优化型自适应频谱感知调度算法

Optimized adaptive spectrum sensing schedule algorithm

针对传统的周期性频谱感知机制对空闲频谱机会的有效利用率不高,且与主用户碰撞概率较大等问题,在多用户协作感知场景下,提出了一种优化型自适应感知调度(OASS)算法.利用部分可观测马尔可夫决策过程理论,将自适应感知调度建模为决策优化控制问题,当每个时隙开始时,从频谱感知、数据传输、休眠等待、信道切换4种可选策略中选择最优策略执行,从而尽可能地减少感知开销浪费.仿真结果表明:与传统的周期性感知调度算法相比,OASS算法能够在有效控制冲撞概率的前提下,使次用户对频谱机会的利用效率明显提高.

The traditional mechanism of periodic spectrum sensing the effective utilization of idle spectrum opportunity is not high,and it has Greater probability of collision with the main users.An optimized adaptive sensing scheduling（OASS） algorithm under the multi-user-cooperation circumstance was proposed.On the basis of the partially observable Markov decision process（POMDP）,the adaptive sensing schedule was modeled as a control problem of policy optimization by the algorithm.At the beginning of each slot,an optimal action was chosen from the action set,including spectrum sensing,data transmission,sleeping and switching,which could reduces the unnecessary sensing cost as much as possible.As represented by the simulation results,the proposed OASS algorithm could obviously improve the secondary users′ valid spectrum utilization,and meanwhile,effectively restrict the interference to the primary system.