Experimental investigations and behavior modeling for monolithic quasi-class E SiGe PA linearization

Yan Li, Jerry Lopez, Donald Y. Lie, Jeremy D. Popp

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

We have developed a modified bias-dependent Cann's model and performed IC design and hardware experiments to study the linearization of a highly-efficient monolithic quasi-class E SiGe power amplifier (PA) IC using both Envelope-Tracking (ET) and Envelope-Elimination-and-Restoration (EER) techniques. Our simple PA behavior model fits the measured SiGe PA IC data very well across a wide range of bias and supply voltages. Both measurement and simulations show that the ET-linearized PA system is significantly less sensitive to the timing misalignment between the amplitude and the RF signal path than the EER-linearized PA system. Our experimental results also show that ET successfully linearized the SiGe PA to pass the stringent EDGE transmit mask at 900MHz, while EER could not. Simulations also predict that the optimal timing alignment for ET linearization can be achieved at PA base bias voltage V bb=0.55-6V, which is consistent with our measurement results as well.

Original languageEnglish
Title of host publication2008 International Conference on Communications, Circuits and Systems Proceedings, ICCCAS 2008
Pages1337-1341
Number of pages5
DOIs
StatePublished - 2008
Event2008 International Conference on Communications, Circuits and Systems, ICCCAS 2008 - Xiamen, Fujian Province, China
Duration: May 25 2008May 27 2008

Publication series

Name2008 International Conference on Communications, Circuits and Systems Proceedings, ICCCAS 2008

Conference

Conference2008 International Conference on Communications, Circuits and Systems, ICCCAS 2008
Country/TerritoryChina
CityXiamen, Fujian Province
Period05/25/0805/27/08

Fingerprint

Dive into the research topics of 'Experimental investigations and behavior modeling for monolithic quasi-class E SiGe PA linearization'. Together they form a unique fingerprint.

Cite this