Identification of non-linear RF systems using backpropagation

Andreas Toftegaard Kristensen; Andreas Burg; Alexios Balatsoukas-Stimming

doi:10.1109/ICCWorkshops49005.2020.9145367

Identification of non-linear RF systems using backpropagation

Andreas Toftegaard Kristensen, Andreas Burg, Alexios Balatsoukas-Stimming

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

10 Citations (Scopus)

Abstract

In this work, we use deep unfolding to view cascaded non-linear RF systems as model-based neural networks. This view enables the direct use of a wide range of neural network tools and optimizers to efficiently identify such cascaded models. We demonstrate the effectiveness of this approach through the example of digital self-interference cancellation in full-duplex communications where an IQ imbalance model and a non-linear PA model are cascaded in series. For a self-interference cancellation performance of approximately 44.5 dB, the number of model parameters can be reduced by 74 % and the number of operations per sample can be reduced by 79 % compared to an expanded linear-in-parameters polynomial model.

Original language	English
Title of host publication	2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers
Number of pages	6
ISBN (Electronic)	978-1-7281-7440-2
DOIs	https://doi.org/10.1109/ICCWorkshops49005.2020.9145367
Publication status	Published - Jun 2020
Event	2020 IEEE International Conference on Communications (ICC 2020) - Dublin, Ireland Duration: 7 Jun 2020 → 11 Jun 2020

Conference

Conference	2020 IEEE International Conference on Communications (ICC 2020)
Country/Territory	Ireland
City	Dublin
Period	7/06/20 → 11/06/20

Keywords

Backpropagation
Memory polynomial
Parallel Hammerstein model
SI cancellation
Volterra series

Access to Document

10.1109/ICCWorkshops49005.2020.9145367

Cite this

@inproceedings{1ace96e1657846c493a9b46f15a3c5e3,

title = "Identification of non-linear RF systems using backpropagation",

abstract = "In this work, we use deep unfolding to view cascaded non-linear RF systems as model-based neural networks. This view enables the direct use of a wide range of neural network tools and optimizers to efficiently identify such cascaded models. We demonstrate the effectiveness of this approach through the example of digital self-interference cancellation in full-duplex communications where an IQ imbalance model and a non-linear PA model are cascaded in series. For a self-interference cancellation performance of approximately 44.5 dB, the number of model parameters can be reduced by 74 % and the number of operations per sample can be reduced by 79 % compared to an expanded linear-in-parameters polynomial model.",

keywords = "Backpropagation, Memory polynomial, Parallel Hammerstein model, SI cancellation, Volterra series",

author = "Kristensen, {Andreas Toftegaard} and Andreas Burg and Alexios Balatsoukas-Stimming",

year = "2020",

month = jun,

doi = "10.1109/ICCWorkshops49005.2020.9145367",

language = "English",

booktitle = "2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2020 IEEE International Conference on Communications (ICC 2020) ; Conference date: 07-06-2020 Through 11-06-2020",

}

Kristensen, AT, Burg, A & Balatsoukas-Stimming, A 2020, Identification of non-linear RF systems using backpropagation. in 2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020 - Proceedings., 9145367, Institute of Electrical and Electronics Engineers, 2020 IEEE International Conference on Communications (ICC 2020), Dublin, Ireland, 7/06/20. https://doi.org/10.1109/ICCWorkshops49005.2020.9145367

Identification of non-linear RF systems using backpropagation. / Kristensen, Andreas Toftegaard; Burg, Andreas; Balatsoukas-Stimming, Alexios.
2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020 - Proceedings. Institute of Electrical and Electronics Engineers, 2020. 9145367.

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

TY - GEN

T1 - Identification of non-linear RF systems using backpropagation

AU - Kristensen, Andreas Toftegaard

AU - Burg, Andreas

AU - Balatsoukas-Stimming, Alexios

PY - 2020/6

Y1 - 2020/6

N2 - In this work, we use deep unfolding to view cascaded non-linear RF systems as model-based neural networks. This view enables the direct use of a wide range of neural network tools and optimizers to efficiently identify such cascaded models. We demonstrate the effectiveness of this approach through the example of digital self-interference cancellation in full-duplex communications where an IQ imbalance model and a non-linear PA model are cascaded in series. For a self-interference cancellation performance of approximately 44.5 dB, the number of model parameters can be reduced by 74 % and the number of operations per sample can be reduced by 79 % compared to an expanded linear-in-parameters polynomial model.

AB - In this work, we use deep unfolding to view cascaded non-linear RF systems as model-based neural networks. This view enables the direct use of a wide range of neural network tools and optimizers to efficiently identify such cascaded models. We demonstrate the effectiveness of this approach through the example of digital self-interference cancellation in full-duplex communications where an IQ imbalance model and a non-linear PA model are cascaded in series. For a self-interference cancellation performance of approximately 44.5 dB, the number of model parameters can be reduced by 74 % and the number of operations per sample can be reduced by 79 % compared to an expanded linear-in-parameters polynomial model.

KW - Backpropagation

KW - Memory polynomial

KW - Parallel Hammerstein model

KW - SI cancellation

KW - Volterra series

UR - http://www.scopus.com/inward/record.url?scp=85090280641&partnerID=8YFLogxK

U2 - 10.1109/ICCWorkshops49005.2020.9145367

DO - 10.1109/ICCWorkshops49005.2020.9145367

M3 - Conference contribution

AN - SCOPUS:85090280641

BT - 2020 IEEE International Conference on Communications Workshops, ICC Workshops 2020 - Proceedings

PB - Institute of Electrical and Electronics Engineers

T2 - 2020 IEEE International Conference on Communications (ICC 2020)

Y2 - 7 June 2020 through 11 June 2020

ER -

Identification of non-linear RF systems using backpropagation

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this