Dual Input Outphasing Power Amplifier

Abstract

Dual Input Outphasing Power Amplifier: The Outphasing amplifier is an amplifier with high efficiency over a large dynamic range. This project will require the student to design a high efficiency amplifier that covers the 5G mobile radio frequency band. The student is expected to design, simulate using ADS/HFSS, build, test and verify the design. Improvements are also required to obtain the highest efficiency possible at 9dB output power backoff. Core Issue: Current day applications of Outphasing remain limited; the core issue being that multiple inputs are hard to implement due to Outphasing being impractical in real life scenarios. This is because the amplifier is extremely sensitive to the phase angle. Any change or variation in phase causes the efficiency of the amplifier to drastically decrease. It was necessary to find a solution to this issue; mainly because the Outphasing amplifier is a very good trade-off between linearity and efficiency. When implemented correctly; shows promising results in the 50-60% efficiency range. For modern day 5G networks, multiple inputs are quickly becoming the new standard and Outphasing has existed since the 50's. If this technology can be adapted to modern day 5G communications; it would prove beneficial for research and implementation purposes. Conduction: The study was conducted via the use of the Advanced Design System (ADS). The Outphasing amplifier was simulated and tested against the original specifications, a modification was made due to the issue of ADS not allowing 2 sources to converge. A Wilkinson power divider was implemented instead to simulate 2 sources. Observation: From the final simulation, it can be seen that the maximum PAE reached was 61% while the 6dB backoff is 30%. While the S21 (signal gain) is 15dB which is quite high for an amplifier; maintaining a S11 of -13.5 dB; which is very good. The S22 parameter sitting at -9.9 dB. Overall, the performance of the amplifier is acceptable for an Outphasing system. Conclusion: In conclusion the study that was conducted ended with decent performance, however a lack of EM simulation and physical proof of concept; leaves the results a bit ambiguous to whether it would perform in practice.