Here’s part two of our July 23rd Summer Seminar Series double up. These talks are open to all, and are a great opportunity to spur discussion and collaboration. Be sure to make time to attend!
Speaker: Andrew Willis, Associate Professor, Department of Electrical and Computer Engineering, University of North Carolina
Topic: Hardware Acceleration for Synthetic Aperture Radar Simulation
Location: UF-REEF Auditorium; 1350 N. Poquito RD, Shalimar, FL
Time: Tuesday, 23 July, 3:00 PM
RW Sponsor: Ms. Jamie Gantert, RWWI
Abstract: Synthetic Aperture Radar (SAR) is a critical sensing technology that is notably independent of the sensor-to-target distance and has numerous cross-cutting applications, e.g., target recognition, mapping, surveillance, oceanography, geology, forestry (biomass, deforestation), disaster monitoring (volcano eruptions, oil spills, flooding), infrastructure tracking (urban growth, structure mapping). SAR uses a high-power antenna to illuminate target locations with electromagnetic radiation, e.g., 10GHz radio waves, and illuminated surface backscatter is sensed by the antenna to generate images of structures. This talk discusses simulation of SAR systems using newly available ray-tracing hardware made available commercially through the NVIDIA's RTX graphics cards series. The RTX architecture offers completely new GPU software and hardware acceleration structures designed to provide real-time geometric ray tracing and deep learning. Simulation of SAR via this technology promises to significantly reduce the processing time required for SAR simulation which is both compute and data intensive; often requiring high performance computing solutions to produce simulations results in acceptable time limits. The talk will discuss SAR image formation principles and an overview of the most important aspects of the related theories for simulation. A focus is placed on how NVIDIA's OptiX library serves to hardware-accelerate the costly electromagnetic propagation / geometric ray-tracing component of the simulation. Demonstrations of a prototype RTX-accelerated SAR simulator will be provided for monostatic spotlight and stripmap SAR imaging modalities and details of the system design and simulation performance trade-offs. This talk may also be of interest to researchers in graphics or physical optics where ray-tracing plays an important role.
Bio: Andrew Willis is an Associate Professor in the Department of Electrical and Computer Engineering at the University of North Carolina at where he directs the UNC Charlotte VisionLab. He received a Dual Computer Science and Electrical Engineering B.Sc. from Worcester Polytechnic Institute (WPI) in 1995 after which he served as a field engineer in high-speed controls. In 1998, he pursued graduate study at Brown University where he received Sc.M. degrees in Electrical Engineering (2001) and Applied Mathematics (2003) ultimately completing his Engineering PhD 2004. After a 1-year post-doctoral post at Brown, he assumed a professor position at the University North Carolina at Charlotte (2005) where he continues to work. His research focuses on 3D computer vision, image processing, machine learning, computer graphics, and stochastic inference problems. Past projects have been sponsored by numerous agencies including NRC, AFRL, NSF, NIH, and NASA. Dr. Willis has published over 50 peer-reviewed articles on these subjects and serves on several conference committees and proposal review panels in this area of research. He is a Senior member of the IEEE and ACM.