论文部分内容阅读
为解决小型空中飞行平台在接入空天异构无线网络时切换算法对下一时刻网络与用户状态考虑不充分、没有考虑不同业务传输需求的问题,提出了一种支持空天异构无线网络的Q学习优化算法。该算法在回报函数中引入用户体验并且将应用综合预测方法获得的当前与下一时刻网络的SINR、用户移动速度、网络切换代价、信息传输的时延和网络的拥塞程度作为综合评价参数,同时根据层次分析法确定不同业务类型下评价参数权值。仿真结果表明,在用户到达率较高,环境干扰强度较强时,本文算法相比原有基于Q学习的切换决策算法可以有效提高网络切换成功率并降低切换次数;在传输不同类型的业务时,可以提供最优的网络切换策略,有效降低切换阻塞率。
In order to solve the problem that the switching algorithm of the small airborne flight platform is not enough to consider the state of the network and the user at the next moment when accessing the air-space heterogeneous wireless network, and the transmission requirement of different services is not considered, a method is proposed to support the air-space heterogeneous wireless network Q learning optimization algorithm. This algorithm introduces the user experience into the reward function and takes the SINR, the user movement speed, the network switching cost, the delay of information transmission and the congestion degree of the network as the comprehensive evaluation parameters of the current and the next moment obtained by the comprehensive prediction method According to AHP, the weights of evaluation parameters under different business types are determined. The simulation results show that the proposed algorithm can improve the success rate of network handover and reduce the number of handovers effectively compared with the original switch decision algorithm based on Q learning when the user arrival rate is high and the environment disturbance intensity is strong. When transmitting different types of services , Can provide optimal network switching strategy, effectively reduce the switching blocking rate.