摘要
相比于陆地无线传感器网络,水下传感器网络需要考虑能耗问题,如何使水下传感器更加节能成为研究的重要课题;同时,海水的温盐深等特性,决定了海水的声速深度剖面是弯曲的、声速是水平分层的,这是造成水下待定位节点与参考节点位置确定不准确的主要原因。针对水下场景中的能耗问题,构建了水下传感器节点协同定位的定位模型;针对声速的不稳定性,提出了一种基于卷积神经网络的水下声速模型,分析并设计了神经网络的网络结构,同时针对网络的缺陷,利用遗传算法对卷积神经网络参数进行模型优化,用以提高声速模型的精确性。利用南海海洋断面科学考察温盐深观测数据集,完成实验仿真对比。结果表明,协同定位模型可有效解决传感器网络的能耗问题,同时修正了海底声速,降低了定位过程中由于声速不稳定与时变性带来的误差影响。该模型进一步提高了水下传感器网络定位系统的精确性与鲁棒性。
Compared with terrestrial wireless sensor networks,underwater sensor networks need to consider energy consumption.How to make underwater sensors more energy-saving has become an important research topic.At the same time,the characteristics of temperature and salt depth of seawater determine that the sound velocity depth profile of seawater is curved and the sound velocity is horizontally layered,which is the main reason for the inaccurate determination of the localization of underwater nodes to be located and reference nodes.Aiming at the problem of energy consumption in underwater scene,a localization model of cooperative localization of underwater sensor nodes is constructed;Aiming at the instability of sound velocity,an underwater model of sound velocity based on convolutional neural network is proposed,and the network structure of neural network is analyzed and designed.At the same time,aiming at the defects of the network,the parameters of convolutional neural network are optimized by genetic algorithm to improve the accuracy of sound velocity model.The experimental simulation comparison is completed by using the temperature,salinity and depth observation data set of the scientific investigation of the marine section of the South China Sea.The results show that the cooperative positioning model can effectively solve the problem of energy consumption of sensor networks,correct the seabed sound velocity,reduce the error impact caused by the instability and time variability of sound velocity in the positioning process and further improve the accuracy and robustness of underwater sensor network positioning system.The proposed model further improves the accuracy and robustness of underwater sensor network positioning system.
作者
崔学荣
刘金峰
李娟
李世宝
刘建航
CUI Xuerong;LIU Jinfeng;LI Juan;LI Shibao;LIU Jianhang(China University of Petroleum(East China),Qingdao,Shandong 266000,China)
出处
《导航定位学报》
CSCD
2022年第5期46-53,共8页
Journal of Navigation and Positioning
基金
国家自然科学基金项目(61902431,52171341,91938204,61972417)
山东省自然科学基金项目(ZR2020MF005)。
关键词
水下传感器网络
协同定位
声速模型
卷积神经网络
遗传算法
underwater sensor network
collaborative positioning
model of sound velocity
convolutional neural network
geneti c algorithm
