| Abstract: |
This work describes a software tool developed for ESA within the project BIBLOS-2 (ref. TEC-SWM/15-811/CP) to model typical conical scanning microwave radiometers in a generic Earth Observation mission. Typical conical scanners have a number of channels in atmospheric transmission windows that allow to sense the radiation coming from the surface of the Earth, that is why the Earth’s Incidence Angle (EIA) is kept constant, and some other channels in atmospheric attenuation bands, which are more sensitive to atmospheric parameters, such as temperature or water vapor.
An Instrument Data Simulator (IDS) simulates the acquisition of the energy by an instrument onboard the spacecraft, up to the conversion to digital numbers (binary values). The output of the IDS are the inputs to the Ground Processing Prototype (GPP). The goal of the IDS is to simulate as realistically as needed the instrument acquisition chain, and all the associated errors. It is also a tool to validate the GPP algorithms against different configurations of errors. The IDS includes the Geometry, Scene Generator and Instrument Modules.
In the Geometry Module the following Orbit and Attitude files are used: a) Nominal Attitude or the ideal orientation of the platform computed using an attitude law (geodetic pointing, yaw steering, etc.), b) Satellite Orbit & Attitude or the real trajectory and the real orientation of the satellite, and the c) Restituted Orbit & Attitude or knowledge of the trajectory and orientation, as determined from the GNSS sensors, ranging measurements, and sensors on board the spacecraft.
The Scene Generator Module implements the Radiative Transfer Models (RTM) for passive microwave emission. The main parameters included are the physical temperature, the soil moisture, the surface roughness, the vegetation cover/water content, the fraction of the pixel area occupied by water, the snow/ice cover, the presence of snow melt, and the actual atmospheric state including water vapor, hydrometeors, aerosols, ionosphere and Earth’s magnetic field, Sun, Moon and other sky contributions. The models used are a simplification of [1].
The Instrument Module includes the antenna pattern (co- and cross-polar terms), the receiver gain and frequency response, the noise terms (Gaussian and flicker noise, discretization…), detector non-linearities etc. [2], and the presence of radio-frequency interference (RFI), which can seriously degrade the performance of microwave radiometers. Nowadays, the concern about the RFI phenomenon is increasing due to the high number of RFI occurrences detected over the years, and this problem is expected to grow even more in the future because of the pervasive use of wireless technologies around the world. In order to improve the performance of space-borne microwave radiometer instrument, the RFI detection and mitigation is considered essential, and many instruments employ the RFI mitigation in the receivers. Therefore, RFI detection/mitigation systems have also been included for the generic passive microwave simulation [3].
RFI is first modelled (direct problem) in the scenario generation either randomly or defined by the user. It is characterized by the locations, probability of occurrence, event duration, intensity, and type (pulsed, CW, wide/narrow band chirp, PRN, glitch…). Two figures of merit are commonly used to evaluate the performance of a RFI mitigation system for a fixed Probability of false alarm (PFA): the probability of detection PD, and the residual power after mitigation ∆T_(RFI,Res). For the evaluation of the detection performance, ROC (Receiver Operating Characteristic) curve is often used, which relates PD and PFA. The probability of detection PD depends on INR (interference-to-Noise Ratio), type of detection/mitigation approach, type of signal…
The complete BIBLOS-2 software will be available at [4].
References:
[1] doi: 10.3390/jimaging2020017
[2] doi: 10.3390/jimaging2020018
[3] doi: 10.1109/JSTARS.2017.2654541
[4] https://gmv-biblos.gmv.com
|
