A New Classification-Like Scheme for Spectrum Sensing Using Spectral Correlation and Stacked Denoising Autoencoders

Hang LIU; Xu ZHU; Takeo FUJII

doi:10.1587/transcom.2017EBP3447

A New Classification-Like Scheme for Spectrum Sensing Using Spectral Correlation and Stacked Denoising Autoencoders

Hang LIU, Xu ZHU, Takeo FUJII

Full Text Views

0

Cite this

Summary :

In this paper, we propose a novel primary user detection scheme for spectrum sensing in cognitive radio. Inspired by the conventional signal classification approach, the spectrum sensing is translated into a classification problem. On the basis of feature-based classification, the spectral correlation of a second-order cyclostationary analysis is applied as the feature extraction method, whereas a stacked denoising autoencoders network is applied as the classifier. Two training methods for signal detection, interception-based detection and simulation-based detection, are considered, for different prior information and implementation conditions. In an interception-based detection method, inspired by the two-step sensing, we obtain training data from the interception of actual signals after a sophisticated sensing procedure, to achieve detection without priori information. In addition, benefiting from practical training data, this interception-based detection is superior under actual transmission environment conditions. The alternative, a simulation-based detection method utilizes some undisguised parameters of the primary user in the spectrum of interest. Owing to the diversified predetermined training data, simulation-based detection exhibits transcendental robustness against harsh noise environments, although it demands a more complicated classifier network structure. Additionally, for the above-described training methods, we discuss the classifier complexity over implementation conditions and the trade-off between robustness and detection performance. The simulation results show the advantages of the proposed method over conventional spectrum-sensing schemes.

Publication: IEICE TRANSACTIONS on Communications Vol.E101-B No.11 pp.2348-2361

Publication Date: 2018/11/01

Publicized: 2018/04/25

Online ISSN: 1745-1345

DOI: 10.1587/transcom.2017EBP3447

Type of Manuscript: PAPER

Category: Terrestrial Wireless Communication/Broadcasting Technologies

Authors

Hang LIU
  The University of Electro-Communications
Xu ZHU
  The University of Electro-Communications
Takeo FUJII
  The University of Electro-Communications

Keyword

spectrum sensing, cyclostationary analysis, stacked denoising autoencoders

Cite this

Copy

Hang LIU, Xu ZHU, Takeo FUJII, "A New Classification-Like Scheme for Spectrum Sensing Using Spectral Correlation and Stacked Denoising Autoencoders" in IEICE TRANSACTIONS on Communications, vol. E101-B, no. 11, pp. 2348-2361, November 2018, doi: 10.1587/transcom.2017EBP3447.
Abstract: In this paper, we propose a novel primary user detection scheme for spectrum sensing in cognitive radio. Inspired by the conventional signal classification approach, the spectrum sensing is translated into a classification problem. On the basis of feature-based classification, the spectral correlation of a second-order cyclostationary analysis is applied as the feature extraction method, whereas a stacked denoising autoencoders network is applied as the classifier. Two training methods for signal detection, interception-based detection and simulation-based detection, are considered, for different prior information and implementation conditions. In an interception-based detection method, inspired by the two-step sensing, we obtain training data from the interception of actual signals after a sophisticated sensing procedure, to achieve detection without priori information. In addition, benefiting from practical training data, this interception-based detection is superior under actual transmission environment conditions. The alternative, a simulation-based detection method utilizes some undisguised parameters of the primary user in the spectrum of interest. Owing to the diversified predetermined training data, simulation-based detection exhibits transcendental robustness against harsh noise environments, although it demands a more complicated classifier network structure. Additionally, for the above-described training methods, we discuss the classifier complexity over implementation conditions and the trade-off between robustness and detection performance. The simulation results show the advantages of the proposed method over conventional spectrum-sensing schemes.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2017EBP3447/_p

Copy

@ARTICLE{e101-b_11_2348,
author={Hang LIU, Xu ZHU, Takeo FUJII, },
journal={IEICE TRANSACTIONS on Communications},
title={A New Classification-Like Scheme for Spectrum Sensing Using Spectral Correlation and Stacked Denoising Autoencoders},
year={2018},
volume={E101-B},
number={11},
pages={2348-2361},
abstract={In this paper, we propose a novel primary user detection scheme for spectrum sensing in cognitive radio. Inspired by the conventional signal classification approach, the spectrum sensing is translated into a classification problem. On the basis of feature-based classification, the spectral correlation of a second-order cyclostationary analysis is applied as the feature extraction method, whereas a stacked denoising autoencoders network is applied as the classifier. Two training methods for signal detection, interception-based detection and simulation-based detection, are considered, for different prior information and implementation conditions. In an interception-based detection method, inspired by the two-step sensing, we obtain training data from the interception of actual signals after a sophisticated sensing procedure, to achieve detection without priori information. In addition, benefiting from practical training data, this interception-based detection is superior under actual transmission environment conditions. The alternative, a simulation-based detection method utilizes some undisguised parameters of the primary user in the spectrum of interest. Owing to the diversified predetermined training data, simulation-based detection exhibits transcendental robustness against harsh noise environments, although it demands a more complicated classifier network structure. Additionally, for the above-described training methods, we discuss the classifier complexity over implementation conditions and the trade-off between robustness and detection performance. The simulation results show the advantages of the proposed method over conventional spectrum-sensing schemes.},
keywords={},
doi={10.1587/transcom.2017EBP3447},
ISSN={1745-1345},
month={November},}

Copy

TY - JOUR
TI - A New Classification-Like Scheme for Spectrum Sensing Using Spectral Correlation and Stacked Denoising Autoencoders
T2 - IEICE TRANSACTIONS on Communications
SP - 2348
EP - 2361
AU - Hang LIU
AU - Xu ZHU
AU - Takeo FUJII
PY - 2018
DO - 10.1587/transcom.2017EBP3447
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E101-B
IS - 11
JA - IEICE TRANSACTIONS on Communications
Y1 - November 2018
AB - In this paper, we propose a novel primary user detection scheme for spectrum sensing in cognitive radio. Inspired by the conventional signal classification approach, the spectrum sensing is translated into a classification problem. On the basis of feature-based classification, the spectral correlation of a second-order cyclostationary analysis is applied as the feature extraction method, whereas a stacked denoising autoencoders network is applied as the classifier. Two training methods for signal detection, interception-based detection and simulation-based detection, are considered, for different prior information and implementation conditions. In an interception-based detection method, inspired by the two-step sensing, we obtain training data from the interception of actual signals after a sophisticated sensing procedure, to achieve detection without priori information. In addition, benefiting from practical training data, this interception-based detection is superior under actual transmission environment conditions. The alternative, a simulation-based detection method utilizes some undisguised parameters of the primary user in the spectrum of interest. Owing to the diversified predetermined training data, simulation-based detection exhibits transcendental robustness against harsh noise environments, although it demands a more complicated classifier network structure. Additionally, for the above-described training methods, we discuss the classifier complexity over implementation conditions and the trade-off between robustness and detection performance. The simulation results show the advantages of the proposed method over conventional spectrum-sensing schemes.
ER -