Open Access System for Information Sharing

Login Library

 

Thesis
Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

기지국용 RF 전력 증폭기의 효율 개선에 관한 연구

Title
기지국용 RF 전력 증폭기의 효율 개선에 관한 연구
Authors
김정준
Date Issued
2013
Publisher
포항공과대학교
Abstract
The main theme of this thesis is efficiency enhancements of RF power transmitters. Because a large portion of total power consumption in wireless communication system occurs at the power amplifier (PA) stage, an efficient PA is crucial. High efficiency guarantees lower electric power consumption, longer life time of the device and easier thermal management. To improve the efficiency, the various techniques are studied in this dissertation. In particular, the efficiency improvement methods for the unit cell PA and the transmitters such as the Doherty amplifier and envelope tracking (ET) technique are considered. To maximize the efficiency for a high frequency PA, a novel design method based on the optimally saturated PA operation is proposed. As a PA is driven into saturated operation, the current source of the device generates a large third harmonic current, which creates a quasi-rectangular current waveform.The large nonlinear output capacitor of the transistor generates a second harmonic voltage with a very small third harmonic component. The second harmonic voltage is in-phase with the fundamental voltage, making a half-sine wave voltage waveform with voltage peaking. These waveforms are similar to those of a class F−1. The fundamental load at the intrinsic device is resistive with the output capacitance tuned out, which is identical to the class F−1 case. However, the required harmonic impedances are just larger than theimpedance levels of the output capacitance for both the second and third harmonics. Therefore, the circuit topology is similar to that of a class E amplifier, which is very simple. The PA implemented using a GaN HEMT delivers the expected good performance using the simple circuit topology. To enhance the efficiency at an average output power level of a modulated signal, among the various efficiency enhancement techniques, the Doherty amplifier is the most preferred structure. For base station applications, -3 dB hybrid coupler has been commonly used for input power dividing stage of Doherty power amplifier (PA). The input direct dividing technique has only been applied for handset applications, because the coupler occupies a large size on a chip. However, this technique provides advantages on power gain and efficiency for base station applications, also. In this dissertation, we have proposed a design procedure of Doherty PA applying the input direct dividing technique. Implemented Doherty PA using two Cree CGH40045 GaN HEMTachieves excellent performances. Bias modulated PA is also one of the good candidate to enhance the efficiency at an average output power level. In the dissertation, we have proposed two methods of enhancing efficiency of an ET PA from an interlock operation. The first is the utilization of sinking current. The sinking current is a criticalefficiency reduction factor since it is a wasted power. To reduce the sinking current, the gate bias of the PA is increased, so that the sinking current is delivered to the PA and is utilized for amplification. The other one is the RF input shaping method. The input signal of ET PAs is a modulated RF signal, and the signal doesnït guarantee fully saturated operation of the PA at all power levels due to gm nonlinearity of a device. To obtain the maximum efficiency for all of the envelope voltage, we have found the optimum RF inputconditions and applied it to the input of the PA. To verify the methods, the proposed ET PA is implemented using a Cree CGH40045 GaN HEMT. By combining the advantages of both techniques, the proposed ET PA for high efficiency is experimentally verified. Additionally, this dissertation includes the design of supply modulator using integrated circuit. For the base-station applications, most of the bias modulators have been designed by the discrete components, thus the each blocks of the bias modulator cannot be optimized for their purposes. To solvethe problems, we have desiged the bias modulator using integrated circuit. Moreover, it is very easy to apply the additional circuits which improve the efficiency. The circuit configurations and the simulation results are presentedin this dissertation.
URI
http://postech.dcollection.net/jsp/common/DcLoOrgPer.jsp?sItemId=000001554599
https://oasis.postech.ac.kr/handle/2014.oak/1739
Article Type
Thesis
Files in This Item:
There are no files associated with this item.

qr_code

  • mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

Views & Downloads

Browse