CNN demodulation model with cascade parallel crossing for CPM signals

doi:10.19682/j.cnki.1005-8885.2024.1005

中国邮电高校学报(英文) ›› 2024, Vol. 31 ›› Issue (3): 30-42.doi: 10.19682/j.cnki.1005-8885.2024.1005

• Artificial Intelligence • 上一篇下一篇

CNN demodulation model with cascade parallel crossing for CPM signals

杨嘉晨¹,段瑞枫²,李诚驹¹

1. School of Information Science and Technology, Beijing Forestry University, Beijing 100083, China
2. Engineering Research Center for Forestry-Oriented Intelligent Information Processing of National Forestry and
Grassland Administration, Beijing 100083, China

收稿日期:2022-12-27 修回日期:2023-07-02 出版日期:2024-06-30 发布日期:2024-06-30
通讯作者: 段瑞枫 E-mail:drffighting2008@163.com
基金资助:
the Beijing Natural Science Foundation (L202003).

CNN demodulation model with cascade parallel crossing for CPM signals

Yang Jiachen, Duan Ruifeng, Li Chengju

1. School of Information Science and Technology, Beijing Forestry University, Beijing 100083, China
2. Engineering Research Center for Forestry-Oriented Intelligent Information Processing of National Forestry and
Grassland Administration, Beijing 100083, China

Received:2022-12-27 Revised:2023-07-02 Online:2024-06-30 Published:2024-06-30
Contact: Duan Ruifeng E-mail:drffighting2008@163.com
Supported by:
the Beijing Natural Science Foundation (L202003).

摘要/Abstract

摘要： The continuous phase modulation (CPM) technique is widely used in range telemetry due to its high spectral
efficiency and power efficiency. However, the demodulation performance of the traditional maximum likelihood
sequence detection (MLSD) algorithm significantly deteriorates in non-ideal synchronization or fading channels. To
address this issue, this work proposes a convolutional neural network (CNN) called the cascade parallel crossing
network (CPCNet) to enhance the robustness of CPM signals demodulation. The CPCNet model employs a multiple
parallel structure and feature fusion to extract richer features from CPM signals. This approach constructs feature
maps at different levels, resulting in a more comprehensive training of the model and improved demodulation
performance. Simulation results show that under Gaussian channel, the proposed CPCNet achieves the same bit
error rate (BER) performance as MLSD method when there is no timing error, but with 1/4 symbol period timing
error, the proposed method has 2 dB demodulation gain compared with CNN and convolutional long short-term
memory deep neural network (CLDNN). In addition, under Rayleigh channel, the BER of the proposed method is
reduced by 5% -87% compared to that of MLSD in the wide signal-to-noise ratio (SNR) region.

关键词: continuous phase modulation (CPM), convolutional neural network (CNN), maximum likelihood sequence detection (MLSD), Rayleigh
fading, timing error

Abstract: The continuous phase modulation (CPM) technique is widely used in range telemetry due to its high spectral
efficiency and power efficiency. However, the demodulation performance of the traditional maximum likelihood
sequence detection (MLSD) algorithm significantly deteriorates in non-ideal synchronization or fading channels. To
address this issue, this work proposes a convolutional neural network (CNN) called the cascade parallel crossing
network (CPCNet) to enhance the robustness of CPM signals demodulation. The CPCNet model employs a multiple
parallel structure and feature fusion to extract richer features from CPM signals. This approach constructs feature
maps at different levels, resulting in a more comprehensive training of the model and improved demodulation
performance. Simulation results show that under Gaussian channel, the proposed CPCNet achieves the same bit
error rate (BER) performance as MLSD method when there is no timing error, but with 1/4 symbol period timing
error, the proposed method has 2 dB demodulation gain compared with CNN and convolutional long short-term
memory deep neural network (CLDNN). In addition, under Rayleigh channel, the BER of the proposed method is
reduced by 5% -87% compared to that of MLSD in the wide signal-to-noise ratio (SNR) region.

