Exploiting Multipath for Efficient Target Classification

The objective of this effort is to develop innovative methods to improve the efficiency and accuracy of Radar Cross Section (RCS) based target classification by exploiting naturally occurring multipath signals. Our goal is to develop a mathematical basis of the multipath exploitation concepts to be incorporated into an analytical model which will be used to prove (or reject) the utility of the approach. We are developing promising approaches incorporating multipath isolation processing and compressive sensing techniques to reconstruct the target’s free-space scattering behavior. In addition, we plan to utilize electromagnetic-based multipath modeling as an integral part of our Phase I research. Ultimately, we will use experimentation, in which live data from a controlled multipath environment is collected and processed off-line to develop the RCS signature of an actual target. The techniques to be developed will provide surveillance and tracking radars, or similar radars, improved RCS based target classification by exploiting naturally occurring multipath. Our focus during this Phase I effort will be on low frequency UHF surveillance radar applications.