Unsupervised deep learning to solve power ...
Document type :
Pré-publication ou Document de travail
Permalink :
Title :
Unsupervised deep learning to solve power allocation problems in cognitive relay networks
Author(s) :
BENATIA, Yacine [Auteur]
Ecole nationale supérieure Mines-Télécom Lille Douai [IMT Nord Europe]
Ecole Nationale Supérieure de l'Electronique et de ses Applications [ENSEA]
Savard, Anne [Auteur]
Negrel, Romain [Auteur]
Belmega, Elena Veronica [Auteur]
Ecole nationale supérieure Mines-Télécom Lille Douai [IMT Nord Europe]
Ecole Nationale Supérieure de l'Electronique et de ses Applications [ENSEA]
Savard, Anne [Auteur]

Negrel, Romain [Auteur]
Belmega, Elena Veronica [Auteur]
English keyword(s) :
Unsupervised deep learning
Full-duplex relaying
Decode-and-Forward
Cognitive radio
Full-duplex relaying
Decode-and-Forward
Cognitive radio
HAL domain(s) :
Informatique [cs]/Intelligence artificielle [cs.AI]
Informatique [cs]/Réseaux et télécommunications [cs.NI]
Informatique [cs]/Réseaux et télécommunications [cs.NI]
English abstract : [en]
In this paper, an unsupervised deep learning approach is proposed to solve the constrained and non-convex Shannon rate maximization problem in a relay-aided cognitive radio network. This network consists of a primary and ...
Show more >In this paper, an unsupervised deep learning approach is proposed to solve the constrained and non-convex Shannon rate maximization problem in a relay-aided cognitive radio network. This network consists of a primary and a secondary user-destination pair and a secondary full-duplex relay performing Decode-and-Forward. The primary communication is protected by a Quality of Service (QoS) constraint in terms of tolerated Shannon rate degradation. The relaying operation leads to non-convex objective and primary QoS constraint, which makes deep learning approaches relevant and promising. For this, we propose a fully-connected neural network architecture coupled with a custom and communication-tailored loss function to be minimized during training in an unsupervised manner. A major interest of our approach is that the required training data contains only system parameters without the corresponding solutions to the non-convex optimization problem, as opposed to supervised approaches. Our numerical experiments show that our proposed approach has a high generalization capability on unseen data without overfitting. Also, the predicted solution performs close to the brute force one, highlighting the high potential of our unsupervised approach.Show less >
Show more >In this paper, an unsupervised deep learning approach is proposed to solve the constrained and non-convex Shannon rate maximization problem in a relay-aided cognitive radio network. This network consists of a primary and a secondary user-destination pair and a secondary full-duplex relay performing Decode-and-Forward. The primary communication is protected by a Quality of Service (QoS) constraint in terms of tolerated Shannon rate degradation. The relaying operation leads to non-convex objective and primary QoS constraint, which makes deep learning approaches relevant and promising. For this, we propose a fully-connected neural network architecture coupled with a custom and communication-tailored loss function to be minimized during training in an unsupervised manner. A major interest of our approach is that the required training data contains only system parameters without the corresponding solutions to the non-convex optimization problem, as opposed to supervised approaches. Our numerical experiments show that our proposed approach has a high generalization capability on unseen data without overfitting. Also, the predicted solution performs close to the brute force one, highlighting the high potential of our unsupervised approach.Show less >
Language :
Anglais
Source :
Submission date :
2022-03-04T05:10:13Z
Files
- https://hal.archives-ouvertes.fr/hal-03534545/document
- Open access
- Access the document