Auction-based Resource Allocation For Mobile Edge Computing Networks

The explosive traffic growth has forced a change in the mobile network architecture.Traditional cloud computing centers have been unable to meet the requirements of new applications such as Virtual Reality(VR),Augmented Reality(AR)and real-time holographic projection.Computing sinking has become a trend.Mobile Edge Computing(MEC)is a new computing paradigm,which provides computing support for resource-constrained user equipments.However,most of the existing studies only considered the needs of users and ignored the interests of MEC operators.Since MEC operators are not obliged to provide computing services to users,it is important to design an effective incentive framework to encourage MEC operators to provide computing services.In addition,users have different Quality of Service(QoS)requirements in application scenrios such as AR and VR.In order to meet the individual needs of customers,MEC operators should provide differentiated services to users.In this thesis,auction theory is used to solve the resource allocation problem of mobile edge networks,and an auction based multi-round iterative allocation algorithm is proposed,which effectively stimulates MEC operators to provide users with computing services.In addition,in order to guarantee the QoS requirements of users at different levels,this thesis also proposes a QoS-driven mobile edge network resource allocation algorithm.The main contributions of this thesis are shown in the below.(1)For the multi-MEC server and multi-user scenario,an auction based multi-round iterative resource allocation algorithm is proposed to maximize the benefits of MEC operators while ensuring the QoS requieements of users.This thesis designs an effective incentive framework to encourage MEC operators to provide computing services for users.The problem of jointly allocating computing and communication resources to maximize the revenue of MEC operators is studied.In addition,this thesis also proves that the proposed double auction algorithm has nice properties including truthfulness,individual rationality and budget balance.Simulation results show that the proposed resource allocation algorithm can significantly improve the interests of MEC operators.(2)For the single MEC server and multi-level user scenario,a QoS-driven mobile edge network resource allocation algorithm is proposed.The proposed algorithm considers the user levels and the number of users in each level when allocating MEC server resources.In the process of resource allocation,more resources are allocated to high-level and heavily loaded edge network slices.In addition,the proposed algorithm also considers the interests of MEC operators.Simulation results show that the proposed resource allocation algorithm can satisfy the needs of high-level users while protecting the interests of MEC operator.