong ZY, Zhu ZG, et al., 2023. Deep-learning hopping capture model for automatic modulation classification of wireless communication signals. IEEE Trans Aerosp Electron Syst, 59(2):772-783.

[15]Li LX, Huang JS, Cheng QQ, et al., 2021. Automatic modulation recognition: a few-shot learning method based on the capsule network. IEEE Wirel Commun Lett, 10(3):474-477.

[16]Li T, Liu W, Jiang XY, et al., 2020. Modulation classification in successive relaying systems with interference. IEEE/CIC Int Conf on Communications in China, p.1022-1026.

[17]Liu YB, Liu Y, Yang C, 2020. Modulation recognition with graph convolutional network. IEEE Wirel Commun Lett, 9(5):624-627.

[18]Meng F, Chen P, Wu LN, et al., 2018. Automatic modulation classification: a deep learning enabled approach. IEEE Trans Veh Technol, 67(11):10760-10772.

[19]O'Shea TJ, West N, 2016. Radio machine learning dataset generation with GNU radio. Proc 6^th GNU Radio Conf, p.1-6.

[20]O'Shea TJ, Roy T, Clancy TC, 2018. Over-the-air deep learning based radio signal classification. IEEE J Sel Top Signal Process, 12(1):168-179.

[21]Peng SL, Sun SJ, Yao YD, 2022. A survey of modulation classification using deep learning: sign representation and data preprocessing. IEEE Trans Neur Netw Learn Syst, 33(12):7020-7038.

[22]Phukan GJ, Bora PK, 2018. Blind equalization for classification of digital modulations. Int Conf on Signal Processing and Communications, p.472-476.

[23]Rawat W, Wang ZH, 2017. Deep convolutional neural networks for image classification: a comprehensive review. Neur Comput, 29(9):2352-2449.

[24]Salam AOA, Sheriff RE, Hu YF, et al., 2019. Automatic modulation classification using interacting multiple model Kalman filter for channel estimation. IEEE Trans Veh Technol, 68(9):8928-8939.

[25]Schmidhuber J, 2015. Deep learning in neural networks: an overview. Neur Netw, 61:85-117.

[26]Sendur L, Selesnick IW, 2002. Bivariate shrinkage functions for wavelet-based denoising exploiting interscale dependency. IEEE Trans Signal Process, 50(11):2744-2756.

[27]Serbes A, Cukur H, Qaraqe K, 2020. Probabilities of false alarm and detection for the first-order cyclostationarity test: application to modulation classification. IEEE Commun Lett, 24(1):57-61.

[28]Szegedy C, Liu W, Jia YQ, et al., 2015. Going deeper with convolutions. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.1-9.

[29]Szegedy C, Vanhoucke V, Ioffe S, et al., 2016. Rethinking the inception architecture for computer vision. Proc IEEE Conf on Computer Vision and Pattern Recognition, p.2818-2826.

[30]Tayakout H, Dayoub I, Ghanem K, et al., 2018. Automatic modulation classification for D-STBC cooperative relaying networks. IEEE Wirel Commun Lett, 7(5):780-783.

[31]Wang QL, Wu BG, Zhu PF, et al., 2020. ECA-Net: efficient channel attention for deep convolutional neural networks. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.11531-11539.

[32]Wang YS, Ma XJ, Chen ZY, et al., 2019. Symmetric cross entropy for robust learning with noisy labels. Proc IEEE/CVF Int Conf on Computer Vision, p.322-330.

[33]Wei YC, Xiao HX, Shi HH, et al., 2018. Revisiting dilated convolution: a simple approach for weakly- and semi-supervised semantic segmentation. Proc IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.7268-7277.

[34]Wei YJ, Fang SL, Wang XY, 2019. Automatic modulation classification of digital communication signals using SVM based on hybrid features, cyclostationary, and information entropy. Entropy, 21(8):745.

[35]Xu JL, Luo CB, Parr G, et al., 2020. A spatiotemporal multi-channel learning framework for automatic modulation recognition. IEEE Wirel Commun Lett, 9(10):1629-1632.

[36]Xu YQ, Xu GX, Ma C, et al., 2022. An advancing temporal convolutional network for 5G latency services via automatic modulation recognition. IEEE Trans Circ Syst II Expr Briefs, 69(6):3002-3006.

[37]Zhang ZF, Wang C, Gan CQ, et al., 2019. Automatic modulation classification using convolutional neural network with features fusion of SPWVD and BJD. IEEE Trans Signal Inform Process Netw, 5(3):469-478.

[38]Zhu MT, Li YJ, Pan ZS, et al., 2020. Automatic modulation recognition of compound signals using a deep multi-label classifier: a case study with radar jamming signals. Signal Process, 169:107393.