基于压缩感知的稀疏水声阵列信号处理机理研究

项目名称： 基于压缩感知的稀疏水声阵列信号处理机理研究

项目编号： No.11304343

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 数理科学和化学

项目作者： 张扬帆

作者单位： 中国科学院声学研究所

项目金额： 25万元

中文摘要： 随着海洋探索的不断深入，对水下声学探测设备的性能提出更高要求，尤其是在分辨率和实时性方面。阵元的增加可以提高分辨率，然而随之带来的复杂性、功耗、数据传输量的增加又成为制约系统实时性能的瓶颈。理论研究指出，稀疏化处理技术可应用于从信号采集阶段开始到信号重构显示阶段，可以简化前端设备的复杂性，减少数据和运算的冗余度，合理分布各组成部件的负载。本项目拟从分析水声阵列信号的稀疏特性入手，比较稀疏分布阵和稀疏抽取阵的不同特性，引入最新的压缩感知理论，研究水下声阵的优化设计方法和信号处理方法。本项目将在充分的理论研究和大量实验数据分析的基础上，建立基于空、时、频稀疏化处理技术的阵列信号处理模型和理论基本框架，研究与之相应的快速成像算法，实现高性能、低功耗水下声学实时探测设备的技术突破，为水下手持式探测装备和水下机器人的研制提供技术支撑。

中文关键词： 稀疏阵；阵列优化；阵列信号处理；三维声成像；压缩感知

英文摘要： With more and more attention has been paid to underwater exploration, more higher performance of marine acoustic equipments are required, such as real-time and high-resolution. A large-scale elements acoustic array can get higher resolution but company with more complexity structure, power consumption and mass data transmission. Research shows that, the application of sparse process from signal collection to signal reconstruction can get simplified structure, lower data redundancy and appropriate load distribution on every component module. Starting with the analysis sparse structure of underwater signal, Comparing the different Characteristics of sparse array and thinned array, using the latest compressed sensing theory, this program concretely research the optimum design method and signal processing method of underwater acoustic array. Through sufficient theoretical discussion and numerous experimental data analysis, establish model and fundamental framework of sparse process technology in the space-time-frequency domain and fast imaging algorithm, seek for critical technical breakthrough in high-performance and low-power-consumption marine acoustic real-time equipments, provide the technical support of underwater hand-held devices and robots.

英文关键词： sparse array；array optimization；array signal processing；3D acoustic imaging；compressive sensing