基于部分可观察马尔科夫决策过程(POMDP)的贪婪算法次优频谱接入

Greedy Algorithm Sub-optimal Spectrum Access Based on POMDP

在ad hoc网络中,由于受硬件和能量的限制,在没有数据发送的情况下,次用户无法监控到频谱,不能执行全频谱的感知,因此,针对现有的机会频谱接入(OSA)算法大都基于信道状态具有完全知识的马尔科夫建模的情况,提出了基于部分可观察的马尔科夫(POMDP)的决策论方法,该方法综合考虑了机会频谱共享系统的感知和接入优化策略。由于受维数灾难的影响,设计了一种基于贪婪算法次优的频谱接入方法,并从理想感知和感知错误存在两方面进行仿真分析。结果表明,在POMDP模型下该方法能有效地避免主次用户之间的资源使用冲突,实现对频谱的有效利用。

Recognizing hardware and energy constraints in the ad hoc network,it is assumed that a secondary user may not be able to perform full-spectrum sensing or may not be willing to monitor the spectrum when it has no data to transmit. Existing opportunistic spectrum access（ OSA） algorithms are based on channel state with full knowledge of Markov modeling. Based on the network channel modeling Partially Observable Markov Decision Process（ POMDP）,the perception of spectrum sharing systems and access optimization strategies were studied. A greedy algorithm was designed and analyzed based on cross-layer sub-optimal access method from the ideal perception and perception errors existing. The results show that POMDP model can effectively avoid the conflicts and realize the efficient resource sharing.