Using autoencoders for radio signal denoising

Ebtesam Almazrouei; Gabriele Gianini; Corrado Mio; Nawaf Almoosa; Ernesto Damiani

doi:10.1145/3345837.3355949

Using autoencoders for radio signal denoising

Ebtesam Almazrouei, Gabriele Gianini, Corrado Mio, Nawaf Almoosa, Ernesto Damiani

Università degli Studi di Milano

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

12 Scopus citations

Abstract

We investigated the use of a Deep Learning approach to radio signal de-noising. This data-driven approach has does not require explicit use of expert knowledge to set up the parameters of the denoising procedure and grants great flexibility across many channel conditions. The core component used in this work is a Convolutional De-noising AutoEncoder, known to be very effective in image processing. The key of our approach consists in transforming the radio signal into a representation suitable to the CDAE: we transform the time-domain signal into a 2D signal using the Short Time Fourier Transform. We report about the performance of the approach in preamble denoising across protocols of the IEEE 802.11 family, studied using simulation data. This approach could be used within a machine learning pipeline: the denoised data can be fed to a protocol classifier. A perspective advantage of using the AutoEncoders in that pipeline is that they can be co-trained with the downstream classifier, to optimize the classification accuracy.

Original language	British English
Title of host publication	Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks
Pages	11-17
Number of pages	7
ISBN (Electronic)	9781450369060
DOIs	https://doi.org/10.1145/3345837.3355949
State	Published - 25 Nov 2019
Event	15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks, Q2SWinet 2019 - Miami Beach, United States Duration: 25 Nov 2019 → 29 Nov 2019

Publication series

Name	Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks

Conference

Conference	15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks, Q2SWinet 2019
Country/Territory	United States
City	Miami Beach
Period	25/11/19 → 29/11/19

Keywords

AutoEncoders
Deep Learning
Radio Spectrum
Signal Denoising

Access to Document

10.1145/3345837.3355949

Cite this

Almazrouei, E., Gianini, G., Mio, C., Almoosa, N., & Damiani, E. (2019). Using autoencoders for radio signal denoising. In Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks (pp. 11-17). (Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks). https://doi.org/10.1145/3345837.3355949

Almazrouei, Ebtesam ; Gianini, Gabriele ; Mio, Corrado et al. / Using autoencoders for radio signal denoising. Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks. 2019. pp. 11-17 (Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks).

@inproceedings{8e2b97b82d1743b1960499479ae4f272,

title = "Using autoencoders for radio signal denoising",

abstract = "We investigated the use of a Deep Learning approach to radio signal de-noising. This data-driven approach has does not require explicit use of expert knowledge to set up the parameters of the denoising procedure and grants great flexibility across many channel conditions. The core component used in this work is a Convolutional De-noising AutoEncoder, known to be very effective in image processing. The key of our approach consists in transforming the radio signal into a representation suitable to the CDAE: we transform the time-domain signal into a 2D signal using the Short Time Fourier Transform. We report about the performance of the approach in preamble denoising across protocols of the IEEE 802.11 family, studied using simulation data. This approach could be used within a machine learning pipeline: the denoised data can be fed to a protocol classifier. A perspective advantage of using the AutoEncoders in that pipeline is that they can be co-trained with the downstream classifier, to optimize the classification accuracy.",

keywords = "AutoEncoders, Deep Learning, Radio Spectrum, Signal Denoising",

author = "Ebtesam Almazrouei and Gabriele Gianini and Corrado Mio and Nawaf Almoosa and Ernesto Damiani",

note = "Funding Information: The authors acknowledge the support of the ICT Fund at Khalifa University, Abu Dhabi (Project No. 88434000029). The work was partially funded by the EU H2020 Research Programme, within the projects Toreador (No. 688797) and Threat-Arrest (No. 786890). Publisher Copyright: {\textcopyright} 2019 Association for Computing Machinery.; 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks, Q2SWinet 2019 ; Conference date: 25-11-2019 Through 29-11-2019",

year = "2019",

month = nov,

day = "25",

doi = "10.1145/3345837.3355949",

language = "British English",

series = "Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks",

pages = "11--17",

booktitle = "Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks",

}

Almazrouei, E, Gianini, G, Mio, C , Almoosa, N & Damiani, E 2019, Using autoencoders for radio signal denoising. in Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks. Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks, pp. 11-17, 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks, Q2SWinet 2019, Miami Beach, United States, 25/11/19. https://doi.org/10.1145/3345837.3355949

