Research On Resource Allocation Methods For D2D Underlaying Cellular Networks

In the construct of device-to-device(D2D)communication underlying cellular networks,the approach that licensed bands being reused among cellular links and D2 D links not only leverages the underutilized cellular spectrum,but also complies with the developing trend that local services will contribute significantly to the growth of mobile communications,which is mostly stimulated by technologies like Internet of Things(IoT)etc..Research on resource allocation algorithms for DUCN can restrain interference produced by channel reusing and improve multi-kinds of performances of networks.Contradiction between ultra-broadband and ultra high rates characters of future mobile communication networks and limited radio resource brings to light the fact that density of mobile service users per cell relatively increases potentially.Therefore,resource allocation schemes in a single cell with dense mobile users are mainly studied in this paper.Firstly,a bipartite hypergraph based spectrum sharing algorithm is proposed.Our design aims to maximize the system sum-rate assuming that each channel can be assigned to multilinks.To solve this NP-hard problem,we propose the concept of bipartite hypergraph,construction rules of hyper-edges,and optimal matching algorithm.Simulation results confirm that,comparing with weighted bipartite graph based algorithm,the system sum-rate can be increased approximately by 40 b/s/Hz and the system capacity can be improved about 50%.Secondly,based on graph theory,a resource allocation algorithm with low complexity is designed in D2 D underlaid cellular networks where at most one D2 D pair and one cellular user can reuse a same channel.An integer program is formed to maximize sum-rate.Then it is transformed to an integer program to minimize the sum of interference channel gains,because interference is thought to be the most effective factor to decide whether two links can use a same channel.In order to solve the optimization problem which can be seen as a one-to-one matching problem,directed weighted bipartite and relative definitions are proposed,and the optimization problem is modeled and transformed twice accordingly.Finally,a greedy algorithm,whose complexity is only O(n),is designed to search optimal match pairs.Simulation results show that our algorithm can achieve better throughput and capacity than the weighted bipartite algorithm in certain range while the complexity is reduced two orders of magnitude.Thirdly,the power is firstly optimized for the fairness of transmitting rate between links using a same channel simultaneously.Joint channel allocation and power control optimization for maximization of the overall network sum-rate is formed for the system model.In order to reduce the complexity to solve the problem,it is decoupled into two subproblems,such as channel allocation and power control.Providing that optimal matching set has been got previously,we focus on power control subproblem,and optimal transmitting power solutions which are in closed form or can be searched from a finite set are separately induced for the power control for fairness,the power control for maximization of the overall network throughput,and the power control for maximization of the overall network throughput under the condition that fairness has been guaranteed.Simulation results confirm that the fairness between links using same channel can be improved significantly.