Properties of predictor based on relative neighborhood graph localized FIR filters

John Aasted Sørensen

doi:10.1109/ICASSP.1995.479713

Properties of predictor based on relative neighborhood graph localized FIR filters

John Aasted Sørensen

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

411 Downloads (Orbit)

Abstract

A time signal prediction algorithm based on relative neighborhood graph (RNG) localized FIR filters is defined. The RNG connects two nodes, of input space dimension D, if their lune does not contain any other node. The FIR filters associated with the nodes, are used for local approximation of the training vectors belonging to the lunes formed by the nodes. The predictor training is carried out by iteration through 3 stages: initialization of the RNG of the training signal by vector quantization, LS estimation of the FIR filters localized in the input space by RNG nodes and adaptation of the RNG nodes by equalizing the LS approximation error among the lunes formed by the nodes of the RNG. The training properties of the predictor is exemplified on a burst signal and characterized by the normalized mean square error (NMSE) and the mean valence of the RNG nodes through the adaptation

Original language	English
Title of host publication	Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing
Volume	Volume 5
Publisher	IEEE
Publication date	1995
Pages	3391-3394
ISBN (Print)	07-80-32431-5
DOIs	https://doi.org/10.1109/ICASSP.1995.479713
Publication status	Published - 1995
Event	1995 IEEE International Conference on Acoustics, Speech, and Signal Processing - Detroit, United States Duration: 8 May 1995 → 12 May 1995 Conference number: 20

Conference

Conference	1995 IEEE International Conference on Acoustics, Speech, and Signal Processing
Number	20
Country/Territory	United States
City	Detroit
Period	08/05/1995 → 12/05/1995

Bibliographical note

Copyright: 1995 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Access to Document

10.1109/ICASSP.1995.479713

S rensenFinal published version, 259 KB

OpenUrl availability

Full text

Cite this

@inproceedings{9088a945d88f417e9777d83329defc96,

title = "Properties of predictor based on relative neighborhood graph localized FIR filters",

abstract = "A time signal prediction algorithm based on relative neighborhood graph (RNG) localized FIR filters is defined. The RNG connects two nodes, of input space dimension D, if their lune does not contain any other node. The FIR filters associated with the nodes, are used for local approximation of the training vectors belonging to the lunes formed by the nodes. The predictor training is carried out by iteration through 3 stages: initialization of the RNG of the training signal by vector quantization, LS estimation of the FIR filters localized in the input space by RNG nodes and adaptation of the RNG nodes by equalizing the LS approximation error among the lunes formed by the nodes of the RNG. The training properties of the predictor is exemplified on a burst signal and characterized by the normalized mean square error (NMSE) and the mean valence of the RNG nodes through the adaptation",

author = "S{\o}rensen, {John Aasted}",

note = "Copyright: 1995 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 1995 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP '95 ; Conference date: 08-05-1995 Through 12-05-1995",

year = "1995",

doi = "10.1109/ICASSP.1995.479713",

language = "English",

isbn = "07-80-32431-5",

volume = "Volume 5",

pages = "3391--3394",

booktitle = "Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing",

publisher = "IEEE",

address = "United States",

}

Sørensen, JA 1995, Properties of predictor based on relative neighborhood graph localized FIR filters. in Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing. vol. Volume 5, IEEE, pp. 3391-3394, 1995 IEEE International Conference on Acoustics, Speech, and Signal Processing, Detroit, Michigan, United States, 08/05/1995. https://doi.org/10.1109/ICASSP.1995.479713

TY - GEN

T1 - Properties of predictor based on relative neighborhood graph localized FIR filters

AU - Sørensen, John Aasted

N1 - Conference code: 20

PY - 1995

Y1 - 1995

N2 - A time signal prediction algorithm based on relative neighborhood graph (RNG) localized FIR filters is defined. The RNG connects two nodes, of input space dimension D, if their lune does not contain any other node. The FIR filters associated with the nodes, are used for local approximation of the training vectors belonging to the lunes formed by the nodes. The predictor training is carried out by iteration through 3 stages: initialization of the RNG of the training signal by vector quantization, LS estimation of the FIR filters localized in the input space by RNG nodes and adaptation of the RNG nodes by equalizing the LS approximation error among the lunes formed by the nodes of the RNG. The training properties of the predictor is exemplified on a burst signal and characterized by the normalized mean square error (NMSE) and the mean valence of the RNG nodes through the adaptation

AB - A time signal prediction algorithm based on relative neighborhood graph (RNG) localized FIR filters is defined. The RNG connects two nodes, of input space dimension D, if their lune does not contain any other node. The FIR filters associated with the nodes, are used for local approximation of the training vectors belonging to the lunes formed by the nodes. The predictor training is carried out by iteration through 3 stages: initialization of the RNG of the training signal by vector quantization, LS estimation of the FIR filters localized in the input space by RNG nodes and adaptation of the RNG nodes by equalizing the LS approximation error among the lunes formed by the nodes of the RNG. The training properties of the predictor is exemplified on a burst signal and characterized by the normalized mean square error (NMSE) and the mean valence of the RNG nodes through the adaptation

U2 - 10.1109/ICASSP.1995.479713

DO - 10.1109/ICASSP.1995.479713

M3 - Article in proceedings

SN - 07-80-32431-5

VL - Volume 5

SP - 3391

EP - 3394

BT - Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing

PB - IEEE

T2 - 1995 IEEE International Conference on Acoustics, Speech, and Signal Processing

Y2 - 8 May 1995 through 12 May 1995

ER -

Properties of predictor based on relative neighborhood graph localized FIR filters

Abstract

Conference

Bibliographical note

Access to Document

OpenUrl availability

Fingerprint

Cite this