Research On Node Deployment And Topology Optimization Strategy In FSO-based 5G Backhaul Networks

The 5th generation mobile networks(5G)is committed to the construction of ecosystem of information and communication technology.The backhaul network is an important part of it for guaranteeing the end-to-end service,which is evolving to low cost,flexible networking,and scalable networks.Free space optical communication(FSO)has the advantages of high capacity,low interference and long transmission distance.Using FSO and network flying platform(NFP)is a promising solution for backhaul network.Due to the transmission distance threshold of the FSO link and the maximum number of the FSO transceivers of NFP,the deployment of NFP nodes determines the establishment of the FSO links,and at the same time,the establishment of the FSO links influence the utilization of node degree in the network,thereby affecting the deployment of NFP nodes.How to construct the backhaul network and optimize the deployment of NFP nodes and network topology under the premise of meeting the traffic demands and with the consideration of constraints in the network is a key issue.In this paper,in order to realize efficient coverage of NFP nodes,the deployment of NFP and the network topology formation are optimized with the constraints of link reliability,node degree and others in the FSO-based 5G backhaul framework.The main works and innovations of this paper are as follows:(1)The research of the FSO-based 5G backhaul framework.In the FSO-based 5G backhaul network,unlike traditional wireless base station,due to FSO is a line-of-sight propagation point-to-point communication technology and the number of FSO transceivers of NFP node is limited,NFP nodes based on the FSO transceivers is disabled to serve all users in their coverage area.The traditional backhaul framework is unsuitable for the backhaul network in the future,and a new backhaul framework need to be proposed.In this paper,with reference to the hierar-chical FSO backhaul network model and the 5G-Xhaul framework,an FSO-based 5G backhaul framework is proposed,which is to deploy NFP nodes and construct network topology of which goal is to realize efficient coverage of NFP nodes.(2)The strategy of NFP nodes deployment based on efficient coverage.In the FSO-based 5G backhaul framework,there are many constraints in the network,such as the node degree constraint,the route continuous constraint,the link transmission distance constraint.Although NFP has the advantages of flexible and easy-to-deploy,how to realize efficient coverage of NFP nodes is still a key issue.In this paper,we firstly the optimization problem into a mixed integer nonlinear programming and linearized it.Then,a Greedy-based NFP deployment algorithm is proposed.The algorithm deploys NFP nodes one after another and deploy the new NFP node only when there are no base stations can connect to the deployed NFP nodes.The simulation results show that the number of NFP nodes obtained by the algorithm is at most one NFP compared with the optimal solution and has lower computational complexity in the same network.(3)The network topology optimization based on efficient coverage of NFP nodes.In the FSO-based 5G backhaul framework,due to the FSO link transmission distance threshold and the maximum number of FSO transceivers of the NFP node,the NFP nodes deployment and the FSO network topology formation are mutually restricted.In this paper,in order to realize efficient coverage of NFP nodes and minimization of network congestion,a mixed integer linear programming is established to jointly optimize the NFP nodes deployment and network topology formation with the constraints of the node degree constraint,the wavelength conservation constraint,the link reliability constraint and others,and to obtain the optimal location of NFP nodes and the network topology.Then,a joint node deployment and topology formation algorithm is proposed in this paper.The first part of the algorithm is the Greedy-based NFP nodes deployment algorithm to deploy the NFP nodes with the consideration of the traffics of base stations of which goal is to realize efficient coverage of NFP nodes.In order to minimize the network congestion,the second part of the algorithm is the particle swarm optimization-based topology formation algorithm to generate network topology,which is inspired by the idea of particle swarm optimization following the optimal solution for iterative optimization.In this paper,the simulation results show the necessity of joint optimization of NFP nodes deployment and topology formation,and the proposed algorithm can get the approximate optimal solution with lower computational complexity is verified by comparing the simulation results of the algorithm with the optimal solution.