Key words: continuous phase modulation (CPM), convolutional neural network (CNN), maximum likelihood sequence detection (MLSD), Rayleigh
fading, timing error

中图分类号:

TN911.7

Yang Jiachen, Duan Ruifeng, Li Chengju. CNN demodulation model with cascade parallel crossing for CPM signals[J]. The Journal of China Universities of Posts and Telecommunications, 2024, 31(3): 30-42.

参考文献

[1] WANG L, QI J Z, SONG P. A simplified direct-decision synchronization algorithm for coherent CPM receiver using linear phase approximation. IEEE Communications Letters, 2019, 23(1): 100 -103.

[2] FARÈS H, GLATTLI D C, LOUET Y, et al. Power spectrum density of single side band CPM using lorenztian frequency pulses. IEEE Wireless Communications letters, 2017, 6(6): 786 -789.

[3] LEI G W, LIU Y A, XIAO X F. Evaluation of bit error probability for CPM MIMO systems in Rayleigh channel. Wireless Personal Communications, 2015, 85(3): 585 -595.

[4] GÁL J, CÂMPEANU A, NAFORNIΤA I. Noncoherent demodulation of continuous phase modulated signals using extended Kalman filtering. Proceedings of the 12th Conference on Optimization of Electrical and Electronic Equipment (OPTIM'10), 2010, May 20 - 22, Brasov, Romania. Piscataway, NJ, USA: IEEE, 2010: 724 -727.

[5] HOSSEINNEJAD M, ERFANIAN A, KARAMI M A. A fully digital BPSK demodulator for biomedical application. Microelectronics Journal, 2018, 81: 76 -83.

[6] YANG D W, LIU C H, ZHANG L. CPM-PSWFs signal demodulation method based on waveform coherence. Proceedings of the 20th IEEE International Conference on Communication (ICC'20), 2020, Oct 28 - 31, Nanning, China. Piscataway, NJ, USA: IEEE, 2020: 1237 -1241.

[7] CAO M Q, SUN H R, YUAN X Y, et al. Analysis of the spectrum and the demodulation of narrowband continuous phase modulation. Proceedings of the 2021 International Conference on Wireless Communications and Smart Grid (ICWCSG'21), 2021, Aug 13 - 15, Hangzhou, China. Piscataway, NJ, USA: IEEE, 2021: 43 -48.

[8] ALEKSEEV G A, MARTIROSOV V E. Dynamic characteristics of the BPSK-GLSS demodulator. Proceedings of the 2021 Systems of Signals Generating and Processing in the Field of on Board Communications ( SiPS'21 ), 2021, Mar 16 - 18, Moscow, Russia. Piscataway, NJ, USA: IEEE, 2021: 1 -4.

[9] XI Z P, ZHU J, FU Y M. Low-complexity detection of binary CPM with small modulation index. IEEE Communications Letters, 2016, 20(1): 57 -60.

[10] TENG F, LIAO Y R, LI Y T. GMSK coherent demodulation technology based on laurent decomposition. Proceedings of the 3rd International Conference on Computer Engineering, Information Science & Application Technology (ICCIA'19), 2019, May 30 - 31, Chongqing, China. Paris, France: Atlantis Press, 2019: 542 -549.

[11] NOJIMA D, NAGAO Y, KUROSAKI M, et al. Soft decision viterbi decoder for FSK demodulation under fast fading channel. Proceedings of the 2010 International Conference on Communications and Electronics (ICCE'10), 2010, Aug 11 -13, Nha Trang, Vietnam. Piscataway, NJ, USA: IEEE, 2010: 222 - 227.

[12] DAWID H, JOERESSEN O J, MEYR H. Viterbi decoders: High performance algorithms and architectures. Parhi K K, Nishitani T (eds). Digital Signal Processing for Multimedia Systems. Boca Raton, FL, USA: CRC Press, 1999: 417 -459.

