Algorithm Design For Collaborative Resource Scheduling In Mobile Edge Networks

To meet the diverse and high-standard service requirements of mobile network users,various resources such as communication,computing,and caching are densely deployed in mobile edge networks.Existing network resource management methods,which focus on communication resource scheduling,cannot effectively use these limited and distributed resources,making mobile edge networks face the problems of low service efficiency and large energy cost.To improve the utilization efficiency of network resources,it is necessary to leverage the collaborative features of base stations and coordinate the scheduling of various resources that are distributed in the network.Since the resource scheduling of different base stations are coupled and the allocation of different resources are correlated,it is difficult to coordinate the scheduling of mobile edge network resources.To solve the specific problems in collaborative network resource scheduling and improve the energy efficiency,computing service efficiency and data transmission efficiency of mobile edge networks,this dissertation carries out the following four aspects of research.(1)To reduce the energy cost of base stations in mobile edge networks,a collaborative base station switching and communication scheduling method is proposed in this dissertation.Since the energy cost of base stations includes the operational energy cost and the switching energy cost,this method solves the energy cost optimization problem of base stations in two steps.First,it proposes a user equipment connection allocation algorithm based on energy cost criteria to solve the spatial coupling problem of on/off settings for adjacent base stations,aiming to optimize the operational energy cost of base stations.Second,it represents the on/off state transitions of base stations as a network flow graph,and then proposes a network flow algorithm to optimize the switching energy cost of base stations.The simulation results show that,compared with the existing methods,our proposed method can optimize the on/off settings of base stations and the connection allocation of user equipment,thereby greatly reducing the total energy cost of base stations in mobile edge networks.(2)To reduce the energy cost of cloud radio access networks(Cloud-RAN),a collaborative dynamic resource scheduling method is proposed in this dissertation.This method is based on the dynamic on/off switching strategy of the remote radio head(RRH)and the baseband unit(BBU).It proposes an iterative bin packing algorithm to optimize theon/off setting of RRH and the virtual machine(VM)consolidation of BBU,so as to reduce the network energy cost.Meanwhile,in order to reduce the impact of VM migration on the stability of baseband signal processing,it controls the number of BBUs that need to reconsolidate the VMs,thereby achieving a trade-off between the network energy cost and the number of VM migrations.The simulation results show that our proposed method can significantly reduce the energy cost of Cloud-RAN,and effectively control the number of VM migrations.(3)To improve the computing service efficiency of mobile edge networks,a collaborative service placement and task request scheduling method is proposed in this dissertation.Considering that the computing service requirements are diversified and the number of base stations that can be reached varies for different user equipment,this method proposes a service placement strategy based on the utilization efficiency of computing resources.Further,it transforms the service scheduling problem into a matching problem between computing task requests and network resources,and then it proposes a centralized algorithm and a distributed algorithm to solve this matching problem.The centralized algorithm is based on greedy strategy and network flow method,and the distributed algorithm is based on matching theory.The simulation results show that our proposed method can improve the computing service efficiency of mobile edge networks and reduce the task request traffic that needs to be forwarded to the cloud.(4)To improve the data transmission efficiency of mobile edge networks,a collaborative data placement and transmission scheduling method is proposed in this dissertation.This method optimizes the data placement and transmission scheduling of base stations in two steps.First,it proposes an iterative relaxation linear programming algorithm and two network flow algorithms to schedule the connection between user equipment and base stations,thereby optimizing the set of data files that base stations need to cache.Second,it uses the encoding features of data files that are requested by different user equipment,and proposes a graph vertex coloring algorithm using maximum degree priority strategy to optimize the multicast group partitions of base stations.The simulation results show that our proposed method can improve the data transmission efficiency of mobile edge networks and reduce the total cost of data download and transmission of base stations.