The modern synthetic aperture radar (SAR) has evolved into a very sophisticated instrument over the fifty-five years since its introduction in 1951. Historically, SAR has been viewed and utilized primarily as a day/night all-weather imager that can be employed in situations when optical or IR systems cannot function. Because the spatial resolution of SARs now rivals that of optical imagers, SAR images have found widespread use in both military and civilian applications. In the last fifteen years, however, SAR has taken on an entirely new dimension beyond high resolution imaging. This is the subject of coherent pair processing, wherein pairs of SAR images are collected in a manner that renders them mutually coherent. Mutual coherence allows at least two important products to be produced. One product is a high-accuracy digital terrain elevation map (DEM), and the other is a coherent change detection image. This lecture begins with an overview of SAR, built upon a framework using the signal processing concepts of three-dimensional tomography. Following the overview is a discussion of how SAR coherent pair processing can be viewed as a straightforward extension of this model. The mathematical concepts are supported by a variety of examples from real SAR imaging systems.
Dr. Charles V. Jakowatz. Jr. is currently leader of the radar signal processing research and development group at Sandia National Laboratories in Albuquerque, NM. Before joining the Labs in 1976, he received the BS, MS, and Ph.D. degrees from the School of Electrical Engineering at Purdue University in 1972, 1973, and 1976, respectively. He is author of a text on synthetic aperture radar imaging, as well as numerous research papers in the fields of signal and image processing. Dr. Jakowatz is a member of the IEEE and the National Academy of Engineering.
return to EECS500 Homepage