The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
La détection du spectre est la première tâche effectuée par les réseaux de radio cognitive (CR). Dans cet article, nous proposons un algorithme de détection de spectre pour le signal de multiplexage par répartition orthogonale de la fréquence (OFDM) basé sur l'apprentissage profond et un graphique matriciel de covariance. L’avantage de l’apprentissage profond dans le traitement d’images s’applique à la détection spectrale des signaux OFDM. Nous commençons par construire le modèle de détection spectrale du signal OFDM, puis analysons les caractéristiques structurelles de la matrice de covariance (CM). Une fois que CM a été normalisé et transformé en une représentation en niveaux de gris, la carte en échelle de gris de la matrice de covariance (GSM-CM) est établie. Ensuite, le réseau neuronal convolutif (CNN) est conçu sur la base du réseau LeNet-5, qui est utilisé pour apprendre les données d'entraînement afin d'obtenir des fonctionnalités plus abstraites de manière hiérarchique. Enfin, les données de test sont entrées dans le modèle de réseau de détection spectrale entraîné, sur la base duquel la détection spectrale des signaux OFDM est effectuée. Les résultats de simulation montrent que cette méthode peut accomplir la tâche de détection spectrale en tirant parti du modèle GSM-CM, qui offre de meilleures performances de détection spectrale pour les signaux OFDM sous un faible SNR que les méthodes existantes.
Mengbo ZHANG
Electronic Countermeasure Institute
Lunwen WANG
Electronic Countermeasure Institute
Yanqing FENG
Electronic Countermeasure Institute
Haibo YIN
Electronic Countermeasure Institute
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copier
Mengbo ZHANG, Lunwen WANG, Yanqing FENG, Haibo YIN, "A Spectrum Sensing Algorithm for OFDM Signal Based on Deep Learning and Covariance Matrix Graph" in IEICE TRANSACTIONS on Communications,
vol. E101-B, no. 12, pp. 2435-2444, December 2018, doi: 10.1587/transcom.2017EBP3442.
Abstract: Spectrum sensing is the first task performed by cognitive radio (CR) networks. In this paper we propose a spectrum sensing algorithm for orthogonal frequency division multiplex (OFDM) signal based on deep learning and covariance matrix graph. The advantage of deep learning in image processing is applied to the spectrum sensing of OFDM signals. We start by building the spectrum sensing model of OFDM signal, and then analyze structural characteristics of covariance matrix (CM). Once CM has been normalized and transformed into a gray level representation, the gray scale map of covariance matrix (GSM-CM) is established. Then, the convolutional neural network (CNN) is designed based on the LeNet-5 network, which is used to learn the training data to obtain more abstract features hierarchically. Finally, the test data is input into the trained spectrum sensing network model, based on which spectrum sensing of OFDM signals is completed. Simulation results show that this method can complete the spectrum sensing task by taking advantage of the GSM-CM model, which has better spectrum sensing performance for OFDM signals under low SNR than existing methods.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2017EBP3442/_p
Copier
@ARTICLE{e101-b_12_2435,
author={Mengbo ZHANG, Lunwen WANG, Yanqing FENG, Haibo YIN, },
journal={IEICE TRANSACTIONS on Communications},
title={A Spectrum Sensing Algorithm for OFDM Signal Based on Deep Learning and Covariance Matrix Graph},
year={2018},
volume={E101-B},
number={12},
pages={2435-2444},
abstract={Spectrum sensing is the first task performed by cognitive radio (CR) networks. In this paper we propose a spectrum sensing algorithm for orthogonal frequency division multiplex (OFDM) signal based on deep learning and covariance matrix graph. The advantage of deep learning in image processing is applied to the spectrum sensing of OFDM signals. We start by building the spectrum sensing model of OFDM signal, and then analyze structural characteristics of covariance matrix (CM). Once CM has been normalized and transformed into a gray level representation, the gray scale map of covariance matrix (GSM-CM) is established. Then, the convolutional neural network (CNN) is designed based on the LeNet-5 network, which is used to learn the training data to obtain more abstract features hierarchically. Finally, the test data is input into the trained spectrum sensing network model, based on which spectrum sensing of OFDM signals is completed. Simulation results show that this method can complete the spectrum sensing task by taking advantage of the GSM-CM model, which has better spectrum sensing performance for OFDM signals under low SNR than existing methods.},
keywords={},
doi={10.1587/transcom.2017EBP3442},
ISSN={1745-1345},
month={December},}
Copier
TY - JOUR
TI - A Spectrum Sensing Algorithm for OFDM Signal Based on Deep Learning and Covariance Matrix Graph
T2 - IEICE TRANSACTIONS on Communications
SP - 2435
EP - 2444
AU - Mengbo ZHANG
AU - Lunwen WANG
AU - Yanqing FENG
AU - Haibo YIN
PY - 2018
DO - 10.1587/transcom.2017EBP3442
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E101-B
IS - 12
JA - IEICE TRANSACTIONS on Communications
Y1 - December 2018
AB - Spectrum sensing is the first task performed by cognitive radio (CR) networks. In this paper we propose a spectrum sensing algorithm for orthogonal frequency division multiplex (OFDM) signal based on deep learning and covariance matrix graph. The advantage of deep learning in image processing is applied to the spectrum sensing of OFDM signals. We start by building the spectrum sensing model of OFDM signal, and then analyze structural characteristics of covariance matrix (CM). Once CM has been normalized and transformed into a gray level representation, the gray scale map of covariance matrix (GSM-CM) is established. Then, the convolutional neural network (CNN) is designed based on the LeNet-5 network, which is used to learn the training data to obtain more abstract features hierarchically. Finally, the test data is input into the trained spectrum sensing network model, based on which spectrum sensing of OFDM signals is completed. Simulation results show that this method can complete the spectrum sensing task by taking advantage of the GSM-CM model, which has better spectrum sensing performance for OFDM signals under low SNR than existing methods.
ER -