Abstract
This letter presents an automatic power amplifier (PA) design methodology that uses a multidimensional search algorithm to find the best compromise between fundamental and harmonic impedance terminations for a specified bandwidth. In conventional design methodologies, PA designers need to preselect the optimum impedances, which normally follow non-Foster trajectories, and are thus impossible to achieve with passive matching networks (MNs) for a wide frequency range. Instead, the proposed automatic design method directly synthesizes the MNs to achieve the desired output power, efficiency, and gain performance, without forcing any impedance profiles. For that, load-pull data are interpolated using artificial neural networks and an algorithm based on the simplified real frequency technique is used to obtain the MNs. Finally, the method is validated with a single-ended PA implementation.
Highlights
Automatic Methodology for Wideband Power Amplifier DesignCatarina Belchior , Graduate Student Member, IEEE, Luís C
I N WIRELESS communication systems, RF power amplifiers (PAs) assume an essential role and should be designed to improve specific performance metrics such as efficiency, output power (Pout), gain, and bandwidth
Since fundamental and second-harmonic load terminations (L1H and L2H, respectively) have a direct impact on the active device efficiency and output power contours, and the fundamental source and load terminations (S1H and L1H, respectively) will impact the small-signal gain, we divided this mapping into two separate functions in the following equations: Eff3 dB, Pout3 dB = FANN1(L1H, L2H )
Summary
Catarina Belchior , Graduate Student Member, IEEE, Luís C. Abstract— This letter presents an automatic power amplifier (PA) design methodology that uses a multidimensional search algorithm to find the best compromise between fundamental and harmonic impedance terminations for a specified bandwidth. PA designers need to preselect the optimum impedances, which normally follow nonFoster trajectories, and are impossible to achieve with passive matching networks (MNs) for a wide frequency range. The proposed automatic design method directly synthesizes the MNs to achieve the desired output power, efficiency, and gain performance, without forcing any impedance profiles. Load-pull data are interpolated using artificial neural networks and an algorithm based on the simplified real frequency technique is used to obtain the MNs. the method is validated with a single-ended PA implementation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
