摘要
随着海陆空一体覆盖的6G通信技术不断发展,水下通信技术作为其关键组成部分起着至关重要的作用,水声信号调制方式识别技术保障了通信系统的稳定性。针对基于深度学习算法的水声信号调制方式识别技术准确率低、复杂度高的问题,设计了一种高效、准确的基于深度融合神经网络的水声信号调制方式识别算法。试验结果表明,基于实测水声数据集,该算法验证集准确率高达98.21%,水声信号平均识别时间仅为7.164 ms,与常规深度学习算法相比,深度融合神经网络模型既保证了高识别精度又大幅降低了识别时间。
With the development of 6G communication technology with the integrated coverage of sea, land and air, underwater communication technology plays an important role as its key component. Underwater acoustic signal modulation recognition technology ensures the stability of the communication system. Aiming at the problems of low accuracy and high complexity of underwater acoustic signal modulation recognition technology based on deep learning algorithm, an efficient and accurate algorithm is designed for recognizing underwater acoustic signal modulation mode based on deep fusion neural network. The experimental results show that based on the measured underwater acoustic data set, the accuracy of the verification set is as high as 98.21%, and the average recognition time of underwater acoustic signal is only 7.164 milliseconds. Compared with the conventional deep learning algorithm, the deep fusion neural network model not only ensures high recognition accuracy, but also greatly reduces the recognition time.
作者
张威龙
王景景
ZHANG Weilong;WANG Jingjing(School of Information Science and Technology,QUST,Qingdao 266000,China)
出处
《移动通信》
2022年第6期111-116,共6页
Mobile Communications
基金
山东省自然科学基金重大基础研究项目(ZR2021ZD12)。
关键词
6G
水下通信
调制方式识别
神经网络
深度学习
6G
underwater communication
modulation mode recognition
neural network
deep learning
