声信标信号的有效检测识别方法在找寻失事黑匣子的过程中起到关键作用。本文基于卷积神经网络（CNN）的检测识别方法，把已知声信标信号作为卷积神经网络的训练样本，提取梅尔频率倒谱系数（MFCC）特征后输入卷积神经网络进行训练，得到相应的训练标签。把待测的声信标信号输入经过训练的卷积神经网络进行测试，得到相应的识别结果。试验结果表明，基于卷积神经网络的方法可用于声信标信号的检测识别，并且有较好的识别率。
The effective detection and identification method of the acoustic beacon signal plays a key role in the search for the wrecked black box. Based on the detection and identification method of Convolutional Neural Network (CNN), this paper takes the known acoustic beacon signal as the training sample of the convolutional neural network, and then inputs the convolution neural network for training after extracting the characteristics of Mel Frequency Cepstrum Coefficient (MFCC) to obtain the corresponding training label. The acoustic beacon signal to be tested is fed into the trained convolutional neural network for testing, and the corresponding recognition results are obtained. The experimental results show that the method based on convolutional neural network can be used for the detection and recognition of the signal of the acoustic beacon, and has a good recognition rate.
2021,43(1): 150-153 收稿日期：2020-02-23
DOI：10.3404/j.issn.1672-7649.2021.01.028
分类号：U666.7
作者简介：张惠臣(1991-),男,硕士研究生,研究方向为水声信号与信息处理
参考文献：
[1] 邵云生. 水下信标定位接收技术综述[J]. 声学技术, 2015, 34(5): 42-47
SHAO Yun-sheng. Review of the underwater beacon’s acoustic positioning and receiving technology[J]. Technical Acoustics, 2015, 34(5): 42-47
[2] 王祯辉. 一种声信标搜寻系统的设计与实现[D]. 哈尔滨: 哈尔滨工程大学, 2012.
WANG Zhenhui. Design and implementation of an underwater acoustic locator system[D]. Harbin: Harbin Engineering University, 2012.
[3] 程星. 低信噪比水声信标信号处理方法研究及试验[D]. 大连: 大连海事大学, 2018.
CHENG Xing. Research and experiment on signal processing method of underwater acoustic beacon with low signal to noise ratio[D]. Dalian: Dalian Maritime University, 2018.
[4] 李笑媛, 程晶, 那健. 一种基于DSP的声信标信号提取方法[J]. 舰船科学技术, 2014, 36(11): 125-127
LI Xiao-yuan, CHENG Jing, NA Jian. Appoach to realizing weak hydroacoustic beacoing signal based on DSP[J]. SHIP SCIENCE AND TECHNOLOGY, 2014, 36(11): 125-127
[5] 李新欣. 船舶及鲸类声信号特征提取和分类识别研究[D]. 哈尔滨. 哈尔滨工程大学, 2012.
LI Xin-xin. Research on feature extraction and classification of ship noise and whale sound[D]. Harbin: Harbin Engineering University, 2012.
[6] 俞栋, 邓力. 解析深度学习: 语音识别实践[M]. 北京. 电子工业出版社, 2016.
[7] 程锦盛, 杜选民, 周胜增, 等. 基于目标MFCC特征的监督学习方法在被动声呐目标识别中的应用研究[J]. 舰船科学技术, 2018, 40(9): 116-121
CHENG Jin-sheng, DU Xuan-min, ZHOU-Sheng-zeng, et al. Application of the MFCC feature based supervised learning method in passive sonar target rec-ognition[J]. Ship Science and Technology, 2018, 40(9): 116-121
[8] LECUN Y, BENGIO Y. Convolutional networks for images, speech, and time series[C]. The handbook of brain theory and neural networks, 1995.
[9] Ludwig HOUÉGNIGAN, Pooyan SAFARI, Climent NADEU. Machine and deep learning approaches tolocalization and range estimation of underwater acoustic sources[J]. IEEE/OES Acoustics in Underwater Geosciences S-Ymposiumcprm, Urca, Rio de Janeiro, Brazil | July 25-27, 2017
