Page 81 - Layout 1

Department of Engineering, ICT and Technologies for Energy and Transport Patent Title Method for filtering interferometric data acquired by synthetic aperture radar (SAR). Ref. CNR 10285 Assignee(s): CNR CNR Institute: IREA Main Inventor: Gianfranco Fornaro Countries: IT, EP Priority date: 19/07/2013 Abstract The present invention refers to a technique - named CAESAR (Component Extraction and Selection SAR) - for filtering multipass interferometric data acquired by synthetic aperture radar (SAR). This technique allows the extraction of the principal components of the data operating, on one hand, a data filtering aimed to limit the effects of temporal and angular variations of the target response and, on the other hand, a selection of interfering components associated with different structures on ground. In particular this interference effect has a dramatic impact in the analysis of urban areas. The innovation of CAESAR consists of allowing such an extraction and selection even in a preliminary stage of the data processing, thus significantly improving the performances of the method used for interferometric SAR data analysis. Background Multipass Differential SAR interferometry (DInSAR) is a cost effective technique for deformation monitoring from satellite. SAR tomography is an evolution of this technique which enables an advanced monitoring at the scale of single buildings. It is also able to solve the problem of interference of different backscattering contributions. SAR tomography assumes that the data are calibrated through the products generated by the low-resolution processing based on classical SAR interferometry techniques. Technology CAESAR exploits the method of principal components analysis which is based on the estimation of the covariance matrix associated with multipass SAR data. By taking advantage of this statistical analysis method, CAESAR allows one to operate a filtering of the whole SAR dataset and to carry out an effective separation of the components associated with different ground backscattering mechanisms interfering within the same SAR image pixel, prior to the data calibration. Advantages and Applications Within the DInSAR analysis context, CAESAR is capable to filter and separate possible interfering contributions at an early processing stage, to improve the monitoring products in areas prone to natural hazards. On the other hand, with reference to the analysis at the scale of buildings and infrastructure, with respect to SAR tomography, CAESAR has shown its potentiality in significantly increasing the measurements density and / or reducing the dataset size. Development stage A software prototype developed in Interactive Data Language (IDL) has been implemented. The method has been validated on different datasets acquired by the SAR sensors of the COSMO/SKYMED constellation. Recently, the effectiveness of CAESAR has been also verified within the framework of the processing aimed 71 at monitoring at the scale of single buildings and infrastructures.