[13] DEVI T K, PRIYANKA E B, SAKTHIVEL P, et al. Low complexity modified viterbi decoder with convolution codes for power efficient wireless communication. Wireless Personal Communications, 2022, 122(1): 685 -700.

[14] MARQUET A, MONTAVONT N, PAPADOPOULOS G Z. Towards an SDR implementation of LoRa: reverse-engineering demodulation strategies and assessment over Rayleigh channel. Computer Communications, 2020, 153(C): 595 -605.

[15] HAMDI K A. Capacity of MRC on correlated Rician fading channels. IEEE Transactions on Communications, 2008, 56(5): 708 -711.

[16] ALBAWI S, MOHAMMED T A, AL-ZAWI S. Understanding of a convolutional neural network. Proceedings of the 2017 International Conference on Engineering and Technology (ICET'17), 2017, Aug 21 - 23, Antalya, Turkey. Piscataway, NJ, USA: IEEE, 2017: 1 -6.

[17] SHERSTINSKY A. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network. Physica D: Nonlinear Phenomena, 2020, 404: Article 132306.

[18] GREFF K, SRIVASTAVA R K, KOUTNÍK J, et al. LSTM: a search space odyssey. IEEE Transactions on Neural Networks and Learning Systems, 2017, 28(10): 2222 -2232.

[19] WANG Y, LIU M, YANG J, et al. Data-driven deep learning for automatic modulation recognition in cognitive radios. IEEE Transactions on Vehicular Technology, 2019, 68 (4): 4074 - 4077.

[20] YANG C, HE Z M, PENG Y, et al. Deep learning aided method for automatic modulation recognition. IEEE Access, 2019, 7: 109063 -109068.

[21] ZHOU R L, LIU F G, GRAVELLE C W. Deep learning for modulation recognition: A survey with a demonstration. IEEE Access, 2020, 8: 67366 -67376.

[22] CHANG Z X, WANG Y S, LI H, et al. Complex CNN-based equalization for communication signal. Proceedings of the IEEE 4th International Conference on Signal and Image Processing (ICSIP'19), 2019, Jul 19 - 21, Wuxi, China. Piscataway, NJ, USA: 2017: 513 -517.

[23] WADA T, TOMA T, DAWODI M, et al. A denoising autoencoder based wireless channel transfer function estimator for OFDM communication system. Proceedings of the 2019 International Conference on Artificial Intelligence in Information and Communication ( ICAIIC'19), 2019, Feb 11 - 13, Okinawa, Japan. Piscataway, NJ, USA: IEEE, 2019: 530 -533.

[24] CHO Y H, KO H L. Channel estimation based on adaptive denoising for underwater acoustic OFDM systems. IEEE Access, 2020, 8: 157197 -157210.

[25] YE H, LIANG L, LI G Y, et al. Deep learning based end-to-end wireless communication systems with conditional GANs as unknown channels. IEEE Transactions on Wireless Communications, 2020, 19(5): 3133 -3143.

[26] ZHANG H T, ZHANG L, JIANG Y. Overfitting and underfitting analysis for deep learning based end-to-end communication systems. Proceedings of the 11th International Conference on Wireless Communications and Signal Processing ( WCSP'19 ), 2019, Oct 23 -25, Xi'an, China. Piscataway, NJ, USA: IEEE, 2019: 1 -6.

[27] FAN M, WU L N. Demodulator based on deep belief networks in communication system. Proceedings of the 2017 International Conference on Communication, Control, Computing and Electronics Engineering ( ICCCCEE'17), 2017, Jan 16 - 18, Khartoum, Sudan. Piscataway, NJ, USA: IEEE, 2017: 1 -5.

[28] WU T. CNN and RNN-based deep learning methods for digital signal demodulation. Proceedings of the 2019 International Conference on Image, Video and Signal Processing (IVSP'19), 2019, Feb 25 - 28, Shanghai, China. New York, NY, USA: ACM, 2019: 122 -127.

