Blind MIMO channel equalisation using matrix momentum natural gradient adaptation

G.H. Morison, T.S. Durrani

Research output: Contribution to conferencePaper

Abstract

In this paper the problem of single input multiple output (SIMO) and multiple input multiple output (MIMO) channel equalisation in a frequency selective environment is addressed using blind source separation techniques. Using blind techniques removes the requirement for a training sequence giving an increase in the spectral efficiency of the system. Direction of arrival (DOA) information is not required as the statistical independence of the signals is the only requirement for the separation and equalisation of the source signals. The natural gradient algorithm is embedded within the matrix momentum method to improve the convergence of the standard algorithm. The performance of the algorithm is demonstrated for DPSK signals.
LanguageEnglish
Pages1-9
Number of pages9
DOIs
Publication statusPublished - Sep 2003
EventIEE Colloquium on DSP enabled Radio - Edinburgh, United Kingdom
Duration: 22 Sep 200323 Sep 2003

Conference

ConferenceIEE Colloquium on DSP enabled Radio
CountryUnited Kingdom
CityEdinburgh
Period22/09/0323/09/03

Fingerprint

Momentum
Blind source separation
Direction of arrival

Keywords

  • blind MIMO channel
  • equalisation
  • matrix momentum
  • natural gradient
  • adaptation

Cite this

Morison, G. H., & Durrani, T. S. (2003). Blind MIMO channel equalisation using matrix momentum natural gradient adaptation. 1-9. Paper presented at IEE Colloquium on DSP enabled Radio, Edinburgh, United Kingdom. https://doi.org/10.1049/ic.2003.0301
Morison, G.H. ; Durrani, T.S. / Blind MIMO channel equalisation using matrix momentum natural gradient adaptation. Paper presented at IEE Colloquium on DSP enabled Radio, Edinburgh, United Kingdom.9 p.
@conference{c27ebf2067484a5baf28e5fdd932c186,
title = "Blind MIMO channel equalisation using matrix momentum natural gradient adaptation",
abstract = "In this paper the problem of single input multiple output (SIMO) and multiple input multiple output (MIMO) channel equalisation in a frequency selective environment is addressed using blind source separation techniques. Using blind techniques removes the requirement for a training sequence giving an increase in the spectral efficiency of the system. Direction of arrival (DOA) information is not required as the statistical independence of the signals is the only requirement for the separation and equalisation of the source signals. The natural gradient algorithm is embedded within the matrix momentum method to improve the convergence of the standard algorithm. The performance of the algorithm is demonstrated for DPSK signals.",
keywords = "blind MIMO channel, equalisation, matrix momentum, natural gradient, adaptation",
author = "G.H. Morison and T.S. Durrani",
year = "2003",
month = "9",
doi = "10.1049/ic.2003.0301",
language = "English",
pages = "1--9",
note = "IEE Colloquium on DSP enabled Radio ; Conference date: 22-09-2003 Through 23-09-2003",

}

Morison, GH & Durrani, TS 2003, 'Blind MIMO channel equalisation using matrix momentum natural gradient adaptation' Paper presented at IEE Colloquium on DSP enabled Radio, Edinburgh, United Kingdom, 22/09/03 - 23/09/03, pp. 1-9. https://doi.org/10.1049/ic.2003.0301

Blind MIMO channel equalisation using matrix momentum natural gradient adaptation. / Morison, G.H.; Durrani, T.S.

2003. 1-9 Paper presented at IEE Colloquium on DSP enabled Radio, Edinburgh, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Blind MIMO channel equalisation using matrix momentum natural gradient adaptation

AU - Morison, G.H.

AU - Durrani, T.S.

PY - 2003/9

Y1 - 2003/9

N2 - In this paper the problem of single input multiple output (SIMO) and multiple input multiple output (MIMO) channel equalisation in a frequency selective environment is addressed using blind source separation techniques. Using blind techniques removes the requirement for a training sequence giving an increase in the spectral efficiency of the system. Direction of arrival (DOA) information is not required as the statistical independence of the signals is the only requirement for the separation and equalisation of the source signals. The natural gradient algorithm is embedded within the matrix momentum method to improve the convergence of the standard algorithm. The performance of the algorithm is demonstrated for DPSK signals.

AB - In this paper the problem of single input multiple output (SIMO) and multiple input multiple output (MIMO) channel equalisation in a frequency selective environment is addressed using blind source separation techniques. Using blind techniques removes the requirement for a training sequence giving an increase in the spectral efficiency of the system. Direction of arrival (DOA) information is not required as the statistical independence of the signals is the only requirement for the separation and equalisation of the source signals. The natural gradient algorithm is embedded within the matrix momentum method to improve the convergence of the standard algorithm. The performance of the algorithm is demonstrated for DPSK signals.

KW - blind MIMO channel

KW - equalisation

KW - matrix momentum

KW - natural gradient

KW - adaptation

U2 - 10.1049/ic.2003.0301

DO - 10.1049/ic.2003.0301

M3 - Paper

SP - 1

EP - 9

ER -

Morison GH, Durrani TS. Blind MIMO channel equalisation using matrix momentum natural gradient adaptation. 2003. Paper presented at IEE Colloquium on DSP enabled Radio, Edinburgh, United Kingdom. https://doi.org/10.1049/ic.2003.0301