| 摘要: |
| 针对对海探测雷达面临的海面漂浮小目标检测问题，本文主要研究了海杂波多普勒谱Hurst指数特征的统计特性，并提出了一种基于多普勒谱Hurst指数的目标检测方法。首先，本文基于实测数据，详细分析了时间序列长度（脉冲数）、快速傅里叶变换频率点数、极化方式和海况等级对海杂波多普勒谱分形特性及多普勒谱Hurst指数统计特性的影响；其次，利用6种经典统计分布模型（高斯分布、瑞利分布、韦布尔分布、对数正态分布、K分布和KK分布），验证了海杂波多普勒谱Hurst指数的最佳统计分布模型为对数正态分布，进而引入Log?t检测器，提出基于多普勒谱Hurst指数的恒虚警检测方法；最后，使用IPIX数据验证了所提方法的有效性。结果表明，所提方法具有较好的小目标检测性能和稳定性，且在VV极化数据中性能最佳。 |
| 关键词: 目标检测 海杂波 特征统计特性 多普勒谱Hurst指数 恒虚警检测 |
| DOI：DOI:10.3969/j.issn.1672-2337.2023.04.001 |
| 分类号:TN957.51 |
| 基金项目:国家自然科学基金（No.62101583，61871392）；泰山学者工程资助 |
|
| Characteristic Analysis on Hurst Index of Sea Clutter Doppler Spectrum and Target Detection |
|
DONG Yunlong， ZHANG Zhaoxiang， LIU Ningbo， HUANG Yong， DING Hao
|
|
Institute of Information Fusion, Naval Aviation University, Yantai 264001, China
|
| Abstract: |
| Aiming at the problem of detecting small floating targets on the sea surface, this paper mainly studies the statistical characteristics of the Hurst index of sea clutter Doppler spectrum, and proposes a target detection method based on the Hurst index of Doppler spectrum. Firstly, based on the measured data, this paper analyses in detail the effects of the time series length （the number of pulses）, the number of frequency points of fast Fourier transform, polarization mode and sea state level on the fractal characteristics of sea clutter Doppler spectrum and the statistical characteristics of Hurst index of Doppler spectrum. Secondly, this paper uses six classical statistical distribution models （Gaussian distribution, Rayleigh distribution, Weibull distribution, lognormal distribution, K distribution and KK distribution） to verify that the optimal statistical distribution model for the Hurst index of sea clutter Doppler spectrum is lognormal distribution, so the Log?t detector is introduced and a constant false alarm detection method based on the Hurst index of Doppler spectrum is proposed. Finally, the effectiveness of the proposed method is verified using the IPIX dataset. The results show that the proposed method has good small target detection performance and stability, with the best performance in VV polarization data. |
| Key words: target detection sea clutter statistical characteristics of feature Hurst index of Doppler spectrum constant false alarm rate detection |