[29] TAN Q Y, ZHAO L. MSK demodulator and impulsive noise depression based on convolutional neural network with gated layers. Proceedings of the IEEE 5th International Conference on Computer and Communications ( ICCC'19), 2019, Dec 6 - 9, Chengdu, China. Piscataway, NJ, USA: IEEE, 2019: 1975 - 1979.

[30] MA S, DAI J H, LU S T, et al. Signal demodulation with machine learning methods for physical layer visible light communications: prototype platform, open dataset, and algorithms. IEEE Access, 2019, 7: 30588 -30598.

[31] CHEN Z, PENG S L, LI H H, et al . Research on blind demodulation system based on deep learning and software defined radio. Journaj of Signal Processing, 2019, 35(4): 649 -655 (in Chinese).

[32] KRIZHEVSKY A, SUTSKEVER I, HINTON G E. ImageNet classification with deep convolutional neural networks. Communications of the ACM, 2017, 60(6): 84 -90.

[33] ZHANG L, ZHANG H T, JIANG Y, et al. Intelligent and reliable deep learning LSTM neural networks-based OFDM-DCSK demodulation design. IEEE Transactions on Vehicular Technology, 2020, 69(12): 16163 -16167.

[34] QIANG Z K, YAN T F, TANG C Y. Demodulation of low SNR QPSK signal based on CNN. Proceedings of the 2021 International Conference on Intelligent Transportation, Big Data & Smart City (ICITBS'21), 2021, Mar 27 - 28, Xi'an, China. Piscataway, NJ, USA: IEEE, 2021: 638 -641.

[35] ZHAO R B, WANG J, LI J Q. An end-to-end demodulation system based on convolutional neural networks. Journal of Physics: Conference Series, 2021, Vol 2026: Proceedings of the 2nd International Conference on Computer Science and Communication Technology (ICCSCT'21), 2021, Jul 29 - 31, Beijing, China. Bristol, UK: IOP Publishing, 2021: Article 12006.

[36] HE D X, WANG Z C. Deep learning-assisted demodulation for TeraHertz communications under hybrid distortions. IEEE Communications Letters, 2022, 26(2): 325 -329.

[37] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR'16), 2016, Jun 27 - 30, Las Vegas, NV, USA. Piscataway, NJ, USA: IEEE, 2016: 770 -778.

[38] SIMONYAN K, ZISSERMAN A. Very deep convolutional networks for large-scale image recognition. Proceedings of the 3rd International Conference on Learning Representations (ICLR'15), 2015, May 7 -9, San Diego, CA, USA. 2015: 1 -4.

[39] LUO W J, LI Y J, URTASUN R, et al. Understanding the effective receptive field in deep convolutional neural networks. Advances in neural information processing systems 29: Proceedings of the 30th Annual Conference on on Neural Information Processing Systems (NIPS'16), 2016, Dec 5 - 10, Barcelona, Spain. Red Hook, NY, USA: Curran Associates, 2016: 4905 -4913.

[40] IOFFE S, SZEGEDY C. Batch normalization: Accelerating deep network training by reducing internal covariate shift. Proceedings of the 32nd International Conference on Machine Learning (ICML'15), 2015, Jul 6 - 11, Lille, France. New York, NY, USA: ACM, 2015: 448 -456.

[41] SAINATH T N, VINYALS O, SENIOR A, et al. Convolutional, long short-term memory, fully connected deep neural networks. Proceedings of the 2015 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP'15), 2015, Apr 19 - 24, South Brisbane, Australia. Piscataway, NJ, USA: IEEE, 2015: 4580 -4584.

[42] O'SHEA T J, CORGAN J, CLANCY T C. Convolutional radio modulation recognition networks. Engineering Applications of Neural Networks: Proceedings of the 17th International Conference on Engineering Applications of Neural Networks ( EANN'16), 2016, Sept 2 -5, Aberdeen, UK. Boston, MA, USA: Springer, 2016: 213 -216.

[43] ZHOU Y, DUAN R F, JIANG B F. Reduced-state noncoherent MLSD method for CPM in aeronautical telemetry. The Journal of China Universities of Posts and Telecommunications, 2020, 27(4): 47 -53.

