Often, significant filtering requirements exist for planar and drop-in receiver filters at IF frequencies in the 1 to 4 GHz frequency range of typically 3 to 7x fIF (centre frequency) to comply with spurious emission requirements and suppression of spurious outputs. Filter requirements become even more challenging when a number of up- or down-converter stages are cascaded in advanced payloads where specifications for up to 10x fIF arise. Today, to achieve such very demanding requirements several low- and high-pass filters have to be cascaded in order to provide the in-band and out-of-band required performances (alternative solutions are based on a bandpass clean-up filter and a low pass filter).
Bandwidth, frequency and power reconfigurable payload architectures are currently under development by major satellite manufacturers. Today's fixed bandwidth filters are a major bottleneck in such systems and have led to the implementation of reconfigurable filters. These new reconfiguration scenarios require flexible converters and, hence, flexible filters to be accommodated after the down/up conversion
Increasingly complex coding and modulation schemes are being applied as means to enhance spectral efficiency. Coupled with the requirement for low-cost and environmentally friendly communications, the aspect of signal amplification becomes entangled with the actual spectral content of the system. It is now widely recognised that power amplification efficiency can only be optimised when the spectral characteristics of the signal to be processed are considered. To that end, fast and efficient methods to model signal distortion in non-linear circuits has emerged as a key priority. The aim of this project is to investigate methods and techniques for efficient modelling of signal distortion in PAs and focus on the development, improvement and experimental verification of one of the state-of-the-art methodologies, the Bessel-Fourier power amplifier, PA, behavioural model.