ShodhGangotri Indian Research in Progress

ShodhGangotri: Repository of Indian Research in Progress >
Maharishi Markandeshwar University >
Department of Electronics & Communication Engineering >

Please use this identifier to cite or link to this item:

Title: An adaptive lineariser for high power amplifiers
Keywords: Communication engineering
High Power Amplifires
Researcher: Garg, Rajnish
Guide(s): Pal, Maninder
Registration Date: 23/07/2012
Abstract: The high power amplifiers (HPAs) for CDMA and OFDM telecommunication systems need to be highly linear. However in practice, the response of HPA is non-linear which introduces distortion in signals being amplified. This response can be made linear using various linearization techniques. In this thesis, an adaptive pre-distortion based newlinelinearization technique is introduced. It is based on adding pre-distortions into signals, newlinewhich will result in cancellation of non-linear distortions appearing in power amplifier. The characteristics of this non-linear adaptive lineariser are represented by an analogue polynomial. In this thesis, the coefficients of this polynomial are adjusted automatically in a manner that the output of lineariser when fed to an amplifier, the final output becomes error free and the whole system behaves like a linear amplifier. This helps reducing the distortion in power amplifiers, resulting in a better linearised output. In particularly, a perturbation based approximation method is used to adjust the coefficients. Perturbation based approximation method for adaptive lineariser is based on changing the coefficients newlinestep by step and checking the error; whether it is decreasing or increasing. Depending upon the amount of distortion the algorithm may take a few to a large number of cycles before the final output of the overall system (amplifier and lineariser) becomes distortion free. The advantage of this system over the existing (digital) ones is that it has a wider frequency range and lower power consumption. The disadvantage is that the system is slow in nature. In this research, an adaptive lineariser is made using matlab and the results are taken for different values of non-linearities and the optimum value of step size is calculated.
Language: English
Appears in Department:Department of Electronics & Communication Engineering

Files in This Item:

File Description SizeFormat
certificate.pdfAttached File1.43 MBAdobe PDFThumbnail
synopsis.pdf766.87 kBAdobe PDFThumbnail

Items in ShodhGangotri are available on open access mode, unless otherwise indicated.


Valid XHTML 1.0!
Copyright 2011-2012 INFLIBNET Centre, Infocity, Gandhinagar, Gujarat,INDIA  - Feedback or email at webmaster [Powered by DSpace] Disclaimer