认知无线Mesh网络中基于协作学习自动机的自适应信道分配算法*

在认知无线Mesh网络多信道环境中处理繁重的数据业务是一项具有挑战性的任务。为了解决这个问题，提出了一种新的基于协作学习自动机(Learning automata，LA)的自适应信道分配算法(CLACAA)。在所提出的算法中，LA被部署在邻近的互相协作的次用户上。首先，为了LA能自适应地更新其动作概率向量，设计了一种线性奖赏无为方案；其次，定义了一种新的信道利用率因子用于信道选择，以解决信道冲突发生的问题；最后，给出了当信道由于大量输入请求而过载时的信道切换方案。实验仿真结果表明，该算法在提升网络吞吐量和数据传输率，降低切换时延和缓冲时延上有明显的优势。

Cooperative learning-automata based adaptive channel allocation algorithm in cognitive wireless mesh networks

It is a challenging task to deal with heavy data traffic in the multi-channel environment for Cognitive Wireless Mesh Networks. To solve this problem, this paper proposed a new cooperative learning-automata based adaptive channel allocation algorithm(CLACAA). The proposed algorithm deployed LA in the secondary users cooperating with each other in the vicinity. First, for updating their action probability vector by taking adaptive decisions, it designed a linear reward inaction; Secondly, when the channel conflict occurred, it defined a new channel utilization factor for channel selection. Finally, if a channel was overloaded due to a large number of input requests, it gave a channel switching scheme. Experimental results show that the proposed algorithm has obvious advantages in improving network throughput and data transmission rate, reducing switching delay and buffering delay.