Our group is focused on the development of robust and stable radar image formation algorithms for synthetic aperture radar (SAR), inverse synthetic aperture radar (ISAR), and other distributed sensing configurations.  The aim is to fundamentally improve imaging fidelity and spatial resolution through novel and computationally feasible approaches that go beyond the conventional point-scatterer model. Of interest are physics-based 2D/3D inversion algorithms that work in conjunction with techniques from deep learning, artificial intelligence, and non-linear optimization. Other efforts of the group include reconstruction and generative techniques to support large-scale synthetic image generation to improve training of deep learning classification models.

We seek postdoctoral candidates with a strong background in inverse problem theory, electromagnetic scattering, machine learning, array processing, and optimization. Strong software development skills in languages such as MATLAB, Python, C/C++ as well as knowledge of relevant frameworks such as TensorFlow is desired.

key words

Radar Imaging; Inverse Scattering; Deep Learning; Distributed Sensing; Physics-Based Machine Learning

Citizenship:  Open to U.S. citizens and permanent residents

Level:  Open to Postdoctoral applicants