
Through‐line method for transmission/reflection estimation of N ‐way multiport power‐combining network
Author(s) -
Uchida Hiromitsu,
Nakashima Masao,
Handa Seiichi,
Sagara Takehiko,
Hariu Kenichi
Publication year - 2019
Publication title -
iet microwaves, antennas and propagation
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.555
H-Index - 69
eISSN - 1751-8733
pISSN - 1751-8725
DOI - 10.1049/iet-map.2018.5398
Subject(s) - transmission line , reactance , amplifier , electric power transmission , resistive touchscreen , impedance matching , reflection (computer programming) , characteristic impedance , image impedance , input impedance , electronic engineering , transmission (telecommunications) , electrical impedance , smith chart , line (geometry) , topology (electrical circuits) , power (physics) , coupling (piping) , scattering parameters , computer science , electrical engineering , engineering , mathematics , physics , telecommunications , voltage , cmos , geometry , quantum mechanics , damping factor , programming language , mechanical engineering
A transmission/reflection estimation method based on ‘through‐line’ measurements is proposed for N ‐way tree‐shaped power‐combining network, mainly used as an output impedance‐matching network of a multi‐transistor‐cell power amplifier (PA). The method can be applied when all the N branches can be seen identical to each other and are excited uniformly in‐phase, and especially useful when the input impedance of the network has a non‐resistive (reactance) component. In the proposed method, a pair of the N ‐way networks is connected directly in a back‐to‐back manner (‘through’ configuration) and with transmission lines in‐between (‘line’ configuration), and their S ‐parameters are measured. The transmission/reflection coefficients of single N ‐way network can be estimated from the measured ‘through’ and ‘line’ S ‐parameters, where in the latter the electromagnetic coupling between the transmission lines should be dealt with carefully to avoid estimation error.