|Table of Contents|

[1] Ouyang Xingchen, Wu Lenan,. Faster-than-Nyquist rate communicationvia convolutional neural networks-based demodulators [J]. Journal of Southeast University (English Edition), 2016, 32 (1): 6-10. [doi:10.3969/j.issn.1003-7985.2016.01.002]
Copy

Faster-than-Nyquist rate communicationvia convolutional neural networks-based demodulators()
基于CNN解调器的超奈奎斯特速率通信
Share:

Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]

Volumn:
32
Issue:
2016 1
Page:
6-10
Research Field:
Information and Communication Engineering
Publishing date:
2016-03-20

Info

Title:
Faster-than-Nyquist rate communicationvia convolutional neural networks-based demodulators
基于CNN解调器的超奈奎斯特速率通信
Author(s):
Ouyang Xingchen Wu Lenan
School of Information Science and Engineering, Southeast University, Nanjing 210096, China
欧阳星辰吴乐南
东南大学信息科学与工程学院, 南京 210096
Keywords:
bipolar extended binary phase shifting keying(EBPSK) convolutional neural networks(CNNs) faster-than-Nyquist(FTN)rate double-symbol united-decision
双极性EBPSK 卷积神经网络 超奈奎斯特速率 双码元联合判决
PACS:
TN911.3
DOI:
10.3969/j.issn.1003-7985.2016.01.002
Abstract:
A demodulator based on convolutional neural networks(CNNs)is proposed to demodulate bipolar extended binary phase shifting keying(EBPSK)signals transmitted at a faster-than-Nyquist(FTN)rate, solving the problem of severe inter symbol interference(ISI)caused by FTN rate signals. With the characteristics of local connectivity, pooling and weight sharing, a six-layer CNNs structure is used to demodulate and eliminate ISI. The results show that with the symbol rate of 1.07 kBd, the bandwidth of the band-pass filter(BPF)in a transmitter of 1 kHz and the changing number of carrier cycles in a symbol K=5, 10, 15, 28, the overall bit error ratio(BER)performance of CNNs with single-symbol decision is superior to that with a double-symbol united-decision. In addition, the BER performance of single-symbol decision is approximately 0.5 dB better than that of the coherent demodulator while K equals the total number of carrier circles in a symbol, i.e., K=N=28. With the symbol rate of 1.07 kBd, the bandwidth of BPF in a transmitter of 500 Hz and K=5, 10, 15, 28, the overall BER performance of CNNs with double-symbol united-decision is superior to those with single-symbol decision. Moreover, the double-symbol united-decision method is approximately 0.5 to 1.5 dB better than that of the coherent demodulator while K=N=28. The demodulators based on CNNs successfully solve the serious ISI problems generated during the transmission of FTN rate bipolar EBPSK signals, which is beneficial for the improvement of spectrum efficiency.
针对超奈奎斯特速率传输信号在传输过程中产生的严重码间干扰问题, 提出了一种基于卷积神经网络(CNN)的解调器, 对双极性扩展的二进制相移键控(bipolar EBPSK)超奈奎斯特速率信号进行解调.利用卷积神经网络局部感受野、池化和权值共享的特点, 提出了一种具有6层结构的卷积神经网络来解调扩展的二进制相移键控调制信号并消除码间干扰.实验结果表明:当码率为1.07 kBd、发送端带宽限制为1 kHz, 且一个码元中跳变载波周期数K=5, 10, 15, 28时, CNN单码元判决方法误码率性能总体优于CNN双码元联合判决方法;当K等于码元载波周期总数N, 即K=N=28时, CNN单码元判决误码率方法优于相干解调约0.5 dB;当码率为1.07 kBd、发送端带宽限制为500 Hz, 且K=5, 10, 15, 28时, CNN双码元联合判决方法优于CNN码元判决方法;当K=N=28时, CNN双码元判决方法优于相干解调约0.5~1.5 dB.基于CNN的解调器成功地解决了由超奈奎斯特速率双极性传输信号产生的严重码间干扰问题, 有利于频谱利用率的提高.

