Research On The Resource Allocation For Device-to-Device Communication Underlaying Cellular Networks

With the development of wireless communication and intelligent terminals,the amount of mobile data has increased dramatically,and the types of mobile services have become increasingly diversified.Mobile operators are facing serious challenges in terms of insufficient spectrum resources.In order to accommodate various communication scenarios and communication services,the fifth-generation mobile communication system proposes to support the peak rates up to tens of Gbps and massive connection densities up to 1 million devices per square kilometer.The traditional cellular network centered on the base station will face serious problems such as low spectrum efficiency and overloaded.D2D communication is introduced into the cellular network in the 5G network architecture,making it a complementary mechanism to the base station by reusing the channel with cellular cells,which can reduce the load on the base station,satisfy the connection requirements of massive user terminals,enhance cell coverage,and increase the overall user rate.Currently,cellular and D2D hybrid networking has become one of the popular network architectures of 5G communication systems.This dissertation focuses on the resource allocation for device-to-device communication underlaying cellular networks.The main content is divided into the following two parts:First,in the scenario of content distribution based on multicast D2D communication underlaying a cellular network,a novel optimization scheme of clustering strategy and resource allocation is proposed aiming at the co-channel interference caused by channel reuse and the limited capacity of multicast D2D communication caused by the worst channel quality in the cluster.We formulate the optimization problem to maximize the total system capacity to ensure the minimum SINR for cellular and D2D users,which is difficult to find the optimal solution.We consider decomposing it into two sub-problems to find the sub-optimal solution which called power control sub-problem and joint clustering and channel allocation sub-problem.Based on k-medoids clustering algorithm,a Channel Quality-based Clustering Algorithm is proposed to improving the multicast D2D capacity.Then,through the joint clustering and channel allocation strategy,the co-channel interference is reduced.Simulation reveals that,the proposed CQCA can significantly improve the minimum throughput of multicast D2D users and result in a great improvement on the total system throughput.Secondly,aiming at the problem that instantaneous CSI is difficult to obtain in actual communication scenarios,as well as the influence of the mobility of D2D terminals and the social interaction between users on the reliability of D2D links,this dissertation adopts statistic CSI and user social statistic information to characterize the wireless links state and the reliability of D2D links and then proposes a resource allocation based on statistical information for D2D communication underlaying cellular networks.The resource allocation scheme realizes the maximization of the ergodic capacity of cellular users on the premise of ensuring the QoS of minimum ergodic capacity of cellular users and minimum successful transmission probability of D2D users.To characterize the reliability of D2D communication links,the average interaction duration of users in social networks is used as a social statistic,and a successful transmission probability is defined for D2D users according to the user's average interaction duration.A reliable D2D link should satisfy that the successful transmission probability is larger than the specified minimum successful transmission probability threshold.Since the interference only exists between each CU-D2D pair,we first obtains the optimal power allocation for each CU-D2D pair,and then finds the optimal spectrum reuse through the matching algorithm.The simulations show that the proposed algorithm using the user social statistics information can improve the reliability of D2D links while ensuring the capacity of the cellular users.