Vector deformation time-series from spaceborne motion compensation InSAR processors
Abstract/Contents
- Abstract
- We have developed an algorithm, which we call V-SBAS, that extends existing InSAR time-series techniques to estimate the temporal evolution of three-dimensional vector deformation. Our approach uses several LOS-InSAR geometries and an along-track technique called multiple aperture interferometry (MAI) to increase the temporal sampling and dimensionality over time-series in a single radar geometry. By combing multiple observations, we reduce the effect of noise, allowing us to estimate deformation in regions where other multidimensional approaches might typically fail. V-SBAS can be implemented at the backend of any InSAR data processor, but since we use hundreds of interferograms, we develop the processing chain in the context of efficient motion compensation processors. Motion compensation uses the satellite orbit information to shift the position and phase of the receive signal to that of an ideal noninertial circular reference orbit where the focusing equations are simplified and SAR scenes are produced in a common coordinate system. This approach reduces the computational complexity of interferogram formation and is well suited to produce stacks of InSAR data for time-series analysis. We first present our approach to motion compensation and develop the equations to produce stacks of interferograms in a single radar geometry. We find that processing to a common reference orbit introduces spectral shifts, which can reduce coherence for large motion compensation baselines. We derive expressions for these shifts and propose processor modifications to restore coherence. We then develop the V-SBAS algorithm and define a set of constraints that are based on the concept of space-time separability. We present an algorithm to verify the existence of this property through a principal component analysis of existing GPS data, which we use to determine whether or not the constraints should be applied. The separability constraints are nonlinear in the time-series components, thus we also propose a linear approximation to maintain the efficiency of our approach. The linear solution involves an initial unconstrained estimate, followed by a constrained re-estimate of the northward component. We choose this approach because the north is defined primarily by the noisy MAI measurements, while the east and up are defined by InSAR. Finally, we present results of the 2007 Father's Day eruption and intrusion at Kilauea Volcano, Hawaii, where we see agreement with GPS on the order of 2 cm in the east and up directions and 6 cm in the north. The separability constraints reduce the overall average errors to approximately 4 cm, however, some individual stations saw a reduction in error by as much as 10-20 cm.
Description
| Type of resource | text |
|---|---|
| Form | electronic; electronic resource; remote |
| Extent | 1 online resource. |
| Publication date | 2014 |
| Issuance | monographic |
| Language | English |
Creators/Contributors
| Associated with | Wortham, Cody Blair |
|---|---|
| Associated with | Stanford University, Department of Electrical Engineering. |
| Primary advisor | Zebker, Howard A |
| Thesis advisor | Zebker, Howard A |
| Thesis advisor | Linscott, Ivan |
| Thesis advisor | Segall, Paul, 1954- |
| Advisor | Linscott, Ivan |
| Advisor | Segall, Paul, 1954- |
Subjects
| Genre | Theses |
|---|
Bibliographic information
| Statement of responsibility | Cody Blair Wortham. |
|---|---|
| Note | Submitted to the Department of Electrical Engineering. |
| Thesis | Thesis (Ph.D.)--Stanford University, 2014. |
| Location | electronic resource |
Access conditions
- Copyright
- © 2014 by Cody Blair Wortham
- License
- This work is licensed under a Creative Commons Attribution Non Commercial 3.0 Unported license (CC BY-NC).
Also listed in
Loading usage metrics...