不同海域真实海面高海情小擦地角全极化海杂波特性及机理分析

The research on the characteristics and mechanism of sea clutter under complex ocean environment is of great significance to the real-time, accuracy and stability of radar target detection. How to recognize the characteristics accurately and deeply has been a hot and difficult issue. Based on the data sets of marine environment in different sea areas, this project establishes the electrically large real sea surface under high sea condition. Combined with the time-evolving multi-scale sea surface electromagnetic scattering theory and high performance parallel technology, the key technologies of multi-scale sea surface multiple scattering, high-order Bragg scattering, foam surface composite scattering and breaking wave Burst scattering are broken down. The law and mechanism of the effects of various physical field parameters such as ocean environment parameters, radar parameters on the characteristics of sea clutter are analyzed by comparing the measured data. Based on the measured and simulated large sea clutter data, the multi-layer self-coding deep learning network is used to explore the high-order, hidden layer and non-linear mapping relationship between multi-physical field parameters and sea clutter characteristics. A deep cognitive model of the relationship between multi-dimensional characteristics of sea clutter and physical field of radar and marine environment is established to realize deep recognition of multi-dimensional characteristics of sea clutter based on physical field parameters of sea clutter detection scene. The research results can provide the basis and technique support for the establishment of practical sea clutter model, clutter suppression and target detection.

复杂海洋环境下海杂波特性及机理研究对雷达目标探测的实时性、准确性及稳定性具有重要意义，如何精确、深入地认知其特性一直是研究热点和难点。本项目利用不同海域海洋环境数据集，建立高海情电大区域真实海面几何精细结构，结合时变多尺度海面电磁散射理论及高性能并行技术，突破多尺度海面多重散射、Bragg散射、泡沫体-面复合散射及破碎波Burst散射关键技术，结合实测数据对比分析不同海洋环境参数、雷达参数等多物理场参数对海杂波特性的影响规律和机理。基于实测及仿真海杂波大数据，利用多层自编码深度学习网络，探究多物理场参数与海杂波特性的高阶、隐层、非线性映射关系，建立海杂波多维特性与雷达、海洋环境物理场关系的深度认知模型，实现基于海杂波探测场景物理场参数的海杂波多维特性深层次认知。研究成果可为建立工程实用化海杂波模型、杂波抑制、目标检测提供理论依据和技术支撑。

随着当前海洋权益国际竞争的战略要求，传统的经验认知方法难以满足“低、小、慢”目标检测对海杂波多维特性深度认知的需求。为保障探测系统在复杂海洋环境下的可靠性以及发挥最佳性能和效能，对海杂波特性研究的需求越发迫切，要求更高。本项目利用海洋环境气象数据同化结合GPU高性能并行计算技术，建立大区域多尺度真实海面精细几何结构；研究重力波调制张力波海面的高阶Bragg散射，以及白冠泡沫、破碎波和多尺度海面的多次散射和机理；建立高海情多尺度电大区域海面小擦地角下电磁散射特性、全极化散射场、海杂波特性和频谱特性模型，分析不同海域海情、擦地角、极化、频率对海杂波特性的影响；结合不同海域海杂波实测数据，基于深度学习研究海杂波统计特征和海洋环境多物理场要素的关联关系，揭示不同海域海面高海情小擦地角下海杂波差异性的主因子和内在机理。实现海杂波特性的深度认知，为建立工程实用化海杂波模型、杂波抑制、目标检测，以及对低高分等需求，提供理论依据和技术支撑。

内容获取失败，请点击重试