In this thesis, a Synthetic Aperture Radar (SAR) calibration method present in the lit- erature is reviewed and implemented with simulated acquisitions. The method, called PSCal, uses a set of stable ground control points called Permanent Scatterers (PS) to find a set of radiometric normalization constants. The entire procedure starts from a set of non-calibrated, but coregistered set of SAR images in which a group of PS are detected. The signal model is written and all the parameters are carefully described within the thesis. The iterative maximum likelihood method to extract the parameters of interest from data is reviewed. The performances of the estimator with different number of PS and different SNR for each PS is presented with simulated results. Moreover, the model is furtherly extended to accommodate residual rolling error of the satellite. The PSs, in this case, form a set of “samplers” of the antenna pattern opening the possibility for the estimation of the residual roll of the satellite. As expected from the Cramer-Rao Bound that has been derived, the estimation of the residual roll is as accurate as the PSs sample the antenna pattern in the areas where the gradient of the latter in maximum. Also in this case, a set of simulations has been carried out to prove the effectiveness of the method.
In questa tesi, un metodo di calibrazione per radar ad apertura sintetica (in inglese: Synthetic Aperture Radar, da cui l’acronimo SAR) che è presente nella letteratura è stato rivisto ed implementato con acquisizioni simulate. Il metodo, chiamato PSCal, utilizza un serie di punti di controllo a terra stabili chiamati Permanant Scatterer (da cui l’acronimo PS) per trovare le costanti di normalizzazione radiometrica in uno stack di immagini SAR. L’intera procedura inizia da una collezione di immagini SAR coregistrate ma non calibrate. Da questo stack vengono estratti una serie di PS, da cui verrà effettuata l’analisi e la stima dei parametri di interesse. Il modello del segnale viene quindi derivato e tutti i parametri sono decritti all’interno della tesi. Per stimare i parametri di interesse viene utilizzata una stima a massima verosimiglianza implementata in modo iterativo. In questa tesi ho analizzato le performance dello stimatore al variare di numero di PS e del loro SNR. Inoltre, il modello è stato ulteriormente esteso a accolgiere l’errore residuo di rollio del satellite. Come previsto dal limite di Cramér-Rao che è stato derivato, la stima dell’errore di rollio è tanto più accurata tanto più i PS campionano il pattern di antenna nelle zone la cui derivata di quest’ultimo è massima.
Synthetic Aperture Radar Calibration by Using Permanent Scatterers
OZKAN, ILGIN
2021/2022
Abstract
In this thesis, a Synthetic Aperture Radar (SAR) calibration method present in the lit- erature is reviewed and implemented with simulated acquisitions. The method, called PSCal, uses a set of stable ground control points called Permanent Scatterers (PS) to find a set of radiometric normalization constants. The entire procedure starts from a set of non-calibrated, but coregistered set of SAR images in which a group of PS are detected. The signal model is written and all the parameters are carefully described within the thesis. The iterative maximum likelihood method to extract the parameters of interest from data is reviewed. The performances of the estimator with different number of PS and different SNR for each PS is presented with simulated results. Moreover, the model is furtherly extended to accommodate residual rolling error of the satellite. The PSs, in this case, form a set of “samplers” of the antenna pattern opening the possibility for the estimation of the residual roll of the satellite. As expected from the Cramer-Rao Bound that has been derived, the estimation of the residual roll is as accurate as the PSs sample the antenna pattern in the areas where the gradient of the latter in maximum. Also in this case, a set of simulations has been carried out to prove the effectiveness of the method.File | Dimensione | Formato | |
---|---|---|---|
Thesis_Ilgin_Ozkan.pdf
accessibile in internet per tutti
Descrizione: Ilgin Ozkan master thesis
Dimensione
5.41 MB
Formato
Adobe PDF
|
5.41 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/198832