基于前馈神经网络的非合作PCMA信号盲分离算法
|
摘要
针对非合作接收PCMA混合信号盲分离中高复杂度束缚,提出一种基于前馈神经网络的分离算法,通过搭建神经网络分离平台,规避传统的发送符号遍历思想,实现PCMA混合信号低复杂度高性能盲分离.仿真实验表明,神经网络能够极大挖掘信号内在信息,针对QPSK调制PCMA混合信号,在信噪比7dB时误比特率达到10~(-3)数量级,并伴随着较PSP分离算法算术平方根级别的复杂度降低.
Aiming at the high complexity in blind separation of PCMA mixed signals with non-cooperative reception,the separation algorithm based on feedforward neural network is proposed. By setting up a neural network separation platform and avoiding the traditional idea of maximum a posteriori probability, the blind separation algorithm with low complexity and high performance can be realized. Simulation results show that the neural network can greatly exploit the intrinsic information of the signal, and 10~(-3) orders of bit error rate performance is achieved with 7 dB of signal-to-noise ratio to QPSK modulated PCMA signals, accompanied by the declining complexity of the arithmetic square root level compared with the PSP algorithm.
Aiming at the high complexity in blind separation of PCMA mixed signals with non-cooperative reception,the separation algorithm based on feedforward neural network is proposed. By setting up a neural network separation platform and avoiding the traditional idea of maximum a posteriori probability, the blind separation algorithm with low complexity and high performance can be realized. Simulation results show that the neural network can greatly exploit the intrinsic information of the signal, and 10~(-3) orders of bit error rate performance is achieved with 7 dB of signal-to-noise ratio to QPSK modulated PCMA signals, accompanied by the declining complexity of the arithmetic square root level compared with the PSP algorithm.
引文
[1] Dankberg M. Paired carrier multiple access(PCMA)forsatellite communication[A]. Pacific TelecommunicationsConference[C]. Honolulu,Haw aii,1998. 787-791.
[2] Li F,Jafarkhani H. Interference cancellation and detectionfor more than tw o users[J]. International Journal of Engi-neering Research&Applications,2014,4(1):901-910.
[3]WuC L,Liu Z,Wang X,et al. Single-channel blind sourceseparation of co-frequency overlapped GM SK signals underconstant-modulus constraints[J]. IEEE Commun Lett,2016,20(3):486-489.
[4]付迪,高勇.非对称PCMA卫星信号的截获方法[J].现代电子技术,2007,7(7):28-30.Fu D,Gao Y. Signal interception method for asymmetricPCM A satellite communication system[J]. M odern Elec-tronic Technique,2007,7(7):28-30.(in Chinese)
[5]Ruder MA,Lehmann A,Schober M R,et al. Single antennainterference cancellation for GSM/VAM OS/EDGE usingLp-norm detection and decoding[J]. IEEE Transactionson Wireless Communications,2015,14(5):2413-2425.
[6]万坚,许华,朱中梁.基于独立分量分析的成对载波多址分离新算法[J].通信学报,2006,27(8):41-44.Wan J,Xu H,Zhu Z L. Separation algorithm of paired car-rier multiple access based on independent component analy-sis[J]. Journal on Communications,2006,27(8):41-44.
[7]Liu K,Li H,Dai X,et al. Single channel blind separation ofcofrequency M PSK signal[A]. Communication Internetand Information Technology[C]. USA,2006. 42-46.
[8]Tu S L,Chen S H,Zheng H,et al. Particle filtering basedsingle-channel blind separation of co-frequency M PSK sig-nals[A]. Proceedings of 2007 International Symposium onIntelligent Signal Processing and Communication Systems[C]. Xiamen,China,2007. 582-585.
[9]Tu S L,Chen S H,Zheng H,et al. On the performance ofsingle-channel blind separation of tw o co-frequency M PSKsignals[C]. 2007 IEEE Region 10 Conference TENCON,Penang,M alaysia,2007. 1-4.
[10] Tu S L,Zheng H,Gu N. Single-channel blind separationof tw o QPSK signals using per-survivor processing[C].IEEE Asia Pacific Conference on Circuits and Systems,M acao,China,2008. 473-476.
