The Development Of A Novel Load-Line Translation Algorithm For Dynamic Power Control Of Power Amplifiers During Power Back-Off

Abstract

The need for highly reliable communication systems that are flexible and adaptable to the changing needs and requirements that occur during space exploration are essential. Output power and efficiency of the system's transmitter are two dynamic requirements that can be improved with an adaptable communication system. Bias adaptation updates the transmitter DC bias to optimize efficient use of power in the transmitter. Within a wireless communication system, the RF power amplifier has significant consequence over the efficiency and linearity of the overall system. Further concern of non-linearity is increased for the power amplifier because the amplifier's best output power and efficiency performance occurs at points when the amplifier is exhibiting non-linear characteristics. These non-linear characteristics are produced by the amplifier entering a saturation region; as a result gain compression and signal distortion become noticed. Furthermore, with the integration of a bias adaption scheme into the amplifier design which uses voltage variation, there is also the additional concern of amplitude modulation to phase modulation distortion. A bias adaption approach was implemented that provides the amplifier with the ability to obtain better efficiency, over a static Class-A bias case, without additional degradation of the linearity performance. The approach also provides further insight into the non-linear behaviors of the amplifier and the measured results are in good agreement with the simulated linearity results provided via Agilent's Advance Design System design software. The Cripps' Class-A Load-line method serves as the foundation for this analysis. The approach translates the original Class-A Load-line to compensate for the change in the input drive level. A load-line bisector was calculated to assist in the change of DC bias proportionately with the change in output RF power. As a result, the adaptable power amplifier showed an average improvement of 69% in power added efficiency (PAE) compared to a static Class-A bias over power levels below the 1dB compression point (P1dB). The Carrier to Interference Ratio (C/I) increased by a maximum 7dB over the static Class-A bias case. Furthermore, the linear variation of DC bias, guided by the load-line bisector, resulted in the linear change of the P1dB . The efficiency performance of the adaptable power amplifier showed and average improvement of 70% when driven by BPSK and QPSK modulated input signals.