MMSE-based MDL method for accurate source number estimation

Lei Huang; Teng Long; Erke Mao; H. C. So

doi:10.1109/LSP.2009.2024785

MMSE-based MDL method for accurate source number estimation

Lei Huang^*, Teng Long, Erke Mao, H. C. So

^*Corresponding author for this work

School of Information and Electronics

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

33 Citations (Scopus)

Abstract

In civilian communication systems, the signature sequence of the desired signal in training phase is known to the receiver. In this letter, using the mutual information, we bridge the probability density function and minimum mean-square error (MMSE) between the observed data and training sequence of the desired signal, and then employ the MMSE to construct a minimum description length (MDL) criterion for accurate source enumeration. Numerical results demonstrate that the proposed method is superior to existing MDL methods in terms of detection performance particularly for small number of snapshots and/or source angular separation.

Original language	English
Article number	2024785
Pages (from-to)	798-801
Number of pages	4
Journal	IEEE Signal Processing Letters
Volume	16
Issue number	9
DOIs	https://doi.org/10.1109/LSP.2009.2024785
Publication status	Published - 2009

Keywords

Eigenvalue decomposition
Minimum description length
Sensor array processing
Source number estimation

Access to Document

10.1109/LSP.2009.2024785

Cite this

Huang, L., Long, T., Mao, E., & So, H. C. (2009). MMSE-based MDL method for accurate source number estimation. IEEE Signal Processing Letters, 16(9), 798-801. Article 2024785. https://doi.org/10.1109/LSP.2009.2024785

@article{5e783226fb174c8a831d50e03b83f1af,

title = "MMSE-based MDL method for accurate source number estimation",

abstract = "In civilian communication systems, the signature sequence of the desired signal in training phase is known to the receiver. In this letter, using the mutual information, we bridge the probability density function and minimum mean-square error (MMSE) between the observed data and training sequence of the desired signal, and then employ the MMSE to construct a minimum description length (MDL) criterion for accurate source enumeration. Numerical results demonstrate that the proposed method is superior to existing MDL methods in terms of detection performance particularly for small number of snapshots and/or source angular separation.",

keywords = "Eigenvalue decomposition, Minimum description length, Sensor array processing, Source number estimation",

author = "Lei Huang and Teng Long and Erke Mao and So, {H. C.}",

year = "2009",

doi = "10.1109/LSP.2009.2024785",

language = "English",

volume = "16",

pages = "798--801",

journal = "IEEE Signal Processing Letters",

issn = "1070-9908",

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

number = "9",

}

TY - JOUR

T1 - MMSE-based MDL method for accurate source number estimation

AU - Huang, Lei

AU - Long, Teng

AU - Mao, Erke

AU - So, H. C.

PY - 2009

Y1 - 2009

N2 - In civilian communication systems, the signature sequence of the desired signal in training phase is known to the receiver. In this letter, using the mutual information, we bridge the probability density function and minimum mean-square error (MMSE) between the observed data and training sequence of the desired signal, and then employ the MMSE to construct a minimum description length (MDL) criterion for accurate source enumeration. Numerical results demonstrate that the proposed method is superior to existing MDL methods in terms of detection performance particularly for small number of snapshots and/or source angular separation.

AB - In civilian communication systems, the signature sequence of the desired signal in training phase is known to the receiver. In this letter, using the mutual information, we bridge the probability density function and minimum mean-square error (MMSE) between the observed data and training sequence of the desired signal, and then employ the MMSE to construct a minimum description length (MDL) criterion for accurate source enumeration. Numerical results demonstrate that the proposed method is superior to existing MDL methods in terms of detection performance particularly for small number of snapshots and/or source angular separation.

KW - Eigenvalue decomposition

KW - Minimum description length

KW - Sensor array processing

KW - Source number estimation

UR - http://www.scopus.com/inward/record.url?scp=79951730524&partnerID=8YFLogxK

U2 - 10.1109/LSP.2009.2024785

DO - 10.1109/LSP.2009.2024785

M3 - Article

AN - SCOPUS:79951730524

SN - 1070-9908

VL - 16

SP - 798

EP - 801

JO - IEEE Signal Processing Letters

JF - IEEE Signal Processing Letters

IS - 9

M1 - 2024785

ER -

MMSE-based MDL method for accurate source number estimation

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this