|
|
|
“Imaging with Radar Returns”
by Chung-Ching Chen
August 1978
This dissertation presents both analytic and processing techniques for various radar imaging systems.
A two dimensional system classification method, which is very general and hence applies to the special case of radar imaging systems as well, is proposed to assist in understanding the structure and describing the limitations of 2-D systems. Once a given system is identified with the simplest possible class, the specific techniques can be directly utilized to process the data or reconstruct the images.
Following a review of radar imaging principles, several coherent radar systems are analyzed and experimented upon. They include synthetic aperture radar (SAR) ground mapping, imaging of an aircraft target from turntable data, and imaging of a flying aircraft target. In each case the point spread function (PSF) of the imaging system is derived or estimated. Physical considerations are then incorporation in mathematical PSF's to categorize the imaging systems according to the aforementioned system classification principle proposed. Degrees of Freedom (DOF) under different imaging geometries are analyzed as a means to determine the amount of information present in the usually huge amount of raw radar data for the purpose of efficient computation and minimal storage requirement. Motion compensation, range curvature, range alignment, de-chirping, FFT, registration and side lobe reduction problems are all addressed and experiments are performed using data from RAT-SCAT (for turnable imaging) and other facilities. The results shown suggest the versatility of coherent radar imaging.
Possible extensions of the current work are discussed. The understanding of the system characteristics, in particular the formulation of the radar image will aid in the advancement of techniques for radar image enhancement, encoding, quantization, and restoration.