Research On User Association And Resource Allocation Based On Virtual Cells In Ultra-dense Networks

With the rapid development of mobile communication and Internet technologies,the number of smart terminals has been growing explosively,and various services are emerging constantly.To meet the demands for massive intelligent device connectivity and high transmission rates for emerging services,ultra-dense network(UDN)technology has emerged.UDNs can improve spectrum utilization and system throughput by deploying a large number of low-power small base stations,and it have become one of the key technologies in the 5th generation mobile communication systems(5G)and beyond 5G(B5G)era.Although UDNs can improve network performance by utilizing close-range transmission,the densely deployed small base stations also bring many new challenges,such as severe interference,unbalanced base station load,and high network energy consumption.To address these challenges and fully exploit the potential of UDNs,user-centric virtual cells technology can effectively enhance the system throughput and energy efficiency of UDNs while ensuring the reliability of user associations.Among them,user association and resource allocation can effectively improve system performance and user service experience through flexible control of user association to base stations and reasonable resource scheduling.Therefore,this thesis conducts an in-depth study on the topic of "User Association and Resource Allocation Based on Virtual Cells in UDNs" and designs a series of solutions for load balancing,interference management,scarce resources,backhaul capacity burden,and energy efficiency in virtual cells-based UDNs,aiming to improve system throughput and energy efficiency and meet the quality of service(QoS)rate requirements for users.The main work and research contributions of this paper are as follows:1.A user association and resource allocation scheme based on load and QoS constraints is proposed in ultra-dense networks.In response to the problems of unbalanced load,scarce resources,unsatisfied user QoS rate requirements,and severe interference in ultra-dense networks,a joint optimization scheme for user association,physical resource block(PRB)allocation,and power allocation based on load and QoS constraints with the goal of maximizing system throughput is proposed,combined with virtual cells technology.This scheme decouples the original optimization problem,which is difficult to solve directly,into three independent subproblems and proposes three algorithms to solve them,respectively.Firstly,to achieve load balancing and meet user QoS rate requirements,a user association algorithm based on user QoS rate requirements is proposed.Secondly,an extended low-complexity graph theory method is proposed for PRB allocation,effectively reducing interference and improving system throughput.Finally,the difference of convex(DC)programming method is used for power allocation,transforming the original non-convex problem into a convex problem and obtaining a convergent power allocation algorithm,further improving system throughput.Simulation results show that the proposed algorithms have significant performance improvements in load balancing,interference mitigation,satisfying QoS rate requirements,and improving system throughput compared to other algorithms.2.A user association and power allocation scheme based on backhaul capacity constraints is proposed in ultra-dense millimeter-wave networks.Millimeter wave communication technology is introduced into ultra-dense network.To address the severe signal blockage and beam interference issues in ultra-dense millimeterwave networks,a joint optimization scheme for user association and power allocation based on backhaul capacity constraints is proposed,which aims to maximize system throughput and satisfy user QoS rate requirements,combined with virtual cells technology and taking into account both access and backhaul links.This scheme models the joint optimization problem as a throughput maximization problem.To effectively solve this complex mixed-integer nonlinear programming optimization problem,it is decoupled into user association and power allocation as two independent subproblems and solved iteratively using an alternating optimization approach.For each iteration,a many-to-many matching method with externalities is used to complete millimeter-wave SBS clustering and successive convex approximation(SCA),Lagrangian dual decomposition,and Newton-Raphson iteration method are used for power allocation.The effectiveness and convergence of the algorithm are verified through system simulations with different network parameter settings.Simulation results show that the proposed algorithm is close to the performance of exhaustive searching algorithms,significantly outperforms various traditional algorithms in improving system throughput and satisfying user QoS rate requirements,but with greatly reduced complexity.3.A high energy-efficient user association,backhaul bandwidth allocation,and power allocation scheme is proposed in ultra-dense millimeter-wave networks.To address the problems of severe beam interference,fixed backhaul bandwidth allocation,high energy consumption,and low energy efficiency in ultra-dense millimeter-wave networks,a joint optimization scheme for user association,backhaul bandwidth allocation,and power allocation is proposed,with the aim of maximizing energy efficiency and satisfying user QoS rate requirements,combined with virtual cells technology.Specifically,the joint optimization problem is modeled as an energy efficiency maximization problem,and the difficult-to-solve mixed-integer fractional programming problem is transformed into a reduced non-fractional problem using the Dinkelbach method.Then,the transformed problem is decoupled into user association,backhaul bandwidth allocation factor,and power allocation as three independent subproblems,and solved iteratively using an alternating optimization approach.In each iteration,a user association algorithm based on energy efficiency maximization is proposed,the optimal backhaul bandwidth allocation factor is derived,and a power allocation scheme using Successive Convex Approximation and Difference of Convex programming methods is proposed.Finally,simulation validation is performed,and the results show that the proposed algorithm effectively improves system energy efficiency and satisfies user QoS rate requirements.