脉冲噪声环境下的水声通信信号调制识别方法  

Modulation Recognition Method of Underwater Acoustic Communication Signals in Impulsive Noise Environment

作　　者：王彬 王海旺 李勇斌 Wang Bin;Wang Haiwang;Li Yongbin(School of Information Systems Engineering,PLA Strategic Support Force Information Engineering University,Zhengzhou,Henan 450001.,China)

机构地区：[1]中国人民解放军战略支援部队信息工程大学信息系统工程学院,河南郑州450001

出　　处：《信号处理》2020年第12期2107-2115,共9页Journal of Signal Processing

基　　金：国家自然科学基金(61602511)。

摘　　要：为了提高浅海脉冲噪声环境下水声通信信号调制识别的性能和实用性,提出了基于降噪自编码器和卷积神经网络的调制识别方法。算法构造了联合降噪自编码器和卷积神经网络的框架,利用降噪自编码器对含噪声信号进行降噪处理,利用卷积神经网络对降噪信号的功率谱图进行调制方式的分类识别。为了解决目标水域水声通信信号训练样本不足的问题,采用迁移学习思想,利用典型声剖面构造水声通信信号训练数据集,采用两步迁移策略提升小样本条件下的水声信号调制识别能力。仿真实验和实测数据验证了本文方法的有效性。与现有算法相比,本文所提方法具有较高的识别率,并且提升了目标信道数据不足条件下的识别性能。To improve the performance and practicability of modulation recognition of underwater acoustic communication signals in impulse noise environment of shallow sea,a modulation recognition method based on denoising automatic-encoder(DAE)and convolutional neural network(CNN)is proposed.First,a joint construction of DAE and CNN is proposed,in which the DAE plays an important role to suppress impulsive noise,and the CNN is used to classify the power spectrum of signals after noise reduction.Meanwhile,in order to solve the problem of insufficient training samples of underwater acoustic communication signals in target waters,the idea of transfer learning is adopted.The transfer learning data set is constructed by the typical acoustic profile and the two-step transfer learning strategy is used to improve the modulation recognition ability of underwater acoustic signals with small samples.Simulation results and practical signal tests demonstrate the effectiveness of the proposed method.Compared with the existing methods,the proposed method improves the accuracy rate of modulation recognition in impulse noise environment even with insufficient target data.

关 键 词：调制识别 脉冲噪声 降噪自编码器 卷积神经网络 数据迁移 

分 类 号：TN911.7[电子电信—通信与信息系统]