Blind spatial signature estimation via time-varying user power loading and parallel factor analysis

Yue Rong; Sergiy A. Vorobyov; Alex B. Gershman; Nicholas D. Sidiropoulos

doi:10.1109/TSP.2005.845441

Blind spatial signature estimation via time-varying user power loading and parallel factor analysis

Yue Rong, Sergiy A. Vorobyov, Alex B. Gershman, Nicholas D. Sidiropoulos

Research output: Contribution to journal › Article › peer-review

63 Scopus citations

Abstract

In this paper, the problem of blind spatial signature estimation using the parallel factor (PARAFAC) analysis model is addressed in application to wireless communications. A time-varying user power loading in the uplink mode is proposed to make the model identifiable and to enable application of PARAFAC analysis. Then, identifiability issues are studied in detail and closed-form expressions for the corresponding modified Cramáer-Rao bound (CRB) are obtained. Furthermore, two blind spatial signature estimation algorithms are developed. The first technique is based on the PARAFAC fitting trilinear alternating least squares (TALS) regression procedure, whereas the second one makes use of the joint approximate diagonalization algorithm. These techniques do not require any knowledge of the propagation channel and/or sensor array manifold and are applicable to a more general class of scenarios than earlier approaches to blind spatial signature estimation.

Original language	English (US)
Pages (from-to)	1697-1710
Number of pages	14
Journal	IEEE Transactions on Signal Processing
Volume	53
Issue number	5
DOIs	https://doi.org/10.1109/TSP.2005.845441
State	Published - May 2005
Externally published	Yes

Bibliographical note

Funding Information:
Manuscript received July 21, 2003; revised March 25, 2004. The work of A. B. Gershman was supported by the Wolfgang Paul Award Program of the Alexander von Humboldt Foundation, Germany; the Natural Sciences and Engineering Research Council (NSERC) of Canada; Communications and Information Technology Ontario (CITO); and the Premier’s Research Excellence Award Program of the Ministry of Energy, Science, and Technology (MEST) of Ontario. The work of N. D. Sidiropoulos was supported by the Army Research Laboratory through participation in the ARL Collaborative Technology Alliance (ARL-CTA) for Communications and Networks under Cooperative Agreement DADD19-01-2-0011. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Constantinos B. Papadias.

Keywords

Blind spatial signature estimation
Parallel factor analysis
Sensor array processing

Access

10.1109/TSP.2005.845441

OpenUrl availability

Full text

Cite this

@article{02a86d53a5914e1ab7e3230a2406d99d,

title = "Blind spatial signature estimation via time-varying user power loading and parallel factor analysis",

abstract = "In this paper, the problem of blind spatial signature estimation using the parallel factor (PARAFAC) analysis model is addressed in application to wireless communications. A time-varying user power loading in the uplink mode is proposed to make the model identifiable and to enable application of PARAFAC analysis. Then, identifiability issues are studied in detail and closed-form expressions for the corresponding modified Cram{\'a}er-Rao bound (CRB) are obtained. Furthermore, two blind spatial signature estimation algorithms are developed. The first technique is based on the PARAFAC fitting trilinear alternating least squares (TALS) regression procedure, whereas the second one makes use of the joint approximate diagonalization algorithm. These techniques do not require any knowledge of the propagation channel and/or sensor array manifold and are applicable to a more general class of scenarios than earlier approaches to blind spatial signature estimation.",

keywords = "Blind spatial signature estimation, Parallel factor analysis, Sensor array processing",

author = "Yue Rong and Vorobyov, {Sergiy A.} and Gershman, {Alex B.} and Sidiropoulos, {Nicholas D.}",

