经验模式分解后海杂波的分形特性

针对实测海杂波数据信号进行分析，判定海杂波信号具有多重分形的特性.鉴于海杂波是一种非线性非平稳性的雷达回波信号，充分发挥经验模式分解（EMD）的优势，并结合多重分形的特性，提出一种新的海杂波背景下的目标检测方法.首先，使用EMD方法将海杂波信号分解为若干个固有模态函数分量（IMF）；然后，利用多重分形趋势起伏分析法(MF-DFA)求主IMF分量的广义Hurst指数；最后，通过实测的海杂波数据进行训练和测试.研究结果表明，该方法可有效实现海杂波下的目标探测，且性能优于经典时域和分数阶傅里叶变换(FRFT)域下的广义Hurst指数的目标检测方法.
The data of sea clutter was analyzed to determine the multi-fractal characteristics of sea clutter signals. In view of fact that the sea clutter is nonlinear and non-stationary radar echo signal, a new target detection method in sea clutter was proposed considering the advantages of empirical mode decomposition (EMD) and multi-fractal properties. Firstly, the sea clutter signal was decomposed into several intrinsic mode function components (IMF) by using EMD, then multi-fractal detrended fluctuation analysis was utilized to calculate the generalized Hurst exponent for the main functions of IMF. Finally, the training and testing were completed for measured sea clutter data. Results show that the proposed method can effectively realize the target detection in the sea clutter, and the performance is much better than that of the generalized Hurst exponent detecting method in time domain and fractional Fourier transform (FRFT) domain.