2.11.4 External Characterisation

ASAR can be put into External Characterisation Mode while flying over a calibration transponder. This involves sending a series of pulses from each of the 32 rows in turn followed by each of the 10 columns in turn.
These pulses are detected both by the internal calibration loop and the receiver embedded in the transponder. Comparison of these data allows characterising the passive part of the antenna and the calibration network. The baseline is to repeat measurement every six months.

2.11.4.1 Characterisation of the Antenna Beam Pattern

For the characterisation of the antenna beam pattern, images of the Amazon Rain Forest are used. This is because the rai forest it is a stable, large-scale, isotropic distributed target with a relatively high backscatter and a well-understood relationship between backscatter and incidence angle.

In order to determine the two-way beam pattern, an uncorrected rain forest image is averaged in the azimuth direction. In the final processed image, the inverted beam pattern is applied and hence the effect of the pattern on the backscatter is removed.

Alternative distributed targets at different latitudes are being investigated. Promising results have been found from ERS data over Lake Vostok in Antarctica. Antenna pattern estimates at different latitudes could be used to verify the round-orbit performance of the ASAR.

2.11.4.2 Gain Calibration

The purpose of the ASAR gain calibration is to provide the users of ASAR data with the possibility to determine the absolute level of backscatter from any target, point (s) or distributed (s0). For ASAR this is achieved in fundamentally the same way as for ERS, namely by providing an Absolute gain Calibration Factor (ACF) in the header of the (processed) product. Since ASAR, however, has a total of eight beams and five different modes and up to four polarisations more ACFs will need to be determined for ASAR than for the ERS single beam, single polarisation with two modes.

The method to be used to determine the ACFs is to image a target of known radar cross-section, integrate the power in its Impulse Response Function (IRF) corrected for the associated background (clutter) power and hence calculate the correction (the ACF) which must be applied to the image values in order to arrive at the same cross-section for that target. For this purpose, precision calibration transponders are deployed in the Netherlands. The radar cross-section of these transponders is 65dBm2 and is known to within ±0.13dB and they are stable to 0.08dB Ref. [2.14 ] . It is necessary to use active radar calibrators (transponders) as opposed to passive ones (e.g. corner reflectors) since the ratio of signal to clutter determines the accuracy to which the calibration can be made.

Once the ACF for a particular configuration has been calculated it will be possible to make a direct comparison with the on-ground measurements of the end-to-end system gain carried out during FM testing.

2.11.4.3 Global Monitoring Mode

This is a special case since the spatial resolution of 10001000m makes the normal use of the transponders unfeasible. For a reasonable calibration to be made (3s value of ± 0.5dB), a signal to clutter ratio of better than 30dB is required. If the clutter at the calibration sites typically has a sigma nought of -6dB then this would require a transponder RCS greater than 84dBm2. This would inevitably saturate the receiver invalidating the calibration. As a result, it is necessary to come up with an alternative scenario for calibrating this mode.

The first option (baseline) is using the other modes (namely Wide Swath and Image) to calibrate GM mode by means of the Amazonian rain forest. Since the s0 of the rain forest is stable to within 0.3dB it will be possible to use the s0 value obtained from a previous (or subsequent) pass in WS or IM to calibrate GM mode. In addition, other relatively stable distributed targets may be used such as the ice caps and specific desert regions (Gibson, Gobi etc).

The second option will allow direct calibration using a special global monitoring mode setting and a modified calibration transponder. The intention is to use the ASAR's digital chirp generator to offset the centre frequency by 5MHz. This is possible since the chirp bandwidth in GM mode is only around 1MHz. In the calibration transponder, the received signal is shifted back by 5MHz allowing it to be received by the ASAR. As the clutter return will all be outside the range of the reduced bandwidth filter in GM mode, only the transponder response will be seen in the processed image against a background of noise. The result of this operation is to provide a transponder signal to clutter ratio of between 25 and 30dB allowing for reliable calibration of Global Monitoring Mode.