基于频域“CLEAN”抑制WVD中的交叉项

Crossterm suppression in the Wigner-Ville distribution of multicomponent sinusoids based on “ CLEAN ” in frequency domain

针对多分量正弦信号 ,采用信号分解的思想 ,提出一种基于频域“CLEAN”抑制交叉项的方法 ,即先在频域利用“CLEAN”的技术将信号逐一分解 ,然后对每一个分解后得到的单分量信号进行时频分析(Wigner Ville分布 ) .这样既可以避免交叉项的干扰 ,又能保持较高的信号时频分辨率 .与“CLEAN”法相比 ,该方法可以避免各信号分量持续时间及缓变包络等因素的影响 ,更适合于逆合成孔径雷达成像等实际应用领域 .

Based on the idea of signal decomposition and with multi sinusoids as the signal model, a new method to suppress crossterms in Wigner ville distribution (WVD) by use of the “ CLEAN ” technique in frequency domain is prosposed, in which the signal is first decomposed into different single sinusoids and then WVD of each decomposed signal is computed. The crossterms can be diminished and a high time frequency resolution obtained. Compared to “ CLEAN ”, the influence of different durations and a slowly varying envelope of each single component can be avoided, while instantaneous frequencies are kept correct, which is more attractive in applications such as ISAR imaging. The experiments by use of simulated and real world data confirm the method.