An Advanced Time-Series InSAR Approach Based on Poisson Curve for Soft Clay Highway Deformation Monitoring

Lingjie Zhu,Xuemin Xing,Yikai Zhu,Wei Peng,Zhihui Yuan,Qing Xia
DOI: https://doi.org/10.1109/jstars.2021.3100086
IF: 4.715
2021-01-01
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Abstract:Monitoring surface deformation for highways built on soft clay subgrades is fundamental for understanding the dynamics of the settlement process and preventing the occurrence of safety accidents. Most of the traditional interferometric synthetic aperture radar (InSAR) deformation models for highway monitoring are based on a combination of one or several empirical functions, which ignores the underground mechanism of highway deformation. To overcome this limitation, we proposed an advanced InSAR approach that improved the InSAR deformation model and parameter estimation algorithm for soft clay road deformation monitoring. The improved InSAR deformation model is based on the Poisson curve, which considers the characteristics of the temporal physical deformation evolution of the soft clay. The improved algorithm for the unknown parameter estimation is based on a GARN algorithm, which can solve the parameters with better accuracy by combining the genetic algorithm and the regularized Newton iterative algorithm. To evaluate the better performance for the proposed approach and its feasibility for soft clay highway monitoring, both the simulation and the real data experiments on a soft clay highway in Foshan, China, are performed. The time-series deformation from January 1, 2015 to January 18, 2017, is retrieved, and the temporal deformational characteristics over this area are analyzed, which facilitates a greater understanding of the deformational evolution process of the soft clay highway. The results are verified in terms of the residual high-pass deformation and the final obtained vertical deformation compared with the external leveling measurements, which indicate the greater reliability of our method.
imaging science & photographic technology,remote sensing,engineering, electrical & electronic,geography, physical
What problem does this paper attempt to address?
The paper attempts to address the issue of long-term settlement monitoring of highways built on soft soil subgrades. Specifically, traditional Interferometric Synthetic Aperture Radar (InSAR) deformation models, when monitoring highway settlement, are usually based on empirical function combinations and ignore the impact of underground mechanisms on highway deformation. This not only affects the accuracy of time-series deformation results but also complicates subsequent displacement interpretation. To overcome this limitation, the authors propose an advanced InSAR method based on the Poisson curve for monitoring the deformation of highways on soft soil subgrades. This method improves the InSAR deformation model and parameter estimation algorithm, where: 1. **Improved InSAR Deformation Model**: Based on the Poisson curve, it considers the characteristics of the physical deformation evolution of soft soil over time. 2. **Improved Parameter Estimation Algorithm**: The GARN algorithm, which combines Genetic Algorithm (GA) and Regularized Newton Iteration Algorithm (RN), can more accurately solve for unknown parameters. Through simulation experiments and actual data experiments, the performance and feasibility of this method in monitoring the deformation of highways on soft soil subgrades were verified. The research results indicate that this method can more accurately describe the deformation evolution process of soft soil subgrades and shows higher reliability when compared with external leveling measurement results.