Research On Topology Optimization Technology For FSO Based Mobile Fronthaul/Backhaul Networks

The rapidly development of communication network technology and the explosive growth of traffic demands have put forward higher requirements for 5G(fifth-generation)and future communication networks technology in data rate of bandwidth,network cost,flexibility and scalability.As one of the key technologies of mobile communication network,free space optical(FSO)communication has several advantages,such as high data rate,long transmission range,low cost,flexible configuration,and ignorable spectrum interference.However,FSO communications also has some challenges for the FSO-based mobile fronthaul/backhaul networks.Firstly,FSO links are vulnerable to atmospheric environment,which affects the link reliability,thus influences the reliability of the whole network topology.Secondly,the number of links connected by FSO nodes in the network is limited by the number of FSO transceivers,which makes the construction and maintenance of network topology more difficult.In addition,in space environment,due to the limitation of energy carried by unmanned aerial vehicles(UAVs),FSO links in the network are limited.The link performance is greatly reduced by power limitation,which affects the performance of the whole network topology.This paper studies the FSO-based network topology optimization problem from three aspects,i.e.,the network topology construction with the greatest resilience in terms of reliability,the network reconfigurations with the greatest throughput and limited energy resource,and the joint optimization of FSO node placement and resources allocation.Main contributions include:(1)A topology formation mechanism with the maximum resilience based on link instability is proposed.In free space,the reliability of FSO links is fragility due to the influence of atmospheric environment and weather conditions.When link reliability is low,link interruption is considered to affect the reliability of the whole network topology.To solve this problem,this paper proposes a topology optimization mechanism with the greatest resilience in terms of reliability.Firstly,the network resilience is defined as the sum of path reliability.The greater the networkresilience,the more reliable of other paths in the network topology can be used as backup paths for the interrupt link.Secondly,the reliability of FSO links and paths in the network are analyzed and modeled.Conside the minimum link reliability and the limit number of FSO transceivers in network configuration as constraints,and the maximum resilience in terms of reliability as the optimization objective,the network topology formation problem is described as a mixed integer nonlinear programming problem.Then,two heuristic algorithms are proposed to solve the problem from the centralized idea and the distributed idea respectively.Finally,simulation results suggest that the network topologies constructed by the two algorithms have high resilience.Moreover,the centralized algorithm and distributed algorithm have certain advantages in performance and efficiency respectively.(2)A dynamic topology reconfiguration scheme based on link and traffic instability is proposed.The dynamic performance of FSO-based fronthaul/backhaul network is reflected in two aspects:firstly,the traffic data volume of communication network shows an increasing trend and has strong dynamic in space and time;secondly,the vulnerability of FSO links makes the link state very easy to change.Therefore,due to the change of traffic demand and link states,the FSO-based network topology needs to be adjusted accordingly to ensure the reliability of the network topology,in or-der to meet the service requirements of the network.However,frequently network topology reconfiguration will increase network overhead and bit error rate.In this paper,the dynamic reconfiguration scheme of network topology is studied.In order to improve the utilization of network resources,the dynamic reconfiguration scheme of network topology aims at maximizing network throughput and minimizing network cost(i.e.,the transmission power consumption of all FSO links in the network).For different network scenarios,this paper proposes a proactive network topology reconfigure method and a reactive network topology reconfigure method.In static network scenarios,a proactive network topology reconfigure method formulates a bi-objective optimization problem with the objectives of maximizing throughput and minimizing network cost,then solves the problem according to Pareto optimization theorem,and proposes a heuristic topology optimization algorithm to construct a network topology with maximum throughput and minimum network cost.Based on a proactive topology reconfigure scheme,two kinds of topology reconfiguration algorithms are proposed for link interruption and explosion of traffic demand in the network.This dynamic optimal configuration scheme of network topology can not only reduce the overhead of network topology reset,but also ensure that the network has high performance.The simulation results show that with the limited network energy resource,the network topology constructed by the proposed proactive and reactive network topology reconfiguration schemes has the higher network throughput and minimum network cost in different network scenarios.(3)A joint optimization mechanism of nodes placement and resource allocation for ultra-intensive traffic service is proposed.The massive increase of data traffic in ultra-intensive service networks causes congestion for FSO relay networks,and the vulnerability of FSO links leads to the decline of network topology reliability.Thus,ensuring network reliability while improving network capacity and reducing network congestion are the key to topology optimization of ultra-intensive service networks.Aiming at this problem,this paper makes full use of the mobility of UAV,and optimizes the network topology from two aspects by changing the position of UAV.On the one hand,to solve the problem that the unbalanced link traffic loads and link utilization leads to low network throughput,this paper proposes an optimal network configuration algorithm based on traffic load balancing.This algorithm combines the locations of UAV and the association of UAV and access points on the terrestrial networks.It aims at minimizing the number of UAVs and guarantees the balance of link utilization in the network.The simulation results show that the algorithm can achieve the minimum number of UAVs and the balance of link utilization.On the other hand,a large number of access points leading to the increase of traffic demands,which increases the service delay of the network.To solve this problem,this paper proposes a network optimal configuration algorithm based on minimizing the network delay.This algorithm considers the power consumption of UAV configuration in the space network,and jointly optimizes UAV placement problem and the association between UAV and the access points from ground networks.The simulation results show that the network optimization algorithm based on minimizing network delay has better performance in reducing network power consumption and average network delay than the other two distributed optimization algorithms(i.e.,first determining the location of UAV,then determining the allocation relationship,first grouping the ground network and then determining the location of UAV).In summary,this paper mainly studies the optimization technology of FSO-based mobile fronthaul/backhaul network topology,including the construction of network topology with the greatest resilience in terms of reliability,the reconfiguration of network topology with the minimum network cost and the largest network throughput,and the optimization of node olacement in the network.In the process of topology optimization,this paper considers the reliability of FSO links,the limitation of the number of FSO links,the limitation of energy in the network,the demand for services and the dynamic nature of traffic in the network.A large number of simulation results show that the proposed topology optimization algorithms make the network topology more stable,and can effectively improve the network capacity,increase network throughput and reduce network delay.