Using autoencoders for radio signal denoising. / Almazrouei, Ebtesam; Gianini, Gabriele; Mio, Corrado et al.
Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks. 2019. p. 11-17 (Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks).

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

TY - GEN

T1 - Using autoencoders for radio signal denoising

AU - Almazrouei, Ebtesam

AU - Gianini, Gabriele

AU - Mio, Corrado

AU - Almoosa, Nawaf

AU - Damiani, Ernesto

N1 - Funding Information: The authors acknowledge the support of the ICT Fund at Khalifa University, Abu Dhabi (Project No. 88434000029). The work was partially funded by the EU H2020 Research Programme, within the projects Toreador (No. 688797) and Threat-Arrest (No. 786890). Publisher Copyright: © 2019 Association for Computing Machinery.

PY - 2019/11/25

Y1 - 2019/11/25

N2 - We investigated the use of a Deep Learning approach to radio signal de-noising. This data-driven approach has does not require explicit use of expert knowledge to set up the parameters of the denoising procedure and grants great flexibility across many channel conditions. The core component used in this work is a Convolutional De-noising AutoEncoder, known to be very effective in image processing. The key of our approach consists in transforming the radio signal into a representation suitable to the CDAE: we transform the time-domain signal into a 2D signal using the Short Time Fourier Transform. We report about the performance of the approach in preamble denoising across protocols of the IEEE 802.11 family, studied using simulation data. This approach could be used within a machine learning pipeline: the denoised data can be fed to a protocol classifier. A perspective advantage of using the AutoEncoders in that pipeline is that they can be co-trained with the downstream classifier, to optimize the classification accuracy.

AB - We investigated the use of a Deep Learning approach to radio signal de-noising. This data-driven approach has does not require explicit use of expert knowledge to set up the parameters of the denoising procedure and grants great flexibility across many channel conditions. The core component used in this work is a Convolutional De-noising AutoEncoder, known to be very effective in image processing. The key of our approach consists in transforming the radio signal into a representation suitable to the CDAE: we transform the time-domain signal into a 2D signal using the Short Time Fourier Transform. We report about the performance of the approach in preamble denoising across protocols of the IEEE 802.11 family, studied using simulation data. This approach could be used within a machine learning pipeline: the denoised data can be fed to a protocol classifier. A perspective advantage of using the AutoEncoders in that pipeline is that they can be co-trained with the downstream classifier, to optimize the classification accuracy.

KW - AutoEncoders

KW - Deep Learning

KW - Radio Spectrum

KW - Signal Denoising

UR - http://www.scopus.com/inward/record.url?scp=85077314019&partnerID=8YFLogxK

U2 - 10.1145/3345837.3355949

DO - 10.1145/3345837.3355949

M3 - Conference contribution

AN - SCOPUS:85077314019

T3 - Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks

SP - 11

EP - 17

BT - Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks

T2 - 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks, Q2SWinet 2019

Y2 - 25 November 2019 through 29 November 2019

ER -

Almazrouei E, Gianini G, Mio C , Almoosa N , Damiani E. Using autoencoders for radio signal denoising. In Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks. 2019. p. 11-17. (Q2SWinet 2019 - Proceedings of the 15th ACM International Symposium on QoS and Security for Wireless and Mobile Networks). doi: 10.1145/3345837.3355949

Using autoencoders for radio signal denoising

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this