# InSAR-DInSAR 技术细节（四） Range spectral shift

Range spectral shift这个问题在斜坡效应中其实已经说过了。理解这个问题的关键在于：

• 知道频率是相位的微分（但这个（干涉数据的）频率跟载频、采样频率等其他的频率并不能混为一谈）。这样就能够很好地理解“斜坡效应”部分根据频谱偏移的公式解释条纹密度随着地形坡度的关系，因为频谱偏移本质上就对应相位变化；
• 观测视角的变化对应频率的变化；
• 这个问题和斜坡效应、失相关以及critical baseline都是相通的

## 频谱偏移和相位梯度那些事儿（由此导出了一个条件限制）

Other sources of decorrelation are more significant and non-reversible. The two most important conditions are related to the phase gradient and the temporal variation in the physical distribution of the elementary scatterers.
The phase gradient condition can be conveniently described in the spectral domain. The temporal bandwidth of the SAR images in range corresponds with a spatial bandwidth due to the projection on the earth’s surface. A phase gradient in range of n cycles/pixel corresponds with a spectral shift between the spectra of both acquisitions of n*f Hz, where f is the sampling frequency.
The spectral shift results in a decreased overlap between the corresponding parts of the spectrum (the signal) and an increasing non-overlapping part of the spectrum (the noise). Due to the limited bandwidth, a phase gradient larger than B/f cycles/pixel (approximately 0.822 for ERS) results in a zero overlap between the spectra, hence a complete loss of correlation. The occurrence of this situation is dependent on: the length of the perpendicular baseline, the steepness of the topographic slopes, and/or the gradient of the surface deformation.

## 参考文献

Hooper, A., Bekaert, D., Spaans, K., & Arıkan, M. (2012). Recent advances in SAR interferometry time series analysis for measuring crustal deformation. Tectonophysics, 514, 1-13.
Zhong, L., & Dzurisin, D. (2014). Insar imaging of aleutian volcanoes. Springer Praxis Books, 2014(8), 1778–1786.
Ketelaar, V. (2009). Satellite radar interferometry : subsidence monitoring techniques.