摘要
在认知无线电中,为了最大化次用户的吞吐量,同时对主用户的干扰低于预定值,提出一种基于POMDP的信道感知接入算法。次用户将主用户信道在时间轴上细分成等间隔的时隙,在每个时隙开始时,次用户从频谱感知、以较高的功率接入信道和以较低的功率接入信道三种可选策略中选择最优的策略。将次用户的选择过程建模成一个POMDP问题,并采用一些相应的最优策略求解。计算机仿真结果验证了算法的有效性。
In order to maximize the throughput of secondary user in cognitive radio network, while the interference to pri-mary user under the predefined threshold, a new channel sensing and access algorithm based on Partially Observable Mar-kov Decision Process(POMDP)is proposed in this paper. The primary channel has been divided into several slots equally in time domain. At beginning of each slot, the secondary user selects an optimal strategy from three actions:spectrum sensing, access the channel with a larger power and access the channel with a low power. It formulates the selection of sec-ondary users as a POMDP problem, some optimal strategies are proposed to solve the problem. Illustrative results indicate the efficiency of the proposed algorithm.
出处
《计算机工程与应用》
CSCD
2014年第5期203-207,共5页
Computer Engineering and Applications
基金
安徽省优秀青年人才基金项目(No.2012SQRL256)
安徽省教育厅高校省级自然科学一般研究项目(No.KJ2011ZO92)