Cognitive unmanned aerial vehicle spectrum sensing method based on reinforcement learning

Abstract

The invention belongs to the technical field of mobile communication, provides a cognitive unmanned aerial vehicle spectrum sensing method based on reinforcement learning, relates to a cognitive radio technology, and provides an effective spectrum sensing method for efficient utilization of idle spectrums. Because the unmanned aerial vehicle has the characteristics of convenient deployment, high flexibility and the like, the unmanned aerial vehicle is utilized to carry cognitive radio equipment to sense the idle spectrum in the network, and compared with a traditional base station type sensing method, the method can effectively sense the spectrum of a local area. The optimal unmanned aerial vehicle flight path is designed based on a reinforcement learning algorithm, the unmanned aerial vehicle position and the spectrum sensing result are taken as states, flight is taken as actions, different actions in different states are evaluated, the sensed false alarm and false alarm probability is considered, the path strategy is intelligently formulated and improved, and the unmanned aerial vehicle flight path is optimized. And the perceived idle spectrum bandwidth is maximized. The method does not depend on a specific spectrum state statistical model, and the flight sensing trajectory of the unmanned aerial vehicle can adapt to the dynamic change of the spectrum environment.

Description

technical field [0001] The invention belongs to the technical field of mobile communication, and in particular relates to a cognitive unmanned aerial vehicle spectrum sensing method based on reinforcement learning. Background technique [0002] With the advent of the 5G era, a large number of intelligent terminal devices are connected to mobile communication networks, which brings massive data transmission requirements, which requires a large amount of spectrum resources as support, and spectrum shortage has become a pain point in the development of mobile communication networks. [0003] According to the spectrum monitoring results, the current network utilization rate of spectrum is low, and there are a large number of unused spectrum at different times and different locations, which forms a strong contradiction with the huge demand for spectrum by the network. How to effectively exploit these spectrum holes and make reasonable Utilization is the key to improving the effic...

AI Technical Summary

Problems solved by technology

The probability of judging 0 as 1 is defined as the false alarm probability, which is related to the received signal-to-noise ratio. When the signal-to-noise ratio is small, the cognitive radio may mistakenly judge noise as a signal, resulting in perception errors

The probability of judging 1 as 0 is defined as the probability of missing alarm, which is related to the signal power and the threshold value set by the decision (threshold value means that when the signal power is greater than this value, it is considered that there is a signal). When the sensing device is far away from the target frequency band, the actual occupied signal may be judged to be idle because the received signal power is low, which is lower than the set receiving threshold, resulting in a sensing error

Images