DDQN Based Handover Scheme in Heterogeneous Network
2022 14th International Conference on Wireless Communications and Signal Processing (WCSP)(2022)
摘要
With the widely construction of 5G-related infras-tructure, heterogeneous networks are becoming more and more common nowadays. Heterogeneous networks can meet different requirements of users and improve the utilization efficiency of communication resources. However, in high-dynamic heteroge-neous network, too much handovers may occur for mobile users especially high-speed vehicles, that may lead to poor network experience. In order to reduce handover numbers of the users and improve the system throughput, we take the user's received signal strength (RSS), available channels of the base stations (BSs) and user's residence time in the BSs into account when making the handover decision, then devise a heterogeneous network scenario and propose a handover scheme based on DDQN (Double Deep Q Network). Neural network is used to estimate the action value function and solve the problem with high complexity caused by the increase of state and action dimension. The scheme is proved to be valid from the simulation results.
更多查看译文
关键词
handover,deep reinforcement learning,heterogeneous network
AI 理解论文
溯源树
样例
![](https://originalfileserver.aminer.cn/sys/aminer/pubs/mrt_preview.jpeg)
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要