基于通信载荷声纹特征的身份认证方法

陈露; 赵德鑫; 王俊; 高虹; 陈迎亮

doi:10.11993/j.issn.2096-3920.2023-0027

基于通信载荷声纹特征的身份认证方法

doi: 10.11993/j.issn.2096-3920.2023-0027

陈露^1,,
赵德鑫^1,,
王俊^1,,
高虹^1,,
陈迎亮^2,

1.
国防科技创新研究院前沿交叉技术研究中心, 北京, 100071
2.
中国船舶集团有限公司第 705 研究所, 陕西西安, 710077

详细信息

作者简介:
陈露：陈　露(2000-), 女, 在读硕士, 主要研究方向为水下智能探测技术

中图分类号: TJ630.34; U674
计量
- 文章访问数: 178
- HTML全文浏览量: 108
- PDF下载量: 57
- 被引次数: 0
出版历程
- 收稿日期: 2023-03-16
- 修回日期: 2023-07-11
- 录用日期: 2023-08-17
- 网络出版日期: 2023-11-21

Identity Authentication Method Based on Voiceprint Features of Communication Payloads

CHEN Lu^1
,,
ZHAO Dexin^1
,,
WANG Jun^1
,,
GAO Hong^1
,,
CHEN Yingliang^2
,

1.
Advanced Interdisciplinary Technology Research Center, National Innovation Institute of Defense Technology, Beijing 100071, China
2.
The 705 Research Institute, China State Shipbuilding Corporation Limited, Xi’an 710077, China

摘要

摘要: 水声通信网络安全是水下通信进行信息共享和协同作业的重要保障, 现有技术主要研究认证协议和数据加密, 注重提高网络的安全性, 忽略了网络的效率和能耗问题。为避免上述方法带来的网络拥堵, 受移动智能设备等领域研究启发, 提出将声纹认证融入水下通信网络的身份认证系统, 设计了一种基于通信载荷声纹特征的识别方法。利用注意力机制, 融合非线性倒谱和相位谱特征, 以降低复杂海洋环境噪声的影响, 并通过AlexNet网络进行目标识别。为验证该方法的有效性, 通过采集多种水声通信信号, 检验文中声纹特征识别的差异性和有效性, 论证了文中方法的可行性和可靠性。文中研究可为解决水声通信网络的身份认证问题提供新思路, 对增强水声通信网络的安全、实现高质量的信息共享和高效率的协同控制提供一定参考。
- 水声通信 /
- 声纹特征 /
- 身份认证 /
- 目标识别
Abstract: The security of underwater acoustic communication networks is an important guarantee for information sharing and cooperative operation of underwater communication. Existing technologies mainly study authentication protocol and data encryption, focusing on improving the security of the network but ignoring the efficiency and energy consumption of the network. To avoid network congestion caused by the above methods, this paper, inspired by research in mobile smart devices and other fields, proposed to integrate voiceprint authentication into the identity authentication system of underwater communication networks, and it designed a recognition method based on voiceprint features of communication payloads. This method used the attention mechanism and merged nonlinear cepstrum features and phase spectrum features to reduce the influence of complex marine environment noise. In addition, it identified the target through the AlexNet network. To verify the effectiveness of this method, this paper collected underwater acoustic communication signals, verified the difference and effectiveness of the proposed voiceprint feature recognition, and demonstrated the feasibility and reliability of the proposed method. The research in the paper provides a new idea for solving the identity authentication of underwater acoustic communication networks, which serves as a reference for enhancing security and realizing high-quality information sharing and high-efficiency cooperative control of underwater acoustic communication networks.
- underwater acoustic communication /
- voiceprint features /
- identity authentication /
- target recognition

HTML全文

图 1 水声通信网络身份认证系统结构

Figure 1. Architecture of underwater acoustic communication network authentication system

下载: 全尺寸图片幻灯片

图 2 通信系统结构

Figure 2. Structure of communication system

下载: 全尺寸图片幻灯片

图 3 换能器内部结构

Figure 3. Internal structure of the transducer

下载: 全尺寸图片幻灯片

图 4 基于AUV声纹特征的识别方法流程图

Figure 4. Flow chart of the recognition method based on AUV voiceprint features

下载: 全尺寸图片幻灯片

图 5 特征提取流程图

Figure 5. Flow chart of feature extraction

下载: 全尺寸图片幻灯片

图 6 神经网络结构图

Figure 6. Structure of neural network

下载: 全尺寸图片幻灯片

图 7 特征融合

Figure 7. Feature fusion

下载: 全尺寸图片幻灯片

图 8 实验布局图

Figure 8. Experimental layout

下载: 全尺寸图片幻灯片

图 9 混淆矩阵

Figure 9. Confusion matrix

下载: 全尺寸图片幻灯片

表 1 换能器失真影响

Table 1. Distortion effects of transducer

影响因素	来源	表现
非线性输出	非线性元件响应	幅度、相位差异
频响变化	换能材料灵敏度	频率特性变化
信道噪声	机械振动、耦合	清晰度
时延	信号传输过程	时域特征

下载: 导出CSV

表 2 信号采集表

Table 2. Signal acquisition

型号	标识	信号内容	数据量
M5M025N	01	1~9 种	2 667
M5M025N	02	1~9 种	2 165
0U2030DECK	03	A~E 种	353
0U2030DECK	04	A~E 种	279

下载: 导出CSV

表 3 数据集划分

Table 3. Division of data set

数据集	型号	信号内容
训练集	M5M025N	1~6种
测试集	M5M025N	7~9种
训练集	0U2030DECK	A~C种
测试集	0U2030DECK	D/E

