A unified approach to the restoration of lost samples in discrete-time signals

R.N.J. Veldhuis; A.J.E.M. Janssen

A unified approach to the restoration of lost samples in discrete-time signals

R.N.J. Veldhuis^*, A.J.E.M. Janssen

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

15 Downloads (Pure)

Abstract

We consider the problem of estimating lost sample values in discrete-time signals. The problem is treated as a linear minimum variance estimation problem, which, in principle, requires knowledge of the signal's autocorrelation coefficients. We show that, starting from this general principle, several sample restoration methods studied separately in the literature, can be derived. As particular examples we consider sample restoration methods for band-limited signals, multiple sinusoids, autoregressive processes and (quasi periodic) speech signals. The restoration method for multiple sinusoids is new, and has not, to our knowledge, been published before.

Original language	English
Title of host publication	Conference record : papers presented November 10-12, 1986, Pacific Grove, California
Subtitle of host publication	Twentieth Asilomar Conference on Signals, Systems & Computers
Editors	Delores M. Etter
Publisher	IEEE
Pages	551-558
Number of pages	8
ISBN (Print)	0818608161, 0818688165
Publication status	Published - 1 Jan 1987
Externally published	Yes
Event	20th Asilomar Conference on Signals, Systems & Computers 1986 - Pacific Grove, United States Duration: 10 Nov 1986 → 12 Nov 1986 Conference number: 20

Conference

Conference	20th Asilomar Conference on Signals, Systems & Computers 1986
Country/Territory	United States
City	Pacific Grove
Period	10/11/86 → 12/11/86

Access to Document

Veldhuis1987unifiedFinal published version, 791 KB

Cite this

@inproceedings{cbe6eb41a2dd465b86509c195b7aec26,

title = "A unified approach to the restoration of lost samples in discrete-time signals",

abstract = "We consider the problem of estimating lost sample values in discrete-time signals. The problem is treated as a linear minimum variance estimation problem, which, in principle, requires knowledge of the signal's autocorrelation coefficients. We show that, starting from this general principle, several sample restoration methods studied separately in the literature, can be derived. As particular examples we consider sample restoration methods for band-limited signals, multiple sinusoids, autoregressive processes and (quasi periodic) speech signals. The restoration method for multiple sinusoids is new, and has not, to our knowledge, been published before.",

author = "R.N.J. Veldhuis and A.J.E.M. Janssen",

year = "1987",

month = jan,

day = "1",

language = "English",

isbn = "0818608161",

pages = "551--558",

editor = "Etter, {Delores M.}",

booktitle = "Conference record : papers presented November 10-12, 1986, Pacific Grove, California",

publisher = "IEEE",

address = "United States",

note = "20th Asilomar Conference on Signals, Systems & Computers 1986 ; Conference date: 10-11-1986 Through 12-11-1986",

}

Veldhuis, RNJ & Janssen, AJEM 1987, A unified approach to the restoration of lost samples in discrete-time signals. in DM Etter (ed.), Conference record : papers presented November 10-12, 1986, Pacific Grove, California : Twentieth Asilomar Conference on Signals, Systems & Computers. IEEE, pp. 551-558, 20th Asilomar Conference on Signals, Systems & Computers 1986, Pacific Grove, California, United States, 10/11/86.

A unified approach to the restoration of lost samples in discrete-time signals. / Veldhuis, R.N.J.; Janssen, A.J.E.M.
Conference record : papers presented November 10-12, 1986, Pacific Grove, California : Twentieth Asilomar Conference on Signals, Systems & Computers. ed. / Delores M. Etter. IEEE, 1987. p. 551-558.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - A unified approach to the restoration of lost samples in discrete-time signals

AU - Veldhuis, R.N.J.

AU - Janssen, A.J.E.M.

N1 - Conference code: 20

PY - 1987/1/1

Y1 - 1987/1/1

N2 - We consider the problem of estimating lost sample values in discrete-time signals. The problem is treated as a linear minimum variance estimation problem, which, in principle, requires knowledge of the signal's autocorrelation coefficients. We show that, starting from this general principle, several sample restoration methods studied separately in the literature, can be derived. As particular examples we consider sample restoration methods for band-limited signals, multiple sinusoids, autoregressive processes and (quasi periodic) speech signals. The restoration method for multiple sinusoids is new, and has not, to our knowledge, been published before.

AB - We consider the problem of estimating lost sample values in discrete-time signals. The problem is treated as a linear minimum variance estimation problem, which, in principle, requires knowledge of the signal's autocorrelation coefficients. We show that, starting from this general principle, several sample restoration methods studied separately in the literature, can be derived. As particular examples we consider sample restoration methods for band-limited signals, multiple sinusoids, autoregressive processes and (quasi periodic) speech signals. The restoration method for multiple sinusoids is new, and has not, to our knowledge, been published before.

UR - http://www.scopus.com/inward/record.url?scp=0023171839&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0023171839

SN - 0818608161

SN - 0818688165

SP - 551

EP - 558

BT - Conference record : papers presented November 10-12, 1986, Pacific Grove, California

A2 - Etter, Delores M.

PB - IEEE

T2 - 20th Asilomar Conference on Signals, Systems & Computers 1986

Y2 - 10 November 1986 through 12 November 1986

ER -

A unified approach to the restoration of lost samples in discrete-time signals

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this