National Technical University of Athens (NTUA)
Προσδιορισμός τραχύτητας επιφάνειας από δεδομένα οπισθοσκέδασης ΗΜ κυμάτων με εφαρμογή σε radar τύπου SAR
Abstract
dc:descriptionThere is an extensive literature on classification of Earth’s surface using images derived from satellite or airborne SAR RADARs based on image processing techniques. The idea that led to the unique concept of this dissertation is the classification and in particular the characterization of the roughness of a surface by directly analyzing the backscattered electromagnetic (EM) waves of a RADAR without image formation and image processing techniques. In this PhD dissertation I defend this proposal on theoretical as well as on practical results. As part of the first phase, theoretical approach, I developed an EM wave scattering simulation software. For rough surfaces simulation, I use a modified 'Weierstrass-Mandelbrot' fractal function. To simulate EM wave scattering on these ‘fractal’ surfaces, I use the 'Kirchhoff approximation' (KA) model. By simulating RADAR emissions as ‘burst’ of M successive pulses of EM waves of a stepped frequency, an innovative method of determining the surface roughness is proposed and verified based on the direct analysis of the M successive intensity values of the backscattered EM waves resulting after each burst. As part of the second phase, experimental approach, I extend the research to determine the roughness of a real surface and in particular to determine the ‘sea state’ of a marine area by directly analyzing the backscattered EM waves originating from bursts of an airborne PicoSAR RADAR. For the signal preprocessing stage, namely SNR improvement, noise elimination, I use statistical analysis techniques, ‘averaging’ techniques, as well as intra-quartile range 'IQR' statistical filters. For the signal processing, namely roughness determination, I use two different algorithms. The first algorithm, 'analysis of one-dimensional Radar signatures,' processes the backscattered amplitudes of EM waves per burst of M successive pulses, i.e. as one-dimensional M-sized vectors and calculates the ‘fractal length’ of each vector separately. The mean value of ‘fractal length’ from N successive vectors equals to the MFL value, namely 'Mean Fractal Length'. As demonstrated in this dissertation, the MFL value can successfully function as measure of ‘sea state’ index. The second algorithm, ‘analysis of two-dimensional Radar signatures' processes the intensity measurements of the backscattered EM waves as a two-dimensional MxN array, where M is the number of pulses per burst and N the number of successive bursts. The intensity measurements of E thus create the dimension of height on this MxN surface. By applying 'multi-resolution fractal analysis' on this three-dimensional surface of the intensity measurements, I calculate the ‘fractal signature’. The average value Dμ of the ‘fractal signature’, namely MFS ‘Mean Fractal Signature’ proves to be a highly representative measure of the surface roughness and an accurate index of the ‘sea state’. I conclude this dissertation with a practical application of the above so as to verify the proposed method. I retrieve in chronological order the set of all ‘raw data’ recordings of the airborne RADAR and by using the MFS algorithm, a Dμ value is calculated for every 3.5 seconds of flight time. By comparing the Dμ(t) value as a measure of the ‘sea state’ with the prevailing weather conditions according to meteorological data, it follows that the proposed algorithm determines ‘sea state’ with an excellent success and consistency throughout the flight.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Kotopoulis, Apostolos
- Κοτοπούλης, Απόστολος
Subjects
dc:subject × 38- Σκέδαση ηλεκτρομαγνητικών κυμάτων
- Οπισθοσκέδαση
- Μορφοκλασματικές επιφάνειες
- Μορφοκλασματική διάσταση
- Μορφοκλασματική ανάλυση
- Μορφοκλασματικό μήκος
- Μορφοκλασματική υπογραφή
- Τραχύτητα επιφάνειας
- Θαλάσσια κατάσταση
- Αλγόριθμοι επεξεργασίας σήματος ραντάρ συνθετικής απεικόνισης
- Ραντάρ συνθετικού ανοίγματος
- Αλγόριθμοι επεξεργασίας σήματος ραντάρ
- Scattering of electromagnetic waves
- Backscattering
- Kirchhoff approximation
- Synthetic Aperture Radar (SAR)
- SAR raw data
- Fractal surfaces
- Fractal analysis
- Fractal area
- Fractal dimensions
- Fractal length
- Fractal Signature
- Modified Fractal Signature (MFS) method
- Sea state
- Rough surfaces
- Multi-resolution fractal analysis
- Radar signal processing
- Άλλες Επιστήμες Μηχανικού και Τεχνολογίες
- Επιστήμες Μηχανικού και Τεχνολογία
- Μηχανική, διεπιστημονική προσέγγιση
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Μηχανική και συστήματα επικοινωνιών, Τηλεπικοινωνίες
- Other Engineering and Technologies
- Engineering and Technology
- Engineering, interdisciplinary
- Electrical Engineering, Electronic Engineering, Information Engineering
- Communication engineering and systems, Telecommunications
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/56897