InSAR Tropospheric Delay Correction in Mountainous Terrain
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing InSAR technologies face challenges in accurately correcting tropospheric delay, particularly in modeling seasonal oscillation signals and addressing the inconsistency between SAR spatial resolution and meteorological data spatial resolution.
Innovation Solution
A time-series InSAR tropospheric delay correction method is developed, utilizing ERA-5 atmospheric grid data to establish a negative exponential vertical profile function model. This model incorporates Fourier series functions to capture temporal variations and employs inverse distance weighting for spatial interpolation, ensuring accurate calculation of atmospheric delay at high-coherence points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If spatial and temporal filtering methods are used to correct tropospheric delay, then the random distribution assumption simplifies processing, but seasonal periodic signals cannot be effectively separated from non-seasonal random signals
Solution Approach 1:
The patent applies dynamics by transitioning from static filtering assumptions to dynamic seasonal modeling. It introduces time-varying parameters including seasonal cycle components (annual and semi-annual periods) and linear trend terms that adapt to changing atmospheric conditions, enabling effective separation of seasonal periodic signals from non-seasonal random signals while maintaining processing feasibility
Solution Approach 2:
The patent changes parameters by incorporating seasonal cycle terms (cos(2πt/365) and sin(2πt/365)), semi-annual cycle terms (cos(4πt/365) and sin(4πt/365)), and linear trend terms into the atmospheric delay model. These parameter changes allow the model to capture temporal variations in tropospheric delay and effectively separate seasonal signals from random noise
2Ease of manufacture
If linear empirical models are used to estimate tropospheric delay, then processing is simplified, but spatial variability of the atmosphere is not considered
Solution Approach 1:
The patent applies local quality by assigning unique atmospheric delay parameters to each pixel location in the SAR image. The model estimates separate intercept terms (ε₀(x,y)) and slope terms (ε₁(x,y)) for each spatial location, allowing the atmospheric delay to vary spatially across the scene. This captures local atmospheric conditions and spatial variability while maintaining the simplicity of the linear empirical model structure
3Measurement precision
If power-law model is used to account for spatial variability, then atmospheric delay variation is modeled, but the method is only suitable for stratified delay closely related to terrain with power-law variation
Solution Approach 1:
The patent changes parameters by replacing the restrictive power-law functional form with a more flexible linear model that includes location-specific intercept and slope parameters. This parameter change allows the model to adapt to various atmospheric conditions and terrain types without requiring the assumption of power-law variation, thereby improving model versatility while maintaining the ability to capture spatial variability through the elevation-dependent slope term
4Productivity
If meteorological reanalysis data is used to simulate atmospheric delay, then near-real-time estimation is achieved, but inconsistency between SAR spatial resolution and meteorological data spatial resolution causes interpolation accuracy problems
Solution Approach 1:
The patent applies local quality by estimating separate atmospheric delay parameters for each pixel location rather than using uniform interpolation across the entire scene. The model fits location-specific intercept (ε₀(x,y)) and slope (ε₁(x,y)) parameters that capture local atmospheric conditions, thereby improving interpolation accuracy at the SAR pixel level while maintaining the efficiency of using coarse-resolution meteorological reanalysis data
5Device complexity
If direct interpolation of ZTD at high-coherence point positions is performed using surrounding grid point ZTDr, then spatial variation characteristics are simplified, but interpolation errors occur
Solution Approach 1:
The patent changes parameters by transforming the interpolation approach from direct ZTD value interpolation to interpolation of the underlying model parameters (intercept ε₀ and slope ε₁). This parameter transformation allows the model to capture spatial variations in atmospheric delay more accurately by fitting local linear relationships between delay and elevation, thereby reducing interpolation errors while maintaining computational efficiency
Data Source
AI summary
Disclosed is a time series InSAR tropospheric delay correction method for relieving atmospheric seasonal oscillation, so as to relieve the seasonal oscillation deviation caused by tropospheric delay. This method is based on the temporal and spatial characteristics of atmospheric delay. Firstly, based on the ERA-5 atmospheric grid data covering the study area, the ZTD negative exponential vertical profile function model is established. Fourier series function is introduced to establish the time series model for β and ZTDr, Finally, according to the radar incidence angle parameters, the delay is changed from zenith direction to radar line of sight direction.