基于学习机的WSNs栅栏覆盖构建算法

A Realization Scheme for Barrier Coverage Based on Learning Automata in WSNs

边界监视是无线传感网络WSNs(Wireless Sensor Networks)的一个重要应用,而栅栏覆盖是实现边界监视的有效覆盖技术。为此,提出基于学习机的栅栏覆盖算法BCLA(Barrier Coverage based on Learning Automata)。BCLA算法的目的在于以最少的节点数实现对网络边界的监视。BCLA算法先利用学习机形成动作概率矢量,然后,再选择具有最大动作概率的节点构建栅栏,使得每条栅栏的节点数尽可能少。实验数据表明,提出的BCLA算法所构建的栅栏数优于同类算法。与最大强栅栏MSBA(Maximizing Strong Barriers Algorithm)算法相比,提出的BCLA算法所构建的栅栏数提高约8%。

Border surveillance is an application of wireless sensor networks(WSNs).The present issue of implementing this application is a barrier coverage in WSNs.Therefore,a realization Scheme for Barrier Coverage based on Learning Automata is proposed in this paper,which marked as BCLA.BCLA aims to find the minimum possible number of nodes in each barrier to monitor the network borders.The action probability vector of nodes are computed by learning automata,the nodes with maximum probability are used to construct the barriers in order to minimize the numbers of nodes in each barrier.The simulation results show that the BCLA scheme outperformed some state-of-art barrier-coverage algorithms.The BCLA has better 8%better performance in average than maximizing strong barriers Algorithm(MSBA)in term of number of barriers.