[11]杨勇,张冬玲,彭华,等.利用Gibbs采样的同频混合信号单通道盲分离[J].通信学报,2015,36(4):156-162.Yang Y,Zhang D L,Peng H,et al. Single-channel blindseparation of co-frequency modulated signals based onGibbs sampler[J]. Journal on Communications,2015,36(4):156-162.(in Chinese)
[12] Erkelens J,Hendriks R,Heusdens R,et al. Minimummean-square error estimation of discrete Fourier coeffi-cients w ith generalized gamma priors[J]. IEEE TransAudio,Speech,Lang Process,2007,15(6):1741-1752.
[13] Hendriks R,Heusdens R,Jensen J. MMSE based noisePSD tracking w ith low complexity[A]. ICASSP[C].Dallas,Texas,USA,2010. 4266-4269.
[14]Chen J,Wang Y,Wang D. A feature study for classifica-tion-based speech separation at very low signal-to-noiseratio[A]. ICASSP[C]. Florence,Italy,2014. 39-43.
[15]廖灿辉,周世东,朱中梁.基于最大似然的同频混合信号联合定时估计算法[J].系统工程与电子技术,2010,32(6):1121-1124.Liao C H,Zhou S D,Zhu Z L. M L-based joint timing es-timation algorithm for co-frequency signals[J]. SystemsEngineering and Electronics,2010,32(6):1121-1124.(in Chinese)
[16]芮国胜,徐彬,张嵩.单通道混合信号的幅度估计算法[J].通信学报,2011,32(12):82-87.Rui G S,Xu B,Zhang S. Amplitude estimation algorithmfor single channel mixing signals[J]. Journal on Commu-nications,2011,32(12):82-87.(in Chinese)
[17]万坚,涂世龙,廖灿辉等.通信混合信号盲分离理论与技术[M].北京:国防工业出版社,2012.Wan J,Tu S L,Liao C H,et al. Theory and Technologyon Blind Source Separation of Communication Signals[M]. Beijing:National Defense Industry Press,2012.(inChinese)
[18]杨勇,张冬玲,彭华,等.单通道高阶调制线性混合信号的频偏估计[J].电子学报,2015,43(1):160-165.Yang Y,Zhang D L,Peng H,et al. Frequency offset esti-mation in single-channel linear mixture of high-ordermodulated signals[J]. Acta Electronica Sinica,2015,43(1):160-165.(in Chinese)
[19]魏雪娟,等. PCMA系统中干扰信号的频偏与时延联合估计[J].通信学报,2017,38(2):173-182.Wei X J,et al. Joint estimation of interference frequencyoffset and time delay in PCM A system[J]. Journal onCommunications,2017,38(2):173-182.(in Chinese)
[20]Wang Yuxuan,Narayanan Arun,Wang Deliang. On train-ing targets for supervised speech separation[J]. IEEE/ACM Trans Audio Speech Lang Process,2014,22(12):1849-1858.
[2] Li F,Jafarkhani H. Interference cancellation and detectionfor more than tw o users[J]. International Journal of Engi-neering Research&Applications,2014,4(1):901-910.
[3]WuC L,Liu Z,Wang X,et al. Single-channel blind sourceseparation of co-frequency overlapped GM SK signals underconstant-modulus constraints[J]. IEEE Commun Lett,2016,20(3):486-489.
[4]付迪,高勇.非对称PCMA卫星信号的截获方法[J].现代电子技术,2007,7(7):28-30.Fu D,Gao Y. Signal interception method for asymmetricPCM A satellite communication system[J]. M odern Elec-tronic Technique,2007,7(7):28-30.(in Chinese)
[5]Ruder MA,Lehmann A,Schober M R,et al. Single antennainterference cancellation for GSM/VAM OS/EDGE usingLp-norm detection and decoding[J]. IEEE Transactionson Wireless Communications,2015,14(5):2413-2425.