[44] BLAIECH A G, KHALIFA K B, VALDERRAMA C, et al. A survey and taxonomy of FPGA-based deep learning accelerators. Journal of Systems Architecture, 2019, 98: 331 -345.

[45] FENG L, WANG B T, LIU B, et al. Research on hardware acceleration technology of deep reinforcement learning based on FPGA. Computer Measurement & Control, 2022, 30(6): 242 - 247 (in Chinese).

CNN demodulation model with cascade parallel crossing for CPM signals

CNN demodulation model with cascade parallel crossing for CPM signals

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	束丰张玲华丁寅. Iterative subspace matching pursuit for joint sparse recovery[J]. 中国邮电高校学报(英文版), 2023, 30(2): 26-35.
[2]	Guan Sihai, Cheng Qing, Zhao Yong, Liu Fangyao. Variable step-size adaptive filtering algorithm based on an exponent sin function[J]. 中国邮电高校学报(英文版), 2023, 30(1): 56-65.
[3]	Jiang Fan, Chen Jiajun, Gao Youjun, Sun Changyin. Research on ECG classification based on transfer learning [J]. 中国邮电高校学报(英文), 2022, 29(6): 83-96.
[4]	Li Hui, Li Shanshan, Zou Borong, Chen Yannan. Modulation classification based on the collaboration of dual-channel CNN-LSTM and residual network[J]. 中国邮电高校学报(英文版), 2022, 29(1): 113-124.
[5]	Ma Shexiang, Mei Xiaobing. Side-lobe constrained beamforming under virtual expansion of L-shaped array[J]. 中国邮电高校学报(英文版), 2021, 28(2): 79-88.
[6]	Gao Miao, Yu Zhiguo, Li Qingqing, Wei Jinghe, Gu Xiaofeng. Improved ECG signal compression method for E-healthcare[J]. 中国邮电高校学报(英文版), 2019, 26(4): 62-69.
[7]	张凯尹长川 Xu Guixian 刘丹谱. Robust hybrid beamforming design for millimeter-wave massive MIMO systems[J]. 中国邮电高校学报(英文版), 2019, 26(3): 44-49.
[8]	Yang Jinsheng, Xiang Yang, Chen Weigang, Dong Yangyang. Computationally efficient 2-D DOA estimation for non-uniform two-L-shaped array[J]. 中国邮电高校学报(英文版), 2018, 25(6): 81-89.
[9]	Wang Qi, Wang Shigang, Jia Bowen, Du Hailong. Curvelet transform and contrast adaptive clip histogram equalization-based image defogging algorithm[J]. 中国邮电高校学报(英文版), 2018, 25(2): 96-104.
[10]	Li Ying, Wang Jian, Song Zhanjie. Multivariate spectral estimation based on THREE[J]. Acta Metallurgica Sinica(English letters), 2015, 22(4): 26-32.
[11]	王波赵彦平刘娟娟. Sparse recovery method for far-field and near-field sources localization using oblique projection[J]. Acta Metallurgica Sinica(English letters), 2013, 20(3): 90-96.
[12]	耿烜刘锋何晨. Outage analysis of ANC in the FDD two-way fading channel with channel estimation error[J]. Acta Metallurgica Sinica(English letters), 2013, 20(2): 12-18.
[13]	耿烜何晨孙作雷刘坤. Tomlinson-Harashima precoding minimizing total MSE with transmitter spatial correlation[J]. Acta Metallurgica Sinica(English letters), 2012, 19(6): 19-24.
[14]	赵发勇;酆广增. Joint spatial-frequency blind multiuser detection based on LCCMA[J]. Acta Metallurgica Sinica(English letters), 2007, 14(2): 19-22.
[15]	齐轶; 杜世培; 黄永贵; 齐洪钢. Pre-processing for video coding with rate-distortion optimization decision[J]. Acta Metallurgica Sinica(English letters), 2006, 13(2): 79-83.