Article (Scientific journals)
Optimal fractional linear prediction with restricted memory
Skovranek, Tomas; Despotovic, Vladimir; Peric, Zoran
2019In IEEE Signal Processing Letters, 26 (5), p. 760-764
Peer Reviewed verified by ORBi
 

Files


Full Text
double.pdf
Author postprint (224.87 kB)
Request a copy

All documents in ORBilu are protected by a user license.

Send to



Details



Keywords :
Fractional calculus; Linear prediction; Restricted memory; Speech processing
Abstract :
[en] Linear prediction is extensively used in modeling, compression, coding, and generation of speech signal. Various formulations of linear prediction are available, both in time and frequency domain, which start from different assumptions but result in the same solution. In this letter, we propose a novel, generalized formulation of the optimal low-order linear prediction using the fractional (non-integer) derivatives. The proposed fractional derivative formulation allows for the definition of predictor with versatile behavior based on the order of fractional derivative. We derive the closed-form expressions of the optimal fractional linear predictor with restricted memory, and prove that the optimal first-order and the optimal second-order linear predictors are only its special cases. Furthermore, we empirically prove that the optimal order of fractional derivative can be approximated by the inverse of the predictor memory, and thus, it is a priori known. Therefore, the complexity is reduced by optimizing and transferring only one predictor coefficient, i.e., one parameter less in comparison to the second-order linear predictor, at the same level of performance.
Disciplines :
Computer science
Author, co-author :
Skovranek, Tomas;  Technical University of Kosice (TUKE) > BERG Faculty
Despotovic, Vladimir ;  University of Belgrade > Technical Faculty in Bor
Peric, Zoran;  University of Nis > Faculty of Electronic Engineering
External co-authors :
yes
Language :
English
Title :
Optimal fractional linear prediction with restricted memory
Publication date :
May 2019
Journal title :
IEEE Signal Processing Letters
ISSN :
1070-9908
Publisher :
Institute of Electrical and Electronics Engineers, United States
Volume :
26
Issue :
5
Pages :
760-764
Peer reviewed :
Peer Reviewed verified by ORBi
Funders :
Slovak Research and Development Agency; Slovak Grant Agency for Science; Ministry of Education, Science and Technological Development (Republic of Serbia); COST Action
Available on ORBilu :
since 22 October 2019

Statistics


Number of views
67 (7 by Unilu)
Number of downloads
3 (3 by Unilu)

Scopus citations®
 
10
Scopus citations®
without self-citations
7
WoS citations
 
9

Bibliography


Similar publications



Contact ORBilu