Research On Cluster-Based Dynamic Topology Control Algorithm For Optical Wireless Communication Networks

Optical Wireless Communication(OWC)is a new type of broadband wireless communication technology,which is promulgated through atmosphere and laser,and it is used for the transmission of real information for the final users,such as data,voice,image and video etc.Due to high bandwidth,low cost,high security and other advantages,OWC has become one of the most important research topics in the field of communications.Nowadays,point-to-point communication technology of OWC system is widely applied to many fields.However,it is necessary to construct mobile OWC networks to guarantee in-time communication among multiple users with the increasing demand for Qos and capacity of mobile communications.Therefore,mobile OWC networking becomes an important research direction for future mobile communications networks.In mobile OWC networking technology,dynamic topology control is one of the most important technologies for establishing a network.But there are few investigations on the field of dynamic topology control in mobile OWC networks.Considering the characteristics of OWC network such as small divergence angle of laser,small maximum constraint degree of nodes,slow tracking alignment and unstable links,this paper,based on the traditional topology control technology of mobile Ad Hoc network,the technological frontier of the current OWC network and deeply studies the topology control algorithm of OWC network,proposes a Cluster-Based(CB)Dynamic Topology Control Algorithm,which is referred to as CB Algorithm.The proposed algorithm is suitable for solving the topology control problem of OWC network under the condition of unstable links,limited degree of nodes,and node mobility.And it achieves higher connectivity to form a stable topology structure using the minimum node degree.The CB algorithm adopts a hierarchical network structure and contains three sub-algorithms:1)a Link Expiration Time(LET)clustering algorithm,2)a cluster head selection algorithm,and 3)an inter-cluster connection algorithm.Firstly,the predicted values of LET is calculated as link weights to form a stable clustering structure.Then,the cluster head selection algorithm is used to select the node with stable reliability as the cluster head.Finally,the link reliability is used as the link weight,and the inter-cluster connection algorithm is used to form a connected topology graph.In addition,the corresponding strategy is made to the dynamic network by topology maintenance to achieve the topology control of OWC networks.At last,the CB algorithm for the hierarchical structure OWC network is simulated in the modified Smooth Random Mobility(SRM)model to make the trajectory of nodes more realistic and correct,and to prevent nodes from moving out of the simulation area,and the final topology structure is formed.Then the network performance is simulated on the formed topology structure,and the CB algorithm is compared with the Block-Node Based(BNB)topology control algorithm and the typical Minimum Spanning Tree(MST)algorithm.The simulation results show that the CB algorithm can improve the network topology stability,algebraic connectivity and scalability,and has better reliability.Besides,theoretical analysis and simulation results show that the CB algorithm has obvious advantages on topology stability,connectivity and scalability under low maximum degree constraint of mobile node,and it is applicable to control topology for large-scale networks,and can effectively solve the problem of poor robustness,scalability,and stability of dynamic OWC networks.