Research On Clustering Algorithm Based On LEACH Protocol In Free Space Optical Sensor Network

Free space optical(FSO)is a communication technology with high communication bandwidth,abundant spectrum resources,high security,and flexible networking.And the technologies of wireless sensor network(WSN)have already been mature.With the advancement of science and technology,in recent years,the free space optical sensor network(FSOSN)has gradually developed and becomes a hot spot in FSO research.One of the major issues in the FSOSN is the energy consumption of optical wireless sensor nodes.In order to extend the network lifetime,it is important to improve the energy efficiency of the nodes.Clustering algorithm is hot spot of energy-saving technology.In the clustering algorithm,member nodes should avoid transmitting data to the base station directly,and the data should be converged by the cluster head.As a typical clustering algorithm,LEACH protocol can effectively reduce the energy consumption of data transmission in the wireless sensor network.But at the same time,the protocol has some shortcomings,such as: the cluster head selection mechanism is random;in the cluster head election stage,the influences of the location of the nodes,the remaining energy,and the density of nodes are not considered.These shortcomings may cause the occurrence of uneven distribution of clusters in the network and excessive death of nodes.At the same time,the LEACH protocol is applied to the node-stationary sensor network,its performance in the node-mobility sensor network is not ideal.In this paper,we refer to the omni-directional smart optical antenna and the energy consumption models of wireless optical sensor nodes.At the same time,the related technologies in WSN are used for reference.In view of the shortcomings of the classical LEACH protocol,clustering algorithms for node-stationary and node-mobility FSOSNs are studied respectively.In the node-stationary FSOSN,according to the classical LEACH protocol,the data in the cluster is directly transmitted to the base station after being converged by the cluster head.As the transmission distance increases,the energy consumption will also increases.Those cluster heads farther away from the base station will die earlier,and it leads to an unstable network topology.In the node-mobility FSOSN,the topology of the network changes frequently due to the movement of nodes,and the cluster head election should consider the mobility of nodes.Combining with LEACH protocol,this paper proposes the clustering algorithms LEACHNS(Improved LEACH)and LEACH-NM(Node Mobility LEACH)for the node-stationary and node-mobility FSOSN.Firstly,according to the node distribution and energy consumption model,LEACH-NS obtains the cluster radius formula which makes the local network energy consumption lowest,and then we improve the cluster head election threshold formula,focusing on the remaining energy,regional node density,and distance from the base station.Then the data transmission stage is described in detail.In LEACH-NM,the node random movement model is used to simulate the movement of nodes.The cluster radius formula refers to that in LEACH-NS.The remaining energy,the density of regional nodes,and the distance to the base station are still important in the cluster head threshold formula,and the key factor is the speed of node.Moreover,the mobility factor is supplemented for cluster head election,and we introduce the communication cost factor in the clustering stage to help the member nodes to select the appropriate cluster.Then the data transmission stage is described in detail.Finally,LEACH-NS and LEACH-NM are simulated and compared with LEACH protocol by MATLAB.Simulation results show that both LEACH-NS and LEACH-NM reduce network energy consumption,extend network lifetime,and increase network data transmission.