PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation

RJ Piechocki; C Andrieu; M Sandell; JP McGeehan

PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation

RJ Piechocki, C Andrieu, M Sandell, JP McGeehan

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

310 Downloads (Pure)

Abstract

In this paper we develop an efficient algorithm for inference in Factorial Hidden Markov Models (FHMM), which is particularly suitable for turbo equalisation in Multiple Input-Multiple Output (MIMO) systems. The proposed PDABCJR algorithm can be viewed as a generalisation of the PDA algorithm, which in its basic form handles single latent variables only. Our generalisation replaces each of the single latent variables with a HMM

Translated title of the contribution	PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation
Original language	English
Title of host publication	European Signal Processing Conference (EUSIPCO), Florence, Italy
Publisher	European Association for Signal Processing (EURASIP)
Publication status	Published - Sept 2006
Event	14th European Signal Processing Conference (EUSIPCO) - Florence, Italy Duration: 1 Sept 2006 → …

Conference

Conference	14th European Signal Processing Conference (EUSIPCO)
Country/Territory	Italy
City	Florence
Period	1/09/06 → …

Bibliographical note

Conference Proceedings/Title of Journal: EUSIPCO 2004: 14th European Signal Processing Conference
Rose publication type: Conference contribution

Additional information: With accompanying conference presentation

Sponsorship: The authors would like to thank C. Vithanage for helpful comments and Toshiba TRL Bristol UK for supporting this
work

Keywords

MIMO
turbo detection
variational inference
PDA
FHMM
HMM
BCJR

Access to Document

conference paperAccepted author manuscript, 99.7 KB
accompanying presentationAccepted author manuscript, 1.29 MB

Handle.net

Persistent link

Cite this

@inproceedings{44ec55a520f049a1b042db81f53ad672,

title = "PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation",

abstract = "In this paper we develop an efficient algorithm for inference in Factorial Hidden Markov Models (FHMM), which is particularly suitable for turbo equalisation in Multiple Input-Multiple Output (MIMO) systems. The proposed PDABCJR algorithm can be viewed as a generalisation of the PDA algorithm, which in its basic form handles single latent variables only. Our generalisation replaces each of the single latent variables with a HMM",

keywords = "MIMO, turbo detection, variational inference, PDA, FHMM, HMM, BCJR",

author = "RJ Piechocki and C Andrieu and M Sandell and JP McGeehan",

note = "Conference Proceedings/Title of Journal: EUSIPCO 2004: 14th European Signal Processing Conference Rose publication type: Conference contribution Additional information: With accompanying conference presentation Sponsorship: The authors would like to thank C. Vithanage for helpful comments and Toshiba TRL Bristol UK for supporting this work ; 14th European Signal Processing Conference (EUSIPCO) ; Conference date: 01-09-2006",

year = "2006",

month = sep,

language = "English",

booktitle = "European Signal Processing Conference (EUSIPCO), Florence, Italy",

publisher = "European Association for Signal Processing (EURASIP)",

}

Piechocki, RJ , Andrieu, C, Sandell, M & McGeehan, JP 2006, PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation. in European Signal Processing Conference (EUSIPCO), Florence, Italy. European Association for Signal Processing (EURASIP), 14th European Signal Processing Conference (EUSIPCO), Florence, Italy, 1/09/06. <http://hdl.handle.net/1983/849>

PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation. / Piechocki, RJ ; Andrieu, C; Sandell, M et al.
European Signal Processing Conference (EUSIPCO), Florence, Italy. European Association for Signal Processing (EURASIP), 2006.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation

AU - Piechocki, RJ

AU - Andrieu, C

AU - Sandell, M

AU - McGeehan, JP

N1 - Conference Proceedings/Title of Journal: EUSIPCO 2004: 14th European Signal Processing Conference Rose publication type: Conference contribution Additional information: With accompanying conference presentation Sponsorship: The authors would like to thank C. Vithanage for helpful comments and Toshiba TRL Bristol UK for supporting this work

PY - 2006/9

Y1 - 2006/9

N2 - In this paper we develop an efficient algorithm for inference in Factorial Hidden Markov Models (FHMM), which is particularly suitable for turbo equalisation in Multiple Input-Multiple Output (MIMO) systems. The proposed PDABCJR algorithm can be viewed as a generalisation of the PDA algorithm, which in its basic form handles single latent variables only. Our generalisation replaces each of the single latent variables with a HMM

AB - In this paper we develop an efficient algorithm for inference in Factorial Hidden Markov Models (FHMM), which is particularly suitable for turbo equalisation in Multiple Input-Multiple Output (MIMO) systems. The proposed PDABCJR algorithm can be viewed as a generalisation of the PDA algorithm, which in its basic form handles single latent variables only. Our generalisation replaces each of the single latent variables with a HMM

KW - MIMO

KW - turbo detection

KW - variational inference

KW - PDA

KW - FHMM

KW - HMM

KW - BCJR

M3 - Conference Contribution (Conference Proceeding)

BT - European Signal Processing Conference (EUSIPCO), Florence, Italy

PB - European Association for Signal Processing (EURASIP)

T2 - 14th European Signal Processing Conference (EUSIPCO)

Y2 - 1 September 2006

ER -

PDA-BCJR algorithm for factorial hidden Markov models with application to MIMO equalisation

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Handle.net

Fingerprint

Cite this