EMD算法及其在电磁成像中的应用

发布时间：2018-10-07 19:17

【摘要】：逆合成孔径雷达(Inverse Synthetic Aperture Radar,ISAR)成像作为一种重要的雷达成像方式,具有全天候、全天时、远距离对目标进行观测和定位的能力。传统的基于傅里叶变换的ISAR成像方法能够对合作运动目标得到较为清晰的目标图像,而对于非合作运动目标,基于傅里叶变换的方法得到的目标图像存在严重的散焦现象。利用时频变换(Time-Frequency Transform,TFT)替代传统的傅里叶变换的方法成像,可以较好的解决由目标复杂运动导致的图像散焦问题,从而得到较清晰的目标图像。在各类时频变换方法中,WVD(Wigner-Ville Distribution)拥有较好的的时频分辨率,但同时多分量信号的WVD变换存在较严重的交叉项干扰问题,使得直接利用它进行ISAR时频成像效果仍不太理想。为了提高成像效果,本文在基于WVD变换的ISAR时频成像方法的基础上,重点研究了利用经验模态分解降低WVD变换交叉项干扰的方法,并结合仿真验证了该方法的有效性。在此基础上进一步研究了基于CEMD(Complex Empirical Mode Decomposition)-WVD的ISAR时频成像方法,通过仿真分析可以看出基于CEMD-WVD的ISAR时频成像方法在一定程度上抑制了WVD变换的交叉项干扰,使得成像效果有了很大程度的提高。但由于CEMD在对成像数据中的距离单元进行分解时会产生一定的虚假频率,会导致目标图像中任然存在一些虚假的散射点。为了进一步提高成像效果,本文分析对比了BEMD(Bivariate Empirical Mode Decomposition)和CEMD对多分量信号的分解能力,发现BEMD在对多分量信号进行分解时产生的虚假频率更少,能够更准确的分解出原信号中频率分量,因此本文提出了基于BEMD-WVD的ISAR时频成像方法,与基于CEMD-WVD的ISAR时频成像方法相比,基于BEMD-WVD的时频成像方法产生的虚假散射点更少,成像效果也更理想。同时受到希尔伯特-黄变换(Hilbert-Huang Transform,HHT)的启发,本文提出了基于BEMD和瞬时频率的ISAR时频成像方法,该方法从根本上避免了WVD变换交叉项干扰对成像效果造成的影响,与基于WVD变换的ISAR时频成像方法相比虚假散射点大大减少,成像效果得到显著的提高,但由于瞬时频率的计算误差导致成像效果同基于BEMD-WVD的成像方法相比,并没有明显的提高,优点是成像速度要明显快于后者。
[Abstract]:Inverse synthetic Aperture Radar (Inverse Synthetic Aperture Radar,ISAR) imaging, as an important radar imaging method, has the ability to observe and locate the target at a long distance, all weather and all day. The traditional ISAR imaging method based on Fourier transform can get a clear image of the cooperative moving target, but for the non-cooperative moving target, there is a serious defocusing phenomenon in the target image based on the Fourier transform method. Using time-frequency transform (Time-Frequency Transform,TFT) instead of the traditional Fourier transform imaging can solve the defocus problem caused by the complex motion of the target and get a clear target image. WVD (Wigner-Ville Distribution) has good time-frequency resolution in all kinds of time-frequency transform methods, but the WVD transform of multi-component signal has serious crossover interference problem, which makes the direct use of ISAR time-frequency imaging not very ideal. In order to improve the imaging effect, based on the ISAR time-frequency imaging method based on WVD transform, this paper focuses on the method of reducing the cross-term interference of WVD transform by empirical mode decomposition. The effectiveness of the method is verified by simulation. On this basis, the ISAR time-frequency imaging method based on CEMD (Complex Empirical Mode Decomposition) -WVD is further studied. Through the simulation analysis, it can be seen that the ISAR time-frequency imaging method based on CEMD-WVD can restrain the cross-term interference of WVD transform to a certain extent. So that the imaging effect has been greatly improved. However, due to the false frequency generated by CEMD when decomposing the distance unit in the imaging data, there are still some false scattering points in the target image. In order to further improve the imaging effect, the ability of BEMD (Bivariate Empirical Mode Decomposition) and CEMD to decompose multi-component signals is analyzed and compared. It is found that BEMD produces less false frequency when decomposing multi-component signals. The frequency component of the original signal can be decomposed more accurately. Therefore, a ISAR time-frequency imaging method based on BEMD-WVD is proposed. Compared with the ISAR time-frequency imaging method based on CEMD-WVD, the time-frequency imaging method based on BEMD-WVD produces fewer false scattering points. The imaging effect is also better. Inspired by Hilbert-Huang transform (Hilbert-Huang Transform,HHT), a ISAR time-frequency imaging method based on BEMD and instantaneous frequency is proposed in this paper. The method avoids the influence of the cross-term interference of WVD transform on the imaging effect. Compared with the ISAR time-frequency imaging method based on WVD transform, the false scattering points are greatly reduced, and the imaging effect is greatly improved. However, because of the calculation error of instantaneous frequency, the imaging effect is not obviously improved compared with the method based on BEMD-WVD. The advantage is that the imaging speed is much faster than the latter.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN957.52

【相似文献】