Wireless Network Optimization Scheme Recommendation Using Monte Carlo Tree Search
Xu Jian, Xi Wang · 2022
Wireless network users have increased significantly, and the wireless network scale has expanded rapidly. It is imperative to improve the efficiency of network optimization and network quality. Inspired by AlphaGo-Zero, this paper proposes a wireless network optimization scheme recommendation method based on Monte Carlo Tree Search (MCTS). Combined with the characteristics of the problem studied in this paper, The Upper Confidence Bounds applied for Trees (UCT) algorithm in MCTS iteration is improved and compared with traditional UCT and two reinforcement learning algorithms. The results show that the recommendation model of wireless network optimization scheme based on the improved UCT algorithm with Monte Carlo tree search as the framework has the highest average value of comprehensive improvement of network indicators, which provides a new model and means for network optimization problems.