Improved TEC retrieval based on spaceborne PolSAR data

Abstract

It is well known that for Earth observations, the spaceborne synthetic aperture radar (SAR) systems at L‐band can be seriously affected by the ionosphere. Fortunately, the destructive ionospheric information, total electron content (TEC), can be retrieved and removed by model inversion from SAR echoes and becomes a subject of interest for ionospheric research. Considering the system noise and channel phase imbalance, this paper focuses on accurate TEC retrieval from the Faraday rotation angle, which is embedded in Advanced Land Observing Satellite phased array L‐band synthetic aperture radar (PALSAR) full‐pol data. For the consideration of denoising, averaging filtering is widely used in existing studies to remove the system noise. In order to further reduce the noise effects, a fast total variation (FTV) denoising method is applied in this paper. By using the PALSAR full‐pol data sets, a series of simulation results show that the FTV denoising is effective for noise suppression and markedly reduce the standard deviation. For the error of channel phase imbalance, a new Faraday estimator based on the covariance matrix of circular basis is proposed. Compared with previous estimators, theoretical analysis and simulation results show that this new estimator can give the best performance for sensors that are dominated by channel phase imbalance errors. Thus, the proposed estimator can be a candidate method for TEC retrieval when phase imbalance is the domain error.

A Novel Strategy of Ambiguity Correction for the Improved Faraday Rotation Estimator in Linearly Full-Polarimetric SAR Data

“…A linearly-polarized electromagnetic wave propagating through the ionosphere splits into two circularly-polarized waves, which are equal in energy but rotate in opposite directions [ 7 ]. They recombine a new wave once leaving the ionosphere, but as a result, the new wave experiences a rotation of polarization plane compared with the original wave in the presence of the geomagnetic field and the free electrons in the ionosphere [ 7 , 8 , 9 , 12 ], and this phenomenon is well known as FR. The one-way FRA in radians can be approximately expressed as where is a constant of value of , is the carrier frequency in hertz, is the geomagnetic field intensity in webers per square meter, is the angle between the wave propagation and geomagnetic vector in radians, is the local Total Electron Content (TEC) integrated along the propagation path, measured in TEC units (TECU) of electrons/ , and is the incident angle at the ionospheric altitude in radians.…”

“…Under the assumption that has no obvious variation in the course of propagation, and ignoring the noise, the measured scattering matrix can be written as [ 6 , 8 , 9 , 11 , 12 ]: where is the overall gain of the system; represents the comprehensive reflection consist of the phase delay of the propagation and the system-dependent phase; represents the channel cross-talk; is the channel imbalance; are the four elements of the true scattering matrix, are the four elements of the measured scattering matrix, respectively.…”

“…Authors have modeled and studied the FR effects on scattering matrix measurements by employing measured scattering matrix. A set of published FR estimators have been proposed to solve this issue of FR [ 8 , 9 , 11 , 12 , 13 ], and an improved FR estimator based on Bickel and Bates’s method has also been proposed, but the performance of the improved estimator has never been tested and proposed. In this paper, the improved estimator is tested in term of system noise, channel imbalance and cross-talk in detail [ 8 , 9 , 11 ].…”

“…In this paper, the improved estimator is tested in term of system noise, channel imbalance and cross-talk in detail [ 8 , 9 , 11 ]. However, all of published estimators suffer from a potential FRA ambiguity [ 5 , 8 , 9 , 11 , 12 ]. Some approaches have been devised to eliminate the FRA ambiguity, such as using specific terrain identities, phase unwrapping and FRA prediction [ 8 , 14 ].…”

Improved Faraday Rotation Estimation in Spaceborne PolSAR Data Using Total Variation Denoising

“…where N matrix is the system noise and Ω denotes the FR angle. According to Equation (1), many methods for FR retrieval have been developed [5][6][7][8]13,14]. In this work, the B&B method is selected as a reference due to its well-known robustness [15,16].…”