Show simple item record

dc.contributor.advisorCole-Rhodes, Arleneen
dc.contributor.advisorMoazzami, Farzaden
dc.contributor.authorTaiwo, Peter
dc.contributor.authorDossongui, Itie Serge Kone
dc.date.accessioned2017-06-20T16:17:09Z
dc.date.available2017-06-20T16:17:09Z
dc.date.issued2016-11
dc.identifier.issn0884-5123
dc.identifier.issn0074-9079
dc.identifier.urihttp://hdl.handle.net/10150/624260
dc.description.abstractWe consider the problem of equalizing data blocks of signals, which have been transmitted over an aeronautical channel using two different modulation schemes. The equalization is performed using the block-processing constant modulus algorithm (CMA), and in order to achieve real-time processing a Fast Fourier Transform (FFT) is used to compute the gradient of this cost function during equalization. The equalizer length is chosen to be five times of the channel length. For the first experiment, we present the result of equalizing a set of measured data, which was modulated and transmitted using the iNET packet structure with SOQPSK modulation. In this case, the CMA equalizer is first initialized using MMSE and the equalizer coefficients are then updated once, using each entire block (iNET packet). In the second experiment, we apply the FFT-based block processing equalizer to received data blocks of QPSK signals, which have been randomly generated and transmitted over an aeronautical channel. A modified constant modulus algorithm and alphabet matched algorithm (CMA + AMA) equalizer is used to recover these data blocks. For this case of QPSK signals, the equalizer performance is evaluated over 500 Monte Carlo runs, using the average symbol error rate (SER).
dc.description.sponsorshipInternational Foundation for Telemeteringen
dc.language.isoen_USen
dc.publisherInternational Foundation for Telemeteringen
dc.relation.urlhttp://www.telemetry.org/en
dc.rightsCopyright © held by the author; distribution rights International Foundation for Telemeteringen
dc.subjectBlind Equalizationen
dc.subjectConstant Modulus Algorithmen
dc.subjectAlphabet Matched Algorithmen
dc.subjectFFTen
dc.subjectChannel Estimationen
dc.subjectSOQPSKen
dc.subjectiNETen
dc.subjectAeronautic channelsen
dc.subjectMMSEen
dc.subjectEqualizersen
dc.titleTowards Real-Time CMA Equalization by using FFT for Signal Blocks transmitted over an Aeronautical channelen_US
dc.typetexten
dc.typeProceedingsen
dc.contributor.departmentMorgan State University, Dept Elect and Comp Engen
dc.identifier.journalInternational Telemetering Conference Proceedingsen
dc.description.collectioninformationProceedings from the International Telemetering Conference are made available by the International Foundation for Telemetering and the University of Arizona Libraries. Visit http://www.telemetry.org/index.php/contact-us if you have questions about items in this collection.en
refterms.dateFOA2018-06-06T04:11:57Z
html.description.abstractWe consider the problem of equalizing data blocks of signals, which have been transmitted over an aeronautical channel using two different modulation schemes. The equalization is performed using the block-processing constant modulus algorithm (CMA), and in order to achieve real-time processing a Fast Fourier Transform (FFT) is used to compute the gradient of this cost function during equalization. The equalizer length is chosen to be five times of the channel length. For the first experiment, we present the result of equalizing a set of measured data, which was modulated and transmitted using the iNET packet structure with SOQPSK modulation. In this case, the CMA equalizer is first initialized using MMSE and the equalizer coefficients are then updated once, using each entire block (iNET packet). In the second experiment, we apply the FFT-based block processing equalizer to received data blocks of QPSK signals, which have been randomly generated and transmitted over an aeronautical channel. A modified constant modulus algorithm and alphabet matched algorithm (CMA + AMA) equalizer is used to recover these data blocks. For this case of QPSK signals, the equalizer performance is evaluated over 500 Monte Carlo runs, using the average symbol error rate (SER).


Files in this item

Thumbnail
Name:
ITC_2016_16-10-03.pdf
Size:
1.039Mb
Format:
PDF
Description:
Conference Proceedings

This item appears in the following Collection(s)

Show simple item record