Reduction of Computational Time for Cooperative Sensing Using Reinforcement Learning Algorithm
No Thumbnail Available
Files
Date
2019
Journal Title
Journal ISSN
Volume Title
Publisher
Published by IEEE Nigeria, Computer Science Section
Abstract
Cooperative spectrum sensing in cognitive radio systems is characterized by high computational
time for decision making due to the fusing of individual decisions of cognitive radios involved in
the cooperative scheme. This increases the communication overhead of the network. In this
paper, an adaptive cooperative spectrum sensing algorithm is developed with improved detection
algorithm. Reinforcement learning is thenincorporated to improve the decision making efficiency
of the cooperative spectrum sensing such that less time is required to make a decision at the
fusion centre. Three temporal difference learning techniques were compared in order to select the
most efficient to reduce sensing and decision delays. Appropriate learning rate was utilized in the
sensing and decision making algorithm to enhance the performance ofthe adaptive cooperative
spectrum sensing. Results reveal significant reduction in the computation time required in
cooperative spectrum sensing and decisions. This permits greater efficiency in dynamic spectrum
management as the limited electromagnetic spectrum is being utilized for telecommunication
services.
Description
Keywords
Cognitive Radio, Cooperative Spectrum Sensing, Computational Time, Reinforcement Learning