复杂多径环境下的无人机集群通信波形识别

doi:10.12305/j.issn.1001-506X.2023.10.36

摘要/Abstract

摘要：

无人机(unmanned aerial vehicle, UAV)集群通信电磁环境复杂, 存在用户干扰、多径衰落与频移等现象, 传统加性高斯白噪声信道下的波形识别算法在此场景下性能大幅降低。针对此问题, 提出了一种复杂多径环境下的UAV集群通信波形识别算法。首先, 建立Alpha脉冲干扰下的UAV集群通信多径衰落信道模型; 然后, 针对集群用户间存在Alpha脉冲干扰的问题, 提取信号广义循环均值和广义循环谱特征, 建立复杂多径环境下的UAV集群通信波形特征矩阵; 最后, 建立稀疏自编码器深度神经网络UAV集群通信波形识别模型。仿真结果表明: 提出的算法在Alpha脉冲干扰、多径衰落与频移存在的复杂环境下具有较强的鲁棒性, 实现了7种UAV集群通信波形的识别, 且在信噪比为-10 dB时仍能保证80%以上的识别准确率。

关键词: 无人机集群, 调制识别, Alpha脉冲干扰, 稀疏自编码

Abstract:

The electromagnetic environment of unmanned aerial vehicle (UAV) swarm communication is complex, and phenomena such as user interference, multipath fading and frequency shift frequently occur. The performance of the waveform recognition algorithm under the traditional additive white Gaussion noise channel is greatly reduced in this scenario. To solve this problem above, a waveform recognition algorithm for UAV swarm communication in complex multipath environment is proposed. Firstly, a multipath fading channel model for UAV swarm communication under Alpha pulse interference is established. Secondly, aiming at the problem of Alpha pulse interference among swarm users, the generalized cyclic mean and generalized cyclic spectrum features of the signals are extracted, and the waveform feature matrix of UAV swarm communication in complex multipath environment is established. Finally, the sparse autoencoder deep neural network UAV swarm communication waveform recognition model is established. The simulation results show that the proposed algorithm has strong robustness in the complex environment of Alpha pulse interference, multipath fading and frequency shift, and realizes the recognition of seven kinds of UAV swarm communication waveforms. At the same time, the recognition accuracy of more than 80% can be guaranteed when the signal to noise ratio is -10 dB.

Key words: unmanned aerial vehicle (UAV) swarm, modulation recognition, Alpha pulse interference, sparse autoencoder

中图分类号:

TN911

翟茹萍, 张书衡, 平嘉蓉. 复杂多径环境下的无人机集群通信波形识别[J]. 系统工程与电子技术, 2023, 45(10): 3312-3320.

Ruping ZHAI, Shuheng ZHANG, Jiarong PING. Waveform recognition of unmanned aerial vehicle swarm communication in complex multipath environment[J]. Systems Engineering and Electronics, 2023, 45(10): 3312-3320.

