back to projects

A Study of Inverse Methods for Processing of Radar Data

Application of geophysical inversion algorithms for their potential application to the radar image processing and improving the quality of radar signals by reducing effects of clutter.

Participants: Phil Bording, Ph.D., Aya Sayedelahl, Mohamed F. Chouikha, Ph.D., Jianchao Zeng, Ph.D, Jim Humphries

Description

We are studying the known geophysical inversion algorithms for radar image processing to improve signal and reduce the effects of clutter. These seismic algorithms use a variety of mathematical and numerical techniques for the imaging problem. However, the key feature of most of these inverse methods is a successful forward modeling technique. So if a good forward radar modeling scheme exists then we have some hope for developing a successful imaging algorithm.
In geophysics the subsurface sound speeds are not uniform and further have to be estimated from the data as an inverse problem. This sound speed parameter estimation problem is a key part of the geophysical inverse problem as well as the reconstruction of the geological subsurface structure. When sufficient source/receiver coverage is used the specular reflection data has sufficient aperture to allow both accurate structural resolution and parameter estimation of the pressure wave (acoustic) velocity's. In practice these seismic imaging methods are used worldwide to construct reliable geological subsurface images on the scale of hundreds of wavelengths.

RADAR

Radar (Radio Detection And Ranging), is used in many applications to locate and identify objects. Radar signals bounce off objects in their path, and the radar system detects the echoes of signals that return. Radar can determine a number of properties of a distant object, such as its distance, speed, direction of motion, and shape. Radar can detect objects out of the range of sight and works in all weather conditions, making it a vital and versatile tool for many industries.

Radar Transmissions Reflections & Clutter

Inverse Problems in Geophysics

A compute expensive but very successful method that uses the approach of running the wave equation backward in time. This uses the inverse of the modeling equation for imaging. First published by Hemon it was popularized by Whitmore after Kelly demonstrated successful acoustic modeling codes. Currently for real three dimensional elastic imaging pre-stack reverse time methods are the most accurate and are competitive in cost. Used to estimate sound speeds in subsurface from source/receiver acoustic data and reconstruct geological subsurface structure.

Seismic Inverse Approach

Existing Seismic Inverse Approaches:

• Kirchhoff method
• Time-domain approach developed by Bleistein and Cohen at the Colorado School of Mines
• Use an asymptotic integral method to construct a migration image of the point-to-point diffraction seismic data
• Requires accurate estimate of travel time from subsurface targets to source receivers on the surface
• Frequency-domain method
• Ground penetrating radar method: Improvement on frequency-domain method by considering attenuation and dispersion effects

Combined Method

Radar Modeling and Imaging
• Use known geophysical algorithms to model radar clutter and imaging problems
• Clutter in radar return signals are caused by surrounding objects such as trees, dusts, and buildings
• Prototype radar clutter models in this project