Introducing PBSR for Drastic Improvements in Radar Resolution!

Telephonics and Information Systems Laboratories are proud to introduce our advanced radar imaging capability, PBSR, which surpasses previous super resolution technique due to it’s underlying physics-based model! See below for a Real-World example of this technique being applied to a real radar system!

Motivation:

Traditional radar imaging employing linear Fourier based techniques is subject to fundamental resolution limitations referred to as the Rayleigh or diffraction limit that is on the order of the ratio l/d, where l is the operating wavelength and d is the size (length) of the aperture. For 2D SAR, range resolution is set by the reciprocal of the radar bandwidth, while cross-range is set by the length of the “synthetic” aperture. Both of these fundamental “limits” are in many circles assumed to be unavoidable. This is, of course true if and only if linear image formation is assumed. These restrictions do not necessarily hold however if nonlinear imaging is employed.

Our Unique Approach:

The PBSR approach is fundamentally different than previous superresolution techniques such as Maximim Entropy, MUSIC, ESPRIT and others because it involves an accurate physics-based assumption about the underlying received signal as opposed to more general assumptions about the statistics of the signal. This explains why we have been able to successfully apply PBSR to real-world airborne radar data (example below) and existing superresolution techniques often fail in real-world settings.

The example shows how PBSR was successfully applied to Telephonics’ radar data to significantly improved the image resolution without increasing the system bandwidth. In this case a test target array of closely-spaced targets arranged in a triangular pattern were successfully resoloved using PBSR!

Low Resolution Data

High Resolution Data using PBSR

PBSR used to achieve imagery with greater than 4x improvement in resolution!