摘要
传统机载雷达海杂波的抑制方法在估计杂波功率谱时存在人工参与度高、误差大等问题,导致环境适应性较差。为此,提出一种基于智能学习的机载海杂波谱参数估计方法,建立基于一维LeNet-5的海杂波训练模型,并将仿真和实测海杂波数据输入训练好的模型后对功率谱的中心和宽度进行估计,进而实现海杂波谱特性的直接感知。实验结果表明,与传统方法相比,文中所提方法具有更高的估计精度以及更好的鲁棒性。
Traditional airborne radar sea clutter suppression methods have a high degree of human participation and large errors in estimating the clutter power spectrum.With the development of modern signal processing and artificial intelligence,deep learning methods are used to study the sea clutter more quickly and intelligently.This paper proposes an airborne radar sea clutter spectrum parameter estimation method based on intelligent learning.It establishes a sea clutter training model based on the one-dimensional LeNet-5.Then the simulated and measured sea clutter data are input into the trained model to estimate the center and width of the power spectrum,thus realizing the direct perception of the sea clutter spectrum characteristics.The experimental results show that the proposed method has a higher estimation accuracy and better robustness than the traditional methods.
作者
范一飞
王心宝
粟嘉
陶明亮
陈明
王伶
FAN Yifei;WANG Xinbao;SU Jia;TAO Mingliang;CHEN Ming;WANG Ling(School of Electronics and Information,Northwestern Polytechnical University,Xi'an 710072,China;Hangzhou Hikvision Digital Technology Co.,Ltd.,Hangzhou 310051,China)
出处
《西北工业大学学报》
EI
CAS
CSCD
北大核心
2024年第3期446-452,共7页
Journal of Northwestern Polytechnical University
基金
国家自然科学基金面上项目(62171379)
航空科学基金(20182053024)资助。
关键词
海杂波
深度学习
多普勒谱特性
参数估计
sea clutter
deep learning
doppler characteristics
parameters estimation
