Research On Resource Allocation Method In D2D Communication

Device-to-Device(D2D),as one of the key technologies of 5G,is a technique of direct communication between adjacent terminals without base station forwarding,which enables short-range transmissions to coexist with cellular transmissions and improves spectral efficiency by interference reduction or interference avoidance.This thesis analyzes the interference problems in D2 D and cellular hybrid networks,and mainly studies the resource allocation methods in D2 D communication,which can maximize the system sum rate and reduce the interference by allocating resources rationally.1.In view of the fact that the current research on resource allocation is relatively single,this thesis studies the joint power allocation and channel allocation algorithm,and analyzes the direct connection scene and relay scene respectively.And the resource allocation algorithm in the direct connection scene is decomposed into three parts: First,the feasible region of user's transmit power is divided.Then the optimal power allocation is solved in the feasible region of the transmitted power.Finally,optimizing the matching order of Kuhn-Munkres(KM)algorithm,and the channel matching is performed by the optimized KM algorithm.In addition,a relay node selection algorithm based on social relations and distance between users is proposed,and the algorithm in direct connection is applied to solve the resource allocation problem.Simulation results show that compared with greedy algorithm,the spectrum efficiency gain is increased by 10% and the acceptance rate is increased by 13%.2.Aiming at the limitations of the above algorithm in spectrum utilization,a resource allocation algorithm based on fairness is proposed to improve the spectral efficiency and the applicability of the algorithm.In this scheme,channel resources of one cellular user are allowed to be reused by multiple D2 D users.When considering fairness,the number of reusable D2 D users on a channel is limited.When solving this problem,firstly the power allocation based on single user water-filling algorithm is preformed over the total number of available channels for each user.The channel resources are then iteratively allocated to maximizing spectrum efficiency until the maximum allowable number of the D2 D users on the shared channel reaches the above limit.Simulation results show that compared with greedy algorithm,the spectrum efficiency is increased by at least 20% and the fairness is increased by 22%.3.at present,most resource allocation algorithms for D2 D communication have a problem that the algorithms only applies to the specific scenarios or under certain conditions,and there is not a complete resource allocation scheme suitable for actual scenarios.Therefore,a set of resource allocation scheme suitable for the actual scenario is designed in this thesis.First,the service priority of D2 D users is calculated.when there are idle resources,the channel resource is allocated first to the D2 D pair with higher service priority.Next,resources are allocated to the remaining D2 D pairs that are not allocated channel resources.The proposed scheme can coordinate the interference between the D2 D users and the cellular users and the interference between different D2 D users.