Stochastic selfish KL-UCB algorithm is applied in dynamic spectrum access

• 1、School of Electronics and Information Engineering, Liaoning University of Engineering and Technology, Huludao, Liaoning 125105

Abstract：Dynamic spectrum access technology is a flexible and efficient way of spectrum utilization, which can solve the problem of spectrum resource shortage. With the continuous development of wireless communication technology, DSA technology will be more widely used and popularized in the future. The existing algorithms such as Lugosi and Simehrabian algorithm, SIC-MMAB2 algorithm and selfish KL-UCB algorithm can not improve the efficiency of dynamic spectrum access, which leads to the waste of spectrum resources. For this reason, the random selfish KL-UCB algorithm is proposed in this paper. The total regret value of this algorithm and the above three algorithms is always 0 when the number of devices is changed. Therefore, the stochastic selfish KL-UCB algorithm can make the terminal equipment make the best spectrum selection decision when competing for spectrum resources, and at the same time, it can make the most correct choice in the shortest time considering the competition behavior of other terminal equipment, it can help to improve the spectrum utilization and system performance, and to minimize the waste of spectrum resources.

Keywords： Random selfish KL-UCB algorithm dynamic spectrum access total regret value