| With the increasing demand of wireless traffic and the rapid development of wireless technologies,mobile network operators are faced up with two chanllenges.First,the capital expenses(CapEx)and operation expenses(OpEx)of mobile network operators are significantly increased.Second,as different radio access technologies’networks begin to accelerate,there lacks a methodology to coordinate and manage the resources of different networks.To overcome above challenges,wireless network virtualization(WNV)is considered as a potentail technology.With WNV,physical infrastructure can be decoupled from specific services and applications.Consequently,different mobile network operators(MNOs)can share the physical infrastructures,thus significantly improving the flexibility of resource management and reducing capital expenses(CapEx)and operation expenses(OpEx).This thesis studies the air interface virtualization in LTE networks and the virtualization in heterogeneous wireless networks.Importantly,we study virtualization based resource allocation problems.The main contributions of this thesis include the following aspects.First,this thesis studies power allocation in LTE air interface virtualization.Based on the feature of LTE air interface virtualization,we propose a two-stage power allocation scheme.In the first stage,Vickrey-Clark-Groves(VCG)auction game is utilized to generate an initial allocation where service providers are modeled as bidders bidding for power resources on behalf of their users.In the second stage,the power allocation is dynamically adjusted based on the users’ rate requirement.Simulation comapres the network energy efficiency and users’ rate fairness of the proposed scheme comparing with conventional schemes.Furthermore,we study bandwidth-power allocation in LTE air interface virtualization.To reduce complexity,the problem is decoupled into bandwidth resource allocation problem and power resource allocation problem separately.First we solve the bandwidth resource allocation problem by introducing bankcruptcy game.Afterwards,we solve the power resource allocation problem by introducing fractional programming and convex optimization.Through simulation,the network energy efficiency and resource allocation fairness are improved by using the proposed resource allocation strategies.Second,this thesis studies the virtualization in heterogeneous wireless networks.First,we propose a virtual MAC protocol.Virtual MAC is a generic network protocol layer which resides above MAC layer,but below IP layer.Through virtual MAC,different radio access technologies’ networks can be aggregated in a single network protocol,which enables resource sharing among different mobile network operators.Second,a resource allocation strategy is devised for the proposed virtual MAC.To model the fact that different radio access technologies possess different adaptability to different services,an’adaptability ratio’ concept is introduced and calculated using Grey Relational Analysis.After that,a matching theory based framework is proposed to formulate the resource allocation problem and a ’deferred acceptance’ algorithm is introduced to solve it.Third,a mobility managemenet mechanism is devised for virtual MAC.Through simulation,it is proved that the proposed virtual MAC protocol can efficiently reduce the service interruptions and improve the resource usage.Third,based on the virtual MAC concept,this thesis proposes a social-aware virtual medium access control(SV-MAC)protocol.SV-MAC integrates virtualization and social-awareness to D2D communications underlying heterogeneous cellular networks.Through SV-MAC,heterogeneous networks and different communication types(D2D communication and cellular communication)can be unified in a single protocol stack.What’s more,by leveraging users’social information,SV-MAC can realize social-aware D2D discovery,association and resource allocation for improved D2D communication efficiency.Furthermore,in order to validate the advantages of the proposed SV-MAC protocol,we study the cellular and D2D resource allocation problems under the SV-MAC protocol.The problems are formulated as maximizing the cellular and D2D energy efficiency(EE).The cellular resource allocation problem is firstly solved considering its priority to D2D users,which satifies the rate requirement of cellular users.The resource allocation problem for D2D users is solved after cellular users to control the D2D-to-cellular interference within a tolerable threshold.In the simulation,we introduce real mobile user traces to simulate the user social information.Simulation results show that the proposed SV-MAC resource allocation schemes obtain a beneficial EE improvement comparing with conventional schemes. |
