NASA SBIR 2003 Solicitation


PROPOSAL NUMBER:03-E1.05-7146 (For NASA Use Only - Chron: 035861)
SUBTOPIC TITLE:Active Microwave
PROPOSAL TITLE:Novel Polarimetric SAR Interferometry Algorithms

SMALL BUSINESS CONCERN (Firm Name, Mail Address, City/State/Zip, Phone)
Vexcel Corp
4909 Nautilus Court
Boulder ,CO 80301 - 3242
(303) 583 - 0200

PRINCIPAL INVESTIGATOR/PROJECT MANAGER (Name, E-mail, Mail Address, City/State/Zip, Phone)
Mark D Tabb
4909 Nautilus Court
Boulder ,CO  80301 -3242
(303) 583 - 0258
U.S. Citizen or Legal Resident: Yes

Polarimetric radar interferometry (PolInSAR) is a new SAR imaging mode that is rapidly becoming an important technique for bare earth topographic mapping, tree height estimation, and small-scale surface deformation monitoring in vegetated and snow covered regions. Vexcel has developed two processing techniques which may prove to be important for the successful realization of operational PolInSAR systems. First, primarily ad hoc techniques have been developed for extracting information from PolInSAR data with the result that estimated geophysical parameters have lower accuracy than necessary and associated confidence bounds cannot be computed. This is unacceptable for both scientific and commercial mapping applications. We have developed a maximum likelihood inversion formalism which yields the optimal estimates with confidence bounds. In addition, PolInSAR systems are thought to require extremely accurate calibration, possibly resulting in significant extra expense for antennas with high crosspol isolation, etc. We have developed a PolInSAR specific calibration algorithm that alleviates these problems, thereby reducing the cost of a PolInSAR system. Phase I concentrates on developing extensions to the basic techniques and theoretically validating their usefulness. Phase II focuses on experimentally validating the techniques using data from the various PolInSAR systems that will become operational in the next 1-2 years.

An understanding of the system level requirements and capabilities of a PolInSAR sensor is important for any PolInSAR systems that JPL is contemplating building. For example, the proposed PolInSAR calibration algorithm may significantly reduce the calibration requirements of a PolInSAR system, and the maximum likelihood inversion formalism is important for determining anticipated accuracies of various estimated geophysical parameters as a function of various system design decisions. Finally, successful realization of these techniques should enable a number of applications of interest to JPL such as biomass estimation and surface deformation in vegetated and snow covered areas (forests, glaciers, snowcapped volcanoes, etc.).

A number of satellites capable of repeat pass PolInSAR are scheduled for launch in the next 1-2 years. This will enable a number of commercial applications, including bare earth mapping, tree height estimation for forestry applications, and surface subsidence monitoring in vegetated areas. Surface subsidence is of interest to numerous cities that have problems due to aquifer depletion, as well as oil and gas companies. The single polarization implementation of the approach has limited functionality in vegetated areas