note = "Funding Information: Manuscript received July 21, 2003; revised March 25, 2004. The work of A. B. Gershman was supported by the Wolfgang Paul Award Program of the Alexander von Humboldt Foundation, Germany; the Natural Sciences and Engineering Research Council (NSERC) of Canada; Communications and Information Technology Ontario (CITO); and the Premier{\textquoteright}s Research Excellence Award Program of the Ministry of Energy, Science, and Technology (MEST) of Ontario. The work of N. D. Sidiropoulos was supported by the Army Research Laboratory through participation in the ARL Collaborative Technology Alliance (ARL-CTA) for Communications and Networks under Cooperative Agreement DADD19-01-2-0011. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Constantinos B. Papadias.",

year = "2005",

month = may,

doi = "10.1109/TSP.2005.845441",

language = "English (US)",

volume = "53",

pages = "1697--1710",

journal = "IEEE Transactions on Signal Processing",

issn = "1053-587X",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "5",

}

TY - JOUR

T1 - Blind spatial signature estimation via time-varying user power loading and parallel factor analysis

AU - Rong, Yue

AU - Vorobyov, Sergiy A.

AU - Gershman, Alex B.

AU - Sidiropoulos, Nicholas D.

N1 - Funding Information: Manuscript received July 21, 2003; revised March 25, 2004. The work of A. B. Gershman was supported by the Wolfgang Paul Award Program of the Alexander von Humboldt Foundation, Germany; the Natural Sciences and Engineering Research Council (NSERC) of Canada; Communications and Information Technology Ontario (CITO); and the Premier’s Research Excellence Award Program of the Ministry of Energy, Science, and Technology (MEST) of Ontario. The work of N. D. Sidiropoulos was supported by the Army Research Laboratory through participation in the ARL Collaborative Technology Alliance (ARL-CTA) for Communications and Networks under Cooperative Agreement DADD19-01-2-0011. The associate editor coordinating the review of this manuscript and approving it for publication was Dr. Constantinos B. Papadias.

PY - 2005/5

Y1 - 2005/5

N2 - In this paper, the problem of blind spatial signature estimation using the parallel factor (PARAFAC) analysis model is addressed in application to wireless communications. A time-varying user power loading in the uplink mode is proposed to make the model identifiable and to enable application of PARAFAC analysis. Then, identifiability issues are studied in detail and closed-form expressions for the corresponding modified Cramáer-Rao bound (CRB) are obtained. Furthermore, two blind spatial signature estimation algorithms are developed. The first technique is based on the PARAFAC fitting trilinear alternating least squares (TALS) regression procedure, whereas the second one makes use of the joint approximate diagonalization algorithm. These techniques do not require any knowledge of the propagation channel and/or sensor array manifold and are applicable to a more general class of scenarios than earlier approaches to blind spatial signature estimation.

AB - In this paper, the problem of blind spatial signature estimation using the parallel factor (PARAFAC) analysis model is addressed in application to wireless communications. A time-varying user power loading in the uplink mode is proposed to make the model identifiable and to enable application of PARAFAC analysis. Then, identifiability issues are studied in detail and closed-form expressions for the corresponding modified Cramáer-Rao bound (CRB) are obtained. Furthermore, two blind spatial signature estimation algorithms are developed. The first technique is based on the PARAFAC fitting trilinear alternating least squares (TALS) regression procedure, whereas the second one makes use of the joint approximate diagonalization algorithm. These techniques do not require any knowledge of the propagation channel and/or sensor array manifold and are applicable to a more general class of scenarios than earlier approaches to blind spatial signature estimation.

KW - Blind spatial signature estimation

KW - Parallel factor analysis

KW - Sensor array processing

UR - http://www.scopus.com/inward/record.url?scp=18844461688&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18844461688&partnerID=8YFLogxK

U2 - 10.1109/TSP.2005.845441

DO - 10.1109/TSP.2005.845441

M3 - Article

AN - SCOPUS:18844461688

SN - 1053-587X

VL - 53

SP - 1697

EP - 1710

JO - IEEE Transactions on Signal Processing

JF - IEEE Transactions on Signal Processing

IS - 5

ER -

Blind spatial signature estimation via time-varying user power loading and parallel factor analysis

Abstract

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this