An interferometric phase optimization method joining polarimetric and temporal dimensions

The polarimetric phase optimization method has been integrated into the multitemporal synthetic aperture radar interferometry (MT-InSAR) framework to enhance phase quality and deformation coverage, known as multitemporal polarimetric InSAR (MT-PolInSAR) technology. However, most existing MT-PolInSAR...

Descripción completa

Detalles Bibliográficos
Autores: Chen, Yaogang|||0000-0003-3800-2370, Hu, Jun, Mallorquí Franquet, Jordi Joan|||0000-0002-9424-1889, Fu, Haiqiang, Wu, Wenqing, Zhang, Leixin
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/433323
Acceso en línea:https://hdl.handle.net/2117/433323
https://dx.doi.org/10.1109/TGRS.2025.3556141
Access Level:acceso abierto
Palabra clave:Maximum likelihood estimation
Polarization
Synthetic aperture radar
Interferometry
Distributed scatterers (DSs)
Maximum likelihood estimation (MLE)
Multipolarization phase optimization
Synthetic aperture radar interferometry (InSAR)
Àrees temàtiques de la UPC::Enginyeria de la telecomunicació::Radiocomunicació i exploració electromagnètica::Radar
Descripción
Sumario:The polarimetric phase optimization method has been integrated into the multitemporal synthetic aperture radar interferometry (MT-InSAR) framework to enhance phase quality and deformation coverage, known as multitemporal polarimetric InSAR (MT-PolInSAR) technology. However, most existing MT-PolInSAR methods optimize phase separately in the temporal and polarimetric dimensions, failing to leverage the interdimensional relationships fully. This article proposes a novel multipolarization optimization method, which achieves one-step phase optimization by joining temporal and polarimetric dimensions based on a joint probability density function and maximum likelihood estimation (MLE). Additionally, a no-threshold regularization is employed to strengthen the stability of the multipolarization covariance matrix. The proposed approach has been validated through synthetic and real quad-polarization datasets. Regarding the real data, ALOS-2/PARSAR-2 from the Fengjie landslide in China and Radarsat-2 data from the Barcelona airport in Spain are used. The experimental outcomes demonstrate that our proposed approach significantly diminishes phase noise while increasing the density of measurement points.