Citation Link: https://doi.org/10.25819/ubsi/10153
Radar target classification via sparse decomposition
Alternate Title
Radar-Zielklassifizierung über die Lösung eines unterbesetzten Gleichungssystem
Source Type
Doctoral Thesis
Author
Issue Date
2022
Abstract
Recognition and Identification of targets are crucial steps in the radar signal processing chain. Due to their high resolution, imaging radars are a well suitable choice for these tasks. This thesis presents a framework, which is based on sparse decomposition of radar images, to identify different kinds of scattering mechanisms. The aim of this framework is the identification of specific parts of certain targets. These specific scattering centers can be used in further steps of identification, e.g. to limit the number of possible targets.
As application example, the separation of echoes from jet engines and isotropic scattering centers is used in this thesis. To model the echoes from jet engines a waveguide model is used, which is common in the modeling of radar echoes. The basic principles of electromagnetic wave propagation in waveguides, which are necessary to understand the model, are explained in this thesis. Finally, a universal model is available, which can be combined with arbitrary waveforms. As an example the change of the time-frequency behavior of a chirp waveform is shown.
The separation of the echoes themselves is done with an algorithm from the domain of solving sparse systems of equations. From this area of mathematics, some basic relationships and elementary solution methods are presented. The method used in this thesis is called morphological component analysis and was developed to decompose signals into components with different structures. Based on this history, this method seems to be particularly suitable to solve the underlying problem and is therefore presented in detail.
The basic procedure of the method is then illustrated by a simulation with different canonical forms, i.e. points, squares, circles, crosses and lines. To evaluate the ability of the algorithm to separate engine echoes and isotropic point scatterers, a simulation with these two types of echoes is performed. Finally, a real data set of the Tracking and Imaging Radar (TIRA) of Fraunhofer FHR is used to show results with a real radar system.
As application example, the separation of echoes from jet engines and isotropic scattering centers is used in this thesis. To model the echoes from jet engines a waveguide model is used, which is common in the modeling of radar echoes. The basic principles of electromagnetic wave propagation in waveguides, which are necessary to understand the model, are explained in this thesis. Finally, a universal model is available, which can be combined with arbitrary waveforms. As an example the change of the time-frequency behavior of a chirp waveform is shown.
The separation of the echoes themselves is done with an algorithm from the domain of solving sparse systems of equations. From this area of mathematics, some basic relationships and elementary solution methods are presented. The method used in this thesis is called morphological component analysis and was developed to decompose signals into components with different structures. Based on this history, this method seems to be particularly suitable to solve the underlying problem and is therefore presented in detail.
The basic procedure of the method is then illustrated by a simulation with different canonical forms, i.e. points, squares, circles, crosses and lines. To evaluate the ability of the algorithm to separate engine echoes and isotropic point scatterers, a simulation with these two types of echoes is performed. Finally, a real data set of the Tracking and Imaging Radar (TIRA) of Fraunhofer FHR is used to show results with a real radar system.
File(s)![Thumbnail Image]()
Loading...
Name
Dissertation_Simon_Wagner.pdf
Size
7.78 MB
Format
Adobe PDF
Checksum
(MD5):e38e53104be0e0b6e7c875cc66587f34
Owning collection