[6]万坚,许华,朱中梁.基于独立分量分析的成对载波多址分离新算法[J].通信学报,2006,27(8):41-44.Wan J,Xu H,Zhu Z L. Separation algorithm of paired car-rier multiple access based on independent component analy-sis[J]. Journal on Communications,2006,27(8):41-44.
[7]Liu K,Li H,Dai X,et al. Single channel blind separation ofcofrequency M PSK signal[A]. Communication Internetand Information Technology[C]. USA,2006. 42-46.
[8]Tu S L,Chen S H,Zheng H,et al. Particle filtering basedsingle-channel blind separation of co-frequency M PSK sig-nals[A]. Proceedings of 2007 International Symposium onIntelligent Signal Processing and Communication Systems[C]. Xiamen,China,2007. 582-585.
[9]Tu S L,Chen S H,Zheng H,et al. On the performance ofsingle-channel blind separation of tw o co-frequency M PSKsignals[C]. 2007 IEEE Region 10 Conference TENCON,Penang,M alaysia,2007. 1-4.
[10] Tu S L,Zheng H,Gu N. Single-channel blind separationof tw o QPSK signals using per-survivor processing[C].IEEE Asia Pacific Conference on Circuits and Systems,M acao,China,2008. 473-476.
[11]杨勇,张冬玲,彭华,等.利用Gibbs采样的同频混合信号单通道盲分离[J].通信学报,2015,36(4):156-162.Yang Y,Zhang D L,Peng H,et al. Single-channel blindseparation of co-frequency modulated signals based onGibbs sampler[J]. Journal on Communications,2015,36(4):156-162.(in Chinese)
[12] Erkelens J,Hendriks R,Heusdens R,et al. Minimummean-square error estimation of discrete Fourier coeffi-cients w ith generalized gamma priors[J]. IEEE TransAudio,Speech,Lang Process,2007,15(6):1741-1752.
[13] Hendriks R,Heusdens R,Jensen J. MMSE based noisePSD tracking w ith low complexity[A]. ICASSP[C].Dallas,Texas,USA,2010. 4266-4269.
[14]Chen J,Wang Y,Wang D. A feature study for classifica-tion-based speech separation at very low signal-to-noiseratio[A]. ICASSP[C]. Florence,Italy,2014. 39-43.
[15]廖灿辉,周世东,朱中梁.基于最大似然的同频混合信号联合定时估计算法[J].系统工程与电子技术,2010,32(6):1121-1124.Liao C H,Zhou S D,Zhu Z L. M L-based joint timing es-timation algorithm for co-frequency signals[J]. SystemsEngineering and Electronics,2010,32(6):1121-1124.(in Chinese)
[16]芮国胜,徐彬,张嵩.单通道混合信号的幅度估计算法[J].通信学报,2011,32(12):82-87.Rui G S,Xu B,Zhang S. Amplitude estimation algorithmfor single channel mixing signals[J]. Journal on Commu-nications,2011,32(12):82-87.(in Chinese)
[17]万坚,涂世龙,廖灿辉等.通信混合信号盲分离理论与技术[M].北京:国防工业出版社,2012.Wan J,Tu S L,Liao C H,et al. Theory and Technologyon Blind Source Separation of Communication Signals[M]. Beijing:National Defense Industry Press,2012.(inChinese)
[18]杨勇,张冬玲,彭华,等.单通道高阶调制线性混合信号的频偏估计[J].电子学报,2015,43(1):160-165.Yang Y,Zhang D L,Peng H,et al. Frequency offset esti-mation in single-channel linear mixture of high-ordermodulated signals[J]. Acta Electronica Sinica,2015,43(1):160-165.(in Chinese)
[19]魏雪娟,等. PCMA系统中干扰信号的频偏与时延联合估计[J].通信学报,2017,38(2):173-182.Wei X J,et al. Joint estimation of interference frequencyoffset and time delay in PCM A system[J]. Journal onCommunications,2017,38(2):173-182.(in Chinese)
[20]Wang Yuxuan,Narayanan Arun,Wang Deliang. On train-ing targets for supervised speech separation[J]. IEEE/ACM Trans Audio Speech Lang Process,2014,22(12):1849-1858.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |