Energy-efficient for Multi-UAV Cognitive Radio Network with Normalized Spectrum Algorithm

Abstract

Unmanned aerial vehicle(UAV) combined with cognitive radio(CR) is a practical application scenario due to its portability and high maneuverability. Aiming at the low energy efficiency of cognitive UAV networks, this paper introduces the normalized spectrum (NS) sensing algorithm into multi-UAV cognitive radio networks to explore the energy efficiency based on cooperative spectrum sensing. Then with a fixed false alarm probability of a single decision, we compare the energy efficiency of the multi-UAV cognitive radio network using the NS algorithm with the energy detection (ED) algorithm. It shows that the NS detection algorithm can achieve a higher energy efficiency than the ED detection algorithm due to the introduction of an additional tunable parameter “the number of segments”. The further simulation indicates that the NS algorithm performs better than the ED algorithm in dynamic noise scenarios with time-varying noise power. Finally, we obtain the optimal sensing time of the NS algorithm to maximize energy efficiency. It shows that a matched pair of sensing time and the number of segments will achieve better performance.