Abstract
This research aims at providing a system design that reduces the mass and cost of spaceborne Synthetic Aperture Radar (SAR) missions by a factor of two compared to current (TecSAR - 300 kg, ~ £ 127 M) or planned (NovaSAR-S — 400 kg, ~ £ 50 M) mission. This would enable the cost of a SAR constellation to approach that of the current optical constellation such as Disaster Monitoring Constellation (DMC). This research has identified that the mission cost can be reduced significantly by: focusing on a narrow range of applications (forestry and disasters monitoring); ensuring the final design has a compact stowage volume, which facilitates a shared launch; and building the payload around available platforms, rather than the platform around the payload. The central idea of the research has been to operate the SAR at a low instantaneous power level—a practical proposition for a micro-satellite based SAR. The use of a simple parabolic reflector with a single horn at L-band means that a single, reliable and efficient Solid State Power Amplifier (SSPA) can be used to lower the overall system cost, and to minimise the impact on the spacecraft power system. A detailed analysis of basic pulsed (~ 5 - 10 % duty cycle) and Continuous Wave (CW) SAR (100 % duty cycle) payloads has shown their inability to fit directly into existing microsatellite buses without involving major changes, or employing more than one platform. To circumvent the problems of pulsed and CW techniques, two approaches have been formulated. The first shows that a CW SAR can be implemented in a mono-static way with a single antenna on a single platform. In this technique, the SAR works in an Interrupted CW (ICW) mode, but these interruptions introduce periodic gaps in the raw data. On processing, these gapped data result in artefacts in the reconstructed images. By applying data based statistical estimation techniques to “fill in the gaps” in the simulated raw SAR data, this research has shown the possibility of minimising the effects of these artefacts. However, once the same techniques are applied to the real SAR data (in this case derived from RADARSAT-1), the artefacts are shown to be problematic. Because of this the ICW SAR design technique it is—set aside. The second shows that an extended chirp mode pulsed (ECMP) SAR (~ 20 - 54 % duty cycle) can be designed with a lowered peak power level which enables a single SSPA to feed a parabolic Cassegrain antenna. The detailed analysis shows the feasibility of developing a microsatellite based SAR design at a comparable price to those of optical missions.