References:

[1] Yang D J, Fang X, Xue G L, Tang J. Relay station placement for cooperative communications in WiMAX networks [C]//IEEE Global Telecommunications Conference. Miami, USA, 2010: 4244-4245.
[2] Sun T Q, An Y L. Study on video transmission system of central heating based on 3G wireless network [J]. Applied Mechanics and Materials, 2014, 536-537: 157-160.
[3] Du J H, Li Y, Chen S P, Wang W B. Modeling HSDPA in TDD-CDMA/SA systems[C]//IEEE International Symposium on Microwave, Antenna, Propagation and EMC Technologies for Wireless Communications. Beijing, China, 2005, 2: 1501-1505.
[4] Lecun Y, Bengio Y. Convolutional networks for images, speech, and time-series[C]//The Handbook of Brain Theory and Neural Networks. Cambridge: MIT Press, 1995.
[5] Montúfar-Chaveznava R, Guinea D, Garcia-Alegre M C, et al. CNN computer for high speed visual inspection [J]. Proceedings of SPIE—The International Society for Optical Engineering, 2001, 4301: 236-243.
[6] Ossama A H, Mohamed A R. Convolutional neural network for speech recognition [J]. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014, 22(10): 1533-1544.
[7] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks [J]. Advances in Neural Information Processing Systems, 2012, 25(2): 1097-1105.
[8] Zhang S K. EBPSK modulation with very high bandwidth efficiency [C]//2011 International Conference on Information Science and Technology. Nanjing, China, 2011:592-595.
[9] Feng M, Wu L N, Ding J, et al. BER analysis and verification of EBPSK system in AWGN channel [J]. IEICE Transactions on Communications, 2011, E94-B(3): 806-809.
[10] Feng M, Qi C H, Wu L N.Analysis and optimization of power spectrum on EBPSK modulation in throughput-efficient wireless system [J]. Journal of Southeast Univershity: English Edition, 2008, 24(2): 143-148.
[11] Andrisano O, Chiani M. First Nyquist criterion applied to coherent receiver design for generalized MSK signals [J]. IEEE Transactions on Communications, 1994, 42(2/3/4): 449-457.
[12] Olshausen B A, Field D J. Emergence of simple-cell receptive field properties by learning a sparse code for natural images [J]. Nature, 1996, 381(6583): 607-609. DOI:10.1038/381607a0.
[13] Feng M, Wu L N, Gao P. An EBPSK demodulator based on ANN detection [C]//The 2nd International Conference on Information Science and Engineering. Hangzhou, China, 2010: 969-972.
[14] Muller J, Muller J, Tetzlaff R. Hierarchical description and analysis of CNN algorithms [C]//International Workshop on Cellular Nanoscale Networks and their Applications. Notre Dame, IN, USA, 2014: 14579612-1-14579612-2.
[15] Shao Q, Feng C J. Pattern recognition of chatter gestation based on hybrid PCA-SVM [J]. Applied Mechanics and Materials, 2012, 120: 190-194.
[16] Chen X Q, Wu L N. Nonlinear detection for a high rate extended binary phase shift keying system [J]. Sensors, 2013, 13(4): 4327-4347. DOI:10.3390/s130404327.

Memo

Memo:
Biographies: Ouyang Xingchen(1992—), female, graduate; Wu Lenan(corresponding author), male, doctor, professor, wuln@seu.edu.cn.
Foundation item: The National Natural Science Foundation of China(No.6504000089).
Citation: Ouyang Xingchen, Wu Lenan. Faster-than-Nyquist rate communication via convolutional neural networks-based demodulators[J].Journal of Southeast University(English Edition), 2016, 32(1):6-10. DOI:10.3969/j.issn.1003-7985.2016.01.002.
Last Update: 2016-03-20