This dissertation focuses on the design and implementation of an X-band receiver for use in the South African Synthetic Aperture Radar (SASAR II) project. The SAR will be used to demonstrate the capability of building a high resolution X-band imaging radar in South Africa.
The design starts by investigating the maximum power return from different targets over a swath width with changing incidence angles. A receiver-power-level table and diagram were constructed, with the power return from a trihedral corner reflector as maximum input power and thermal noise as the minimum input power to the receiver. The output of the receiver, which has to be fed to the input of an analogue-to-digital converter (ADC), is limited by the ADC’s maximum operating input power. Amplifiers, attenuators and mixers were chosen to implement a dual-stage downconversion from a radio frequency (RF) of 9300 MHz to a 2nd IF of 1300 MHz and then to a 1st IF of 158 MHz. A sensitivity time control (STC) is implemented in the receiver to cater for the limited dynamic range of the ADC. The power return varies with range and hence, time. Thus, an STC will correct for low return power, at far range, by boosting the received signal and attenuating large return power, at close range, ideally providing a fairly constant power return at the receiver output. A manual gain control (MGC) is also needed in the receiver, such that none of the components are driven into saturation. The gain control is switched on when large targets are expected to fall in the swath width, otherwise it is switched to a minimum for targets with low backscattered power.
The tests that were carried out on the receiver components showed that all the components operated very close to their specifications. The cascaded filters work well in tailoring the front-end 3-dB bandwidth to close to the required 3-dB bandwidth. The receiver was designed to have enough gain to boost the maximum power received to within the operating range of the ADC, without saturating any components in the receiver. The noise figure test showed a noise figure of 4:20 dB. This is 1.73 dB higher than the calculated noise figure of 2.47 dB which is a result of an underestimation of the losses in the system.