A robust and energy efficient cooperative spectrum sensing scheme in cognitive wireless sensor networks
Résumé
Cooperative spectrum sensing (CSS) is able to effectively solve the hidden terminal, depth
attenuation, multipath shadows and other issues which are not addressed by the single-user
sensing. Therefore, it has attracted a large amount of interest and several CSS algorithms
have been proposed. However, they are not specifically tailored for cognitive wireless sensor
networks (CWSNs) where transmission reliability, power management and interference
avoidance are critical issues. In this paper, we propose a robust and energy efficient CSS
scheme in CWSNs. Firstly, taking into account the limited energy of sensor node, especially
the mobile node, we introduce the nodes of the network into multiple clusters for the CSS in
order to save energy consumed in reporting results and exchanging information and extend
the lifetime of the network. Secondly, we consider that some cognitive nodes may not work
as expected. Hence, facing the problem of faulty nodes in clusters, we propose an evaluation
method which considers simultaneously the node reliability and the mutually supportive
degree among different nodes to support adapted decisions. Finally, after removing the node
of low credibility, the energy efficiency and reliability of each cluster are improved
significantly. Simulation results allow to validate that the proposed method outperforms the
state of the art in energy efficiency and detection reliability, even in presence of faulty nodes.