基于压缩感知的蒸发波导时空态势获取方法研究

项目名称： 基于压缩感知的蒸发波导时空态势获取方法研究

项目编号： No.41476089

项目类型： 面上项目

立项/批准年度： 2015

项目学科： 天文学、地球科学

项目作者： 康健

作者单位： 中国人民解放军海军航空大学

项目金额： 85万元

中文摘要： 海上蒸发波导是一把双刃剑，急需高效、准确地获取其时空态势以趋利避害，而仅靠增大浮标布设密度来提高态势感知时空分辨率的方式，不但费效比过高，且无法满足日益增长的需求。压缩感知理论为从少量低速的观测数据中获知长时间、大范围蒸发波导态势提供了理论基础。本项目围绕其中存在的三维高效观测、盲稀疏度条件下和含噪环境下精确重构等三大问题展开研究。主要内容包括：⑴提出时空相关条件下蒸发波导态势的三维等效观测方法，为充分利用时空相关性提高采样资源的使用效率提供一种新的途径；⑵研究蒸发波导态势的稀疏度约束及估计方法，弥补常规范数难以衡量波导态势稀疏度的不足，有望解决重构算法要求稀疏度确知的难题；⑶提出蒸发波导态势的去噪重构方法，解决噪声削弱信号稀疏度后造成的重构性能下降的问题。 本项目摒弃传统的先高速采样，后大量丢弃的模式，有望实现蒸发波导态势的高效、准确感知，有助于丰富三维含噪近似稀疏信号的压缩感知

中文关键词： 蒸发波导；压缩感知；近似稀疏；重构优化

英文摘要： Evaporation duct is a double-edged sword, so its time-space situation urgently needs to be sensed efficiently and accurately for profit and avoiding loss. But, it not only is high cost-effective but also can't meet the progressively requirements by just increasing the density of buoys. Compressed sensing provides the theoretical basis for the time-space situation awareness of evaporation duct for a long time and in a large area, which is recovered from a small amount of low speed measurements. There are still some problems like 3-D high efficient measuring and exact reconstruction of blind sparsity and noisy signal. The research centers on those problems mainly includes (1) proposing a 3-D equivalent measuring method for situation awareness of evaporation duct in time-space correlation condition, and providing a new way of enhancing the sampling efficiency; (2) discussing the sparsity estimation criterion and its constraint condition and covering the shortage that the traditional norm is not able to scale the situation sparsity, and appearing to solve the problem that the recovery algorithms require deterministic sparsity; (3) presenting de-noising reconstruction algorithm for time-space situation of evaporation duct to improve on the decreased performance in recovery caused by noise. The traditional mode of high speed sampling first and mass discarding then is abandoned, and it's hopeful that the time-space situation of evaporation duct is sensed efficiently and precisely. Meanwhile, the compressed sensing theory about 3-D approximately sparse noisy signal is enriched.

英文关键词： Evaporation duct;Compressive sensing;Approximately sparse;Reconstruction optimization