| Popis: |
Modeling the ionospheric disturbances behavior is of great importance in many SpaceGeodetic applications, including robust Spaceborne Synthetic Aperture Radar (SAR) signal processing through accurate noise modeling and removal, 3D-monitoring of land and sea surface and geodynamics SAR Interferometry (InSAR) purposes. Estimating ionospheric disturbances behavior is a challenging research issue which involves many non-linearities, dynamics and external factors. In this paper, we propose an Echo State Network (ESN) for Modeling Ionospheric Disturbances Behavior. The echo state network (ESN) is a recurrent neural network with a sparsely connected hidden layer (with typically 1% connectivity). The connectivity and weights of hidden neurons are randomly assigned and are fixed. The weights of output neurons can be learned so that the network can (re)produce specific temporal patterns. The representation is accomplished through network parameters. Thus, we come up against the problems arising from the traditional approaches which try to precompute the noisy effects of the wave propagation through ionosphere on Spaceborne SAR images. The main advantage of the ESN is that although its behavior is non-linear, the error function is thus quadratic with respect to the parameter vector and can be differentiated easily to a linear system. The model has been validated with ionospheric noise derived from real interferograms produced by earthquakes occurred in Greece the last ten years, during various ionospheric disturbances. By using the ESN for Modeling Ionospheric Disturbances Behavior, we extract the noise due to the ionospheric propagation from various comparatively processed interferograms, gathered from different highly seismicity areas in Greece. In addition, we compare the observed Total Electron Content (TEC) during severe ionospheric disturbances, using the local Ionospheric Station at the National Observatory of Athens and the main Global Position System (GPS) Station at Dionysos Satellite Observatory (DSO) of the National Technical University of Athens (NTUA), Greece, with the ionospheric TEC derived from the proposed ESN. |
