Instruments
Passive Optical
Instrument
The Instrument Module is in charge of simulating the sensor behaviour. The sensors for the passive optical instrument measure the energy, the radiant flux that it receives. The main types of passive optical instruments are:
- Multi-band spectrometers
- Radiometers
- Prism-grating spectrometers
- Fourier transform spectrometers.
The quality of the image is determined by the spatial, spectral and radiometric resolution.
- Spatial resolution is the size of the ground sampling distance. This how big is the pixel projected on Earth due to the Instantaneous Field Of View (IFOV).
- The spectral resolution is the size of the wavelength range of the energy recorded. The smaller the wavelength range, the better the spectral resolution.
- The radiometric resolution is the number of energy levels that the instrument can distinguish. The radiometric resolution is dependent on the number of bits the sensor has available to store the information, so the number of radiometric levels is a power of two. A sensor with 4 has 2^4 levels, from 0-15. A sensor with 8 bits has 2^4 levels, from 0-255.
There is a trade-off to be done between the spatial, spectral, and radiometric resolutions. For an instrument to have a high spatial resolution it must have a small Instantaneous Field of View. This limits the amount of energy it can detect, in turn degrading the radiometric resolution. To increase the amount of energy, the wavelength range is increased, which decreases the spectral resolution. These instruments usually operate with one high spatial resolution band in Panchromatic, and several lower spatial resolution multi-spectral bands in Red, Green, Blue, Near-infrared.
On the other hand, for a high spectral resolution, the spatial resolution is decreased, the ground sampling distance is bigger so acquire enough energy. This is the case of spectrometers.
Radiometers have the highest radiometric resolution, the lower the spectral and spatial resolutions. In order to record enough energy, there needs to be a wider wavelength range and a bigger ground sampling distance (bigger instantaneous field of view).
For each instant of time where the instrument is acquiring information (sampling time), the instrument records a matrix of information in units of length^2. In the case of imagers or radiometers this matrix is in across-track (ACT) x along-track (ALT). For spectrometers, this matrix is in ACT x wavelength.
Some types of instruments, like the Fourier Transform Spectrometers with imaging capabilities record a 3 dimension cube of data for each sampling time (ACT x ALT x wavelength). The processing of 3D instead of 2D information adds an additional layer of complexity to the simulator. The number of missions that have this type of processing is less than 5%. Therefore it is not taken into account in the BIBLOS design at this stage. The BIBLOS design assumes a 2D matrix of information per sampling time.
The simulation chain for the instrument converts the input radiant flux into the digital numbers recorded in a binary image, and adds errors at different stages. There are two simulation flows depending on the nature of the instrument:
- Imagers and Radiometers record data in the ACT x ALT direction. Their simulation chain consists of the Spatial Block and the Radiometric Block .
- Spectrometers record data in the ACT x wavelength direction, so spatial and spectral effects are always coupled. Their simulation chain consists of the Spatial and Spectral Block for Spectrometers and Radiometric Block .
The list of Blocks for the Instrument Module is the following
- The Spatial Block does the spatial resampling and introduces the spatial aberration errors. The errors are modelled with a user-defined MTF.
- Spatial and Spectral Block for Spectrometers models the spatial and spectral coupled effects that occur in spectrometers. This is the deformation of the detector line due to the smile/keystone effect. Other spatial and spectral errors are modelled with a user-defined MTF.
- The Radiometric Block converts the radiance into digital numbers. It also introduces the radiometric errors Photonic Noise, PRNU, DSNU and defective pixels.
The simulation flow is the following: