Title: Using cognitive radio to deliver green communications: a reinforcement learning approach

Authors: Mohammed Saleh Bendella; Badr Benmammar; Francine Krief

Addresses: STIC Laboratory, University of Tlemcen, Algeria ' LTT Laboratory, University of Tlemcen, Algeria; LaBRI Laboratory, Bordeaux INP, Talence, France ' LaBRI Laboratory, Bordeaux INP, Talence, France

Abstract: In this paper, we are interested in the concept of green networking and the solutions brought by cognitive radio technology in this field. The purpose of this work is to find a mechanism that minimises energy consumption by integrating it in a cognitive radio network. For this, we have used the Q-learning algorithm, a reinforcement learning technique that will help the cognitive users to find the optimal channel that has a low transmission power by guaranteeing the needs of their application and therefore a reduction in the energy consumption of their batteries while minimising interference in the network. The obtained results are very satisfactory because we have shown that through the integration of the Q-learning algorithm in a cognitive radio network, we have been able to significantly reduce the energy consumption and the interferences of the cognitive radio terminals and therefore we have satisfied the service of green networking.

Keywords: green networking; cognitive radio; energy efficiency; artificial intelligence; reinforcement learning; knowledge base; Markov decision process; MDP.

DOI: 10.1504/IJKEDM.2019.105241

International Journal of Knowledge Engineering and Data Mining, 2019 Vol.6 No.4, pp.331 - 355

Received: 17 Mar 2019
Accepted: 28 May 2019
Published online: 22 Feb 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article