A computational framework for modeling time-frequency distributions

Cesar Aceros, Marisel Villafãne, Domingo Rodriguez

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

Abstract

This work presents a computational modeling framework (CMF) for the algorithm formulation of timefrequency tools used in the signal characterization of wireless channels. The work centers on two important original contributions: the development of the CMF itself, which is based on group harmonic analysis and signal algebra operator theory, and the use of Kronecker products algebra (KPA) as an implementation tool-aid in the algorithm mapping of the timefrequency distributions to multicore computational structures. Performance evaluation results are presented which describe the scalability and extensibility attributes the KPA-based algorithm formulations. Implementation performance speed ups are also presented with respect to a PC-based cluster architecture used as a preliminary computational structure. Finally, the work demonstrates how time-frequency methods are instrumental in mapping one-dimensional impulse response functions of wireless channel links in a MIMO system into two-dimensional energy distributions through the use of time-frequency tools.

Original languageEnglish
Title of host publicationInternational Conference on High Performance Computing Systems 2010, HPCS 2010
Pages71-77
Number of pages7
StatePublished - 2010
Externally publishedYes
Event2010 International Conference on High Performance Computing Systems, HPCS 2010 - Orlando, FL, United States
Duration: 12 Jul 201014 Jul 2010

Publication series

NameInternational Conference on High Performance Computing Systems 2010, HPCS 2010

Conference

Conference2010 International Conference on High Performance Computing Systems, HPCS 2010
Country/TerritoryUnited States
CityOrlando, FL
Period12/07/1014/07/10

Fingerprint

Dive into the research topics of 'A computational framework for modeling time-frequency distributions'. Together they form a unique fingerprint.

Cite this