图/表 19

图1

图2

表1

图3

图4

图5

图6

表2

表3

图7

图8

图9

图10

图11

图12

表4

图13

表5

图14

参考文献 21

1	柳征, 王明阳, 姜文利, 等. 一种新的贝叶斯调制分类算法[J]. 电子与信息学报, 2006, 7, 1233- 1237.
	LIU Z , WANG M Y , JIANG W L , et al. A novel Bayesian modulation classification algorithm[J]. Journal of Electronics and Information, 2006, 7, 1233- 1237.
2	李红光, 郭英, 眭萍, 等. 基于时频能量谱纹理特征的跳频调制方式识别[J]. 通信学报, 2019, 40 (10): 20- 29.
	LI H G , GUO Y , SUI P , et al. Identification of frequency hopping modulation mode based on time-frequency energy spectrum texture features[J]. Journal of Communications, 2019, 40 (10): 20- 29.
3	KUMAR Y , SHEORAN M , JAJOO G , et al. Automatic modulation classification based on constellation density using deep learning[J]. IEEE Communications Letters, 2020, 24 (6): 1275- 1278. doi: 10.1109/LCOMM.2020.2980840
4	XIE W W , HU S , YU C , et al. Deep learning in digital modulation recognition using high order cumulants[J]. IEEE Access, 2019, 63760- 63766.
5	郝云飞. 基于深度学习的无人机信号检测与识别技术研究[D]. 长沙: 国防科技大学, 2019: 42-56.
	HAO Y F. Research on detection and recognition technology for UAV signal based on deep learning[D]. Changsha: National University of Defense Technology, 2019: 42-56.
6	王自维, 姚志成, 王海洋, 等. 基于改进LeNet-5模型的无人机遥控信号调制方式识别算法[J]. 火箭军工程大学学报, 2021, 4, 30- 34.
	WANG Z W , YAO Z C , WANG H Y , et al. Recognition algorithm of UAV remote control signal modulation mode based on improved LeNet-5 model[J]. Journal of Rocket Force Engineering University, 2021, 4, 30- 34.
7	TU Y , LIN Y . Deep neural network compression technique towards efficient digital signal modulation recognition in edge device[J]. IEEE Access, 2019, 7, 58113- 58119. doi: 10.1109/ACCESS.2019.2913945
8	WANG D , ZHANG M , LI Z , et al. Modulation format recognition and OSNR estimation using CNN-based deep learning[J]. IEEE Photonics Technology Letters, 2017, 29 (19): 1667- 1670. doi: 10.1109/LPT.2017.2742553
9	ZHANG Z F , WANG C , GAN C Q , et al. Automatic modulation classification using convolutional neural network with features fusion of SPWVD and BJD[J]. IEEE Trans.on Signal and Information Processing Over Networks, 2019, 5 (3): 469- 478. doi: 10.1109/TSIPN.2019.2900201
10	张天骐, 范聪聪, 葛宛营, 等. 基于ICA和特征提取的MIMO信号调制识别算法[J]. 电子与信息学报, 2020, 42 (9): 2208- 2215.
	ZHANG T Q , FAN C C , GE W Y , et al. A modulation identification algorithm of MIMO signal based on ICA and feature extraction[J]. Journal of Electronics and Information, 2020, 42 (9): 2208- 2215.
11	PENG S L , JIANG H Y , WANG H X , et al. Modulation classification based on signal constellation diagrams and deep learning[J]. IEEE Trans.on Neural Networks and Learning Systems, 2018, 30 (3): 718- 727.
12	张思成, 林云, 涂涯, 等. 基于轻量级深度神经网络的电磁信号调制识别技术[J]. 通信学报, 2020, 41 (11): 12- 21.
	ZHANG S C , LIN Y , TU Y , et al. Electromagnetic signal modulation identification technology based on lightweight deep neural network[J]. Journal of Communications, 2020, 41 (11): 12- 21.
13	李晨, 杨俊安, 刘辉. 基于信息熵和GA-ELM的调制识别算法[J]. 系统工程与电子技术, 2020, 42 (1): 223- 229.
	LI C , YANG J A , LIU H . Modulation identification algorithm based on information entropy and GA-ELM[J]. Systems Engineering and Electronics, 2020, 42 (1): 223- 229.
14	史蕴豪, 许华, 郑万泽, 等. 基于集成学习与特征降维的小样本调制识别方法[J]. 系统工程与电子技术, 2021, 43 (4): 1099- 1109.
	SHI Y H , XU H , ZHENG W Z , et al. Few-sample modulation recognition method based on ensemble learning and feature dimensionality reduction[J]. Systems Engineering and Electro-nics, 2021, 43 (4): 1099- 1109.
15	冯径, 熊鑫立, 蒋磊. 软件通信适配器的调制模式识别算法[J]. 东南大学学报, 2017, 47 (3): 456- 460.
	FENG J , XIONG X L , JIANG L . Modulation pattern recognition algorithm for software communication adapter[J]. Journal of Southeast University, 2017, 47 (3): 456- 460.
16	ETSI TR 38.901. Study on channel model for frequencies from 0.5 to 100 GHz (V17.0.0)[S]. 3GPP, 2022.
17	JAKES W C . Microwave mobile communications[M]. New York: Wiley, 1974.
18	SAMORODNITSKY G , TAQQU M S . Stable non-Gaussian random processes: stochastic models with infinite variance[M]. London: Routledge, 2017.
19	AZZOUZ E , NANDI A K . Automatic modulation recognition of communication signals[M]. BerLin: Springer Science & Business Media, 2013.
20	孔豫京. 基于Alpha稳定分布理论的无人机侦察信道建模与信号调制识别技术研究[D]. 郑州: 解放军信息工程大学, 2015: 52-53.
	KONG Y J. Research on channel modeling and signal modulation recognition for UAV reconnaissance based on Alpha stable distribution theory[D]. Zhengzhou: PLA Information Engineering University, 2015: 52-53.
21	ALI A , YANG Y F . Automatic modulation classification using deep learning based on sparse autoencoders with nonnegativity constraints[J]. IEEE Signal Processing Letters, 2017, 24 (11): 1626- 1630.

调制类型	循环均值	f=0截面的循环谱
MASK	ε=f_c	ε=2f_c
BPSK	无离散峰值	ε=±(2f_c±R_b)
QPSK	无离散峰值	无离散峰值
MFSK	ε=f_c+f_m	ε=±2(f_c+f_m)
MSK	无离散峰值	ε=±(2f_c±R_b/2)

参数	无人机1	无人机2
飞行速度/(m/s)	50.0	50.5
最大多普勒频移/Hz	967	976
TDL-A时延扩展/ns	100	101
TDL-D时延扩展/ns	30	31

SAE网络模型	参数值
输入层	14
隐藏层	(10, 8)
输出层	6
激活函数	Linear, ReLU, Softmax
优化算法	L-BFGS
稀疏系数	0.1
稀疏权重因子	3
权重衰减系数	0.000 1

无人机	飞行速度/(m/s)	最大多普勒频移/Hz	时延扩展/ns
1	50.0	967	100
2	50.5	976	101
3	51.0	986	102
4	51.5	995	103

场景	参数	无人机1	无人机2
所有场景	时延扩展/ns	100	101
场景1	飞行速度/(m/s)	15	15.5
场景1	最大多普勒频移/Hz	290	300
场景2	飞行速度/(m/s)	25	25.5
场景2	最大多普勒频移/Hz	483	493
场景3	飞行速度/(m/s)	50	50.5
场景3	最大多普勒频移/Hz	967	976
场景4	飞行速度/(m/s)	70	70.5
场景4	最大多普勒频移/Hz	1 353	1 363