基于多水下滑翔机的海洋水声信道特征参数测绘研究

For obtaining accurate parameters that characterize the underwater acoustical communication channels in three dimensional ocean environments to support the design, implementation, and efficient operation of underwater acoustic sensor networks, this project proposes a novel approach to direct measurement of oceanic acoustic communication channels, which will be more efficient and economic than other traditional approaches. In this approach, underwater gliders which are equipped with acoustic sensors are employed to move in an area to be measured in two or three dimensional space and measure and record acoustic signals transmitted by a fixed acoustic source deployed in the ocean, and by utilizing the acoustic signals measured along the gliders’ paths a three dimensional distribution map of the parameters characterizing acoustic communication channels between the deployed acoustic source and any other locations in the area will be constructed. The key issues relating to this approach are to be investigated in this project, which include the strategies of mapping the parameters with underwater gliders and methods of constructing the 3D map of the parameters with the data collected by the gliders, the accurate localization of the underwater gliders by fusing the measured acoustic signals, and the planning of the gliders’ paths to collect the acoustic signals. The research results including developed strategies, methods and algorithms will be demonstrated and validated by computer simulation and part in field tests. The proposed approach to acoustic channel measurement and the systematic research results of the key issues will provide a solid foundation for the future application of gliders in acoustic channel mapping and 3D mapping of other environmental parameters in ocean environments. And the research results of this project will also contribute to the research on adaptive and autonomous observation of oceans with mobile sensor networks based on multiple underwater gliders.

针对基于水声通信的海洋无线传感器网络设计、构建和应用等对高分辨率水声信道空间分布特性数据的需求，首次提出利用多水下滑翔机进行海洋水声信道的直接测量和水声信道特征参数的三维空间分布场构建。主要研究其涉及的：基于多水下滑翔机的海洋水声信道特征参数测绘策略和特征参数分布场构建方法、水下滑翔机测量水声信号路径的规划方法和算法、和融合水声信号测量信息的水下滑翔机精确定位策略和算法三方面关键研究内容，对涉及的关键基础问题进行探索和解决。通过项目的研究，以期获得一套支撑应用多水下滑翔机进行海洋水声信道特征参数三维空间分布测绘的基础策略、方法和算法。在对研究结果进行基于计算机仿真的有效性验证和性能评估外，还对部分研究结果进行海洋环境实验演示和验证。项目的研究结果，为基于水下滑翔机的海洋水声信道直接测量以及其它海洋环境参数三维空间分布场的测绘提供重要支撑。

本项目结合水下机器人数据驱动参数场构建理论框架，以无线海洋传感器网络中对全面认识和掌握海洋水声信道特征需求为背景，为提高基于多水下滑翔机的海洋水声信道特征参数场测绘精度，开展了一系列前沿问题的应用基础研究。主要研究内容包括特征参数分布场构建、优化采样、自噪声滤除、高精度导航定位以及对相关技术理论方法的仿真与实验验证研究。本项目已按照研究计划完成了全部研究内容，主要成果为：.（1）提出了加权克里金-压缩感知水声信道特征参数场重构方法，解决了沿轨迹连续采样影响稀疏变换矩阵与测量矩阵相干性的问题，通过基于海洋-声学耦合模式的仿真系统以及外场试验验证所提方法的重构性能明显优于单独克里金或压缩感知方法。.（2）结合压缩感知理论中有限等距性质和水下机器人的连续运动约束，提出一种基于遗传算法的测量矩阵优化方法，形成全局区域内的最优采样路径，仿真实验验证所提方法在重构性能指标上均高于高斯TSP路径和梳状路径。.（3）提出一种联合卷积滤波阈值方法的自噪声滤除方法，解决了声学水下滑翔机作为水声信号测量平台自身噪声对记录声学数据污染的问题，通过海试实验数据处理，验证了此算法为从受水下滑翔机噪声污染的水听器记录信号中获取干净的环境噪声数据提供了有效手段。.（4）提出了基于单信标声传播时间差的水下滑翔机导航定位算法，解决了水下滑翔机由于负载有限导致其水下导航定位精度低、进而影响依赖于样本空间位置的数据驱动空间场重构精度的问题，结合南海实验中获取的角度和深度测量以及海洋-声学耦合模式仿真系统中的海流和声学模拟，开展仿真实验验证水下滑翔机的导航定位精度明显提升。