Abstract: In order to realize the optimal resource allocation between wireless backhaul link and access link in 5G ultra-dense heterogeneous network,the joint user scheduling and power allocation problems of two-layer heterogeneous network in multi-user scenario are studied.In the case of stable queue and limited wireless backhaul resources,by comprehensively considered factors such as user scheduling,power allocation,interference control,the optimal resource allocation problem of in-band wireless backhaul is mathematically modeled and solved,and the optimization algorithm combining user scheduling and power allocation based on Lyapunov optimization theory is proposed.The optimization problem is decoupled into the scheduling of each user in the network and the power allocation process of macro base stations and small base stations.The user scheduling vector is solved by using the MOSEK solver and binary classification method,and the power allocation problem is solved by using the Lagrange multiplier method.On this basis,the optimal resource allocation is achieved through the time update process of the queue.Simulation results show that the proposed scheme can effectively improve the overall network throughput and network utility performance in the multi-user scenario,and its communication performance of the millimeter wave band is better than that of the traditional cellular network band.

Key words: 5G ultra-dense heterogeneous network, wireless backhaul, user scheduling, power allocation, Lyapunov optimization

摘要： 为实现5G超密集异构网络中无线回传链路和接入链路之间的最优资源分配，研究多用户场景下双层异构网络的联合用户调度和功率分配问题，在队列稳定和无线回传资源有限的情况下，综合考虑用户调度、功率分配和干扰控制等因素，对带内无线回传的最优资源分配问题进行数学建模并求解，基于李雅普诺夫优化理论提出联合用户调度和功率分配的优化算法。将优化问题解耦为网络内各个用户的调度以及宏基站和小基站的功率分配过程，采用MOSEK求解器和二分类方法获得用户调度向量，利用拉格朗日乘子法求解功率分配问题，并通过队列的时刻更新过程实现最优资源分配。仿真结果表明，在多用户场景下，该方案能够有效提升网络总吞吐量以及网络效用，并且毫米波频段的通信性能优于传统蜂窝网络频段。

关键词: 5G超密集异构网络, 无线回传, 用户调度, 功率分配, 李雅普诺夫优化

CLC Number: 

• TN929.5