分簇拓扑下分层异构网络延迟优化方法仿真

摘要

为了提升分簇拓扑下分层异构网络通信的性能,需要进行分簇拓扑下分层异构网络延迟优化.但是采用当前方法进行异构网络延迟优化时,无法给出异构网络感知准确度,存在网络延迟优化误差大的问题.为此,提出一种基于POMDP的分簇拓扑下分层异构网络延迟优化方法.该方法先融合于POMDP原理使用状态转移函数和观察函数表述系统完成用户请求任务时所存在的外界干扰噪声和传输冲突等不确定因素,引入回报函数映射出在任务执行过程中异构网络感知的准确度和能量消耗状态,计算出任务调度优先级的参数权值,融合于匈牙利理论获取基站和用户的关联、基站对用户的传输功率,以此为依据完成对分簇拓扑下分层异构网络延迟优化.仿真证明,所提方法延迟优化准确度高,极大的提升了分簇拓扑下分层异构网络的服务质量.%This paper proposes an optimization method for delay of hierarchical heterogeneous network under cluster topology based on Partially Observable Markov Decision Processes (POMDP).Firstly,the state transition function and observation function are used to express uncertain factors such as external interference noise and transmission collision existing in user request completed by system integrated with POMDP.The reward function is introduced to map accuracy and state of energy consumption sensed by heterogeneous network during process of task execution.Then,the weight of parameter of task scheduling priority is worked out,and the correlation between base station and user and transmission power of station to user is acquired.Finally,the optimization for delay is completed.Simulation results show that the proposed method has high accuracy of delay optimization.It improves service quality of hierarchical heterogeneous network under cluster topology apparently.