面向用户需求的空天地一体化车载网络任务分配策略  

Task assignment strategy for space-air-ground integrated vehicular networks oriented to user demand

作　　者：谭诗翰 金凤林 顿聪颖 TAN Shihan;JIN Fenglin;DUN Congying(School of Command and Control Engineering College, Army Engineering University, Nanjing 210007, China)

机构地区：[1]陆军工程大学指挥控制工程学院,江苏南京210007

出　　处：《系统工程与电子技术》2022年第5期1717-1727,共11页Systems Engineering and Electronics

摘　　要：为了提高空天地一体化车载网络(space-air-ground integrated vehicular networks,SAGVN)内用户的网络服务质量体验,解决不同网络间相互协同的问题,提出了面向用户需求的SAGVN任务分配策略。基于用户信号强度、时延、网络费用和带宽需求,利用效用函数理论和层次分析法(analytic hierarchy process,AHP),构建用户需求和满意度描述框架。将网络任务分配过程抽象为半马尔可夫决策过程(semi Markov decision process,SMDP),根据用户需求和网络状态,利用价值迭代算法获得整体用户满意度最大的网络任务分配策略,利用Q-learning算法得到近似最优策略。实验表明,相较于传统策略,所提策略整体用户满意度提高超过30%;在网络拥塞的环境下,可以有效降低对网络服务需求迫切用户服务请求的拒绝率。To improve the user network service quality experience in space-air-ground integrated vehicular networks(SAGVN)and solve the problem of collaboration among different networks,a task allocation strategy for SAGVN oriented on user demand is proposed.Based on the user demand of signal strength,delay,network cost and bandwidth,the utility function theory and analytic hierarchy process(AHP)are used to construct the description framework of user demand and satisfaction.The network task allocation process is abstracted as semi Markov decision process(SMDP).According to the user demand and network states,the value iteration algorithm is used to obtain the network task allocation strategy to maximize the overall user satisfaction and the approximate optimal strategy is obtained by Q-learning algorithm.Experimental results show that compared with the traditional strategy,the overall user satisfaction of the proposed strategy is improved by more than 30%;inthe network congestion environment,it can greatly reduce the rejection rate of the users with urgent network service demand.

关 键 词：空天地一体化车载网络 无线网络管理 半马尔可夫决策过程 用户需求 Q-learning算法 

分 类 号：TN92[电子电信—通信与信息系统]