下载: 导出CSV

表 4 不同特征下识别结果对比

Table 4. Comparison of recognition results under different features

特征	识别率/%
非线性倒谱+相位谱	98.15
非线性倒谱	95.19
线性倒谱+相位谱	94.66
线性倒谱	92.68

下载: 导出CSV

表 5 不同条件变量下识别结果对比

Table 5. Comparison of recognition results under different condition variables

训练集	测试集		数据量	识别率/%
训练集	变量	信号内容	数据量	识别率/%
1~6种信号	深度	1~6种	1 147	95.64
	深度	7~9种	626	88.82
	距离	5~6种	211	86.71
	距离	7~9种	725	77.86
	运动	5/6	271	83.46
	运动	7~9种	859	76.59

下载: 导出CSV

参考文献(17)

[1]	楚立鹏, 鄢宏华, 范强, 等. 国外水下无人潜航器及其通信技术发展综述[J]. 中国电子科学研究院学报, 2022, 17(2): 112-118. Chu Lipeng, Yan Honghua, Fan Qiang, et al. A review on the development of foreign unmanned underwater vehicles and their communication technology[J]. Journal of the Chinese Academy of Electronic Science, 2022, 17(2): 112-118.
[2]	韦韬, 朱遴, 梁世龙. 水下无人系统集群感知与协同技术发展[J]. 指挥控制与仿真, 2022, 44(5): 6-13. Wei Tao, Zhu Lin, Liang Shilong. Development of cluster sensing and cooperative technology for underwater unmanned systems[J]. Command and Control and Simulation, 2022, 44(5): 6-13.
[3]	宋保维, 潘光, 张立川, 等. 自主水下航行器发展趋势及关键技术[J]. 中国舰船研究, 2022, 17(5): 27-44. Song Baowei, Pan Guang, Zhang Lichuan, et al. Development trend and key technologies of autonomous underwater vehicles[J]. China Ship Research, 2022, 17(5): 27-44.
[4]	羊秋玲, 申家琪, 练凯伟, 等. 一种水下无线传感器网络数据传输安全防御框架设计: CN112672347A[P]. 2021-04-16.
[5]	陈惠芳, 余修俊, 谢磊. 水声通信中固定传感器节点与移动汇聚节点安全通信方法: CN110380848A[P]. 2019-07-09.
[6]	黄亮平. 水声通信网的认证和入侵检测技术研究[D]. 南京: 东南大学, 2019.
[7]	Nguyen N T, Zheng G, Zhu H, et al. Device fingerprinting to enhance wireless security using nonparametric Bayesian method[C]//2011 Proceedings IEEE INFOCOM. Shanghai, China: IEEE, 2011.
[8]	冀晓宇, 徐文渊, 程雨诗. 一种基于CPU模块电磁辐射的设备指纹提取和认证方法: CN108664785A[P]. 2018-04-04.
[9]	Das A, Borisov N, Caesar M C. Do you hear what I hear? Fingerprinting smart devices through embedded acoustic components[C]//Proceedings of the 2014 ACM SIGSAC Conference on Computer and Communications Security. [S. l.]: ACM, 2014: 441-452.
[10]	Bojinov H, Michalevsky Y, Nakibly G, et al. Mobile device identification via sensor fingerprinting[EB/OL]. (2014-04-06)[2023-05-06]. https://doi.org/10.48550/arXiv.1408.1416
[11]	莫喜平. 我国水声换能器技术研究进展与发展机遇[J]. 中国科学院院刊, 2019, 34(3): 272-282. doi: 10.16418/j.issn.1000-3045.2019.03.004 Mo Xiping. Research progress and development opportunities of hydroacoustic transducer technology in China[J]. Proceedings of the Chinese Academy of Sciences, 2019, 34(3): 272-282. doi: 10.16418/j.issn.1000-3045.2019.03.004
[12]	Liu M, Wang L, Dang J, et al. Replay attack detection using magnitude and phase information with attention-based adaptive filters[C]//2019 IEEE International Conference on Acoustics, Speech and Signal Processing(ICASSP). Brighton, UK: IEEE, 2019.
[13]	董胡. 基于窗函数与MATLAB的数字FIR滤波器设计[J]. 微型电脑应用, 2016, 32(3): 30-33. Dong Hu. Digital FIR filter design based on window function and MATLAB[J]. Microcomputer Applications, 2016, 32(3): 30-33.
[14]	徐小平, 余香佳, 刘广钧, 等. 利用改进AlexNe卷积神经网络识别石墨[J]. 计算机系统应用, 2022, 31(2): 376-383. Xu Xiaoping, Yu Xiangjia, Liu Guangjun, et al. Graphite recognition using improved AlexNe convolutional neural network[J]. Computer Systems Applications, 2022, 31(2): 376-383.
[15]	Dai Y, Gieseke F, Oehmcke S, et al. Attentional feature fusion[C]//2021 IEEE Winter Conference on Applications of Computer Vision(WACV). Waikoloa, HI, USA: IEEE, 2021.
[16]	Sun C, Xu Y, Wu Z, et al. ReAFFPN: Rotation-equivariant attention feature fusion pyramid networks for aerial object detection[C]//2022 IEEE International Geoscience and Remote Sensing Symposium. Kuala Lumpur, Malaysia: IEEE, 2022.
[17]	Wang Z, Shao W, Chen Y, et al. A cross-scale iterative attentional adversarial fusion network for infrared and visible images[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2023, 33(8): 3677-3688.