Research On Ionospheric Effects In Spaceborne P Band Synthetic Aperture Radar

Spaceborne synthetic aperture radar(SAR) is capable of imaging large area of the earth at all-time and all-weather with fine resolution. It is most valuable in application of military reconnaissance and economic constructions. P band SAR possesses remarkable advantages over higher frequency SAR systems in two aspects: Firstly, due to P band SAR’s strong capability of penetrating things, it can image areas through superficial ground and vegetation, which makes it an effectual technique for discovering and detecting hidden targets covered by superficial ground, vegetation, ice, snow and etc. Second, P band SAR is sensitive to biomass, which enables it be capable of measuring the forest biomass with high precision. Thus, spaceborne P band SAR will possess P band SAR’s speciality of strong penetration capability and high sensitivity to biomass, as well as spaceborne SAR’s virtue of large area imaging, this indicates extensive applications of future spaceborne P band SAR at reconnaissance of hidden military targets, biomass measurement, ice structure detecting and etc. However, spaceborne P band SAR’s development is bottlenecked by severe ionospheric propagation effects such as dispersion, scintillation and Faraday rotation(FR). This paper focuses on the ionospheric effects on spaceborne P band SAR. Three related problems, i.e. analyzing, simulating and correcting the ionospheric effects, are studied systematically. Major works of this paper are listed as follows:Ionospheric propagation effects are introduced in chapter 2. Firstly, the dispersive, random, anisotropic characteristic of the ionosphere are expatiate from the aspect of its refractive index and the distribution of the free electrons. Secondly, mechanism of the background ionospheric dispersion and FR effects are derived from Maxwell’s equations, whilst ionospheric scintillation is explained theoretically by phase screen theory.Numerical simulating methods of ionospheric effects in SAR based on the multiple phase screen theory are studied in chapter 3. The single phase screen mothed which is widely used is valid only when the scintillation is weak. The multiple phase screen method is used to solve the problom of in simulating the ionospheric effects for strong scinttilation in spaceborne SAR. The parabolic wave equation is solved to model the diffraction effects between the phase screens, which is obeyed by a spherical wave propagating out of the phase screen with an inclined incident angle. A SAR simulator is designed and realized using the proposed method to simulate ionospheric effects in spaceborne P band SAR data.Chapter 4 is dedicated to study the ionospheric effects on SAR based on modeling the impulse response matrix(IRM) of a spaceborne polarimetric SAR. Ionospheric dispersion, scintillation and FR effects are modeled in SAR’s IRM and their effects on SAR imaging are analyzed based on this model. Ionospheric un-linear phase error(IUPE) is proposed as a measure of the extent how bad SAR imaging’s performance is under ionospheric effects. Thus the measurement is less dependent onnumerical simulation and the evaluating progress is simplified. Performances of SAR imaging are evaluated quantitatively according to the ambiguity function’s(AF) second moment. Numerical simulation validates the theoretical analysis. Polarimetric dispersion effects on SAR imaging are analyzed based on the system’s IRM.Correcting of the ionospheric effects in SAR data are studied in chapter 5. A TEC(Total Electron Contents) estimation method based on the split-spectrum ideal is proposed to correct the dispersion effects. This method does not need to measure the TEC by accessorial equipment, and the dispersive effects in SAR image can then be self-corrected. Relationship between the signal models of ionospheric scinttilation affected SAR and the signal models from which autofocus method are deduced are revealed. Robusetness of autofocus method for correcting the ionospheric scinttilation effects is analyzed and validated. A new FR angle(FRA) estimator based on polarimetric covariance matrix data is proposed. The proposed estimator has particularly low RMS(Root Mean Square) error than the pre-existing estimators under the affect of phase imbalance and the error changes little with the true FRA. Simulations and ALOS PALSAR data processing validate the mentioned correcting methods.