Research On Resource Allocation And Power Control In Hybrid D2D Cellular Networks

D2D communication refers to a kind of communication method that mobile devices which is near apart transmit data directly without through the base station.The introduction of D2 D communication in traditional cellular networks can improve network throughput and resource utilization and improve the quality of edge user communications.But D2 D users need to reuse resources of cellular system,which would also introduce the complex Co-channel interference to the network.Therefore,it is very important to design a reasonable resource allocation and power control algorithm to improve the communication quality of hybrid D2 D cellular networks.In reality,there is no algorithm that can be applied to any scene,so we should take different measures according to the different networks models.In this thesis,the hybrid D2 D cellular network is divided into three types: intensive,saturated and sparse,and the corresponding resource allocation algorithms are studied respectively.Firstly,a resource allocation algorithm based on graph coloring clustering has been proposed to solve the problem of resource allocation in dense D2 D cellular networks in this paper.The so-called dense models is a relative concept,and it refers to a scene that the number of D2 D users is much more than the number of cellular users.The algorithm calculates the interference graph between D2 D links firstly,and then divides all the D2 D links into several clusters with small internal interference degree by using the graph coloring theory,so the resource allocation process is transformed into the TSP problem.Finally using the Lagrange multiplier method and the linear programming idea to find the resource allocation result and the corresponding power solution to maximize the system throughput.Secondly,a cooperative game resource allocation algorithm has been proposed to solve the resource allocation problem in saturated D2 D cellular networks based on the competitive game theory.The saturation type is a model between the dense and sparse type.All D2 D users match the cellular channel according to the interference threshold,the power threshold and minimum SINR threshold conditions under the control of the base station.In this paper,it has added constraints matching factor to adjust the unfairness of the network,and the optimal solution of the Pareto is obtained for each link to achieve the minimum signal to interference ratio taking the total data rate of the system as a utility function.Finally,a distributed algorithm based on simulated annealing is proposed to solve the problem of resource allocation in sparse D2 D cellular networks.Sparse refers to the number of D2 D users is far less than the number of cellular users.When the number of D2 D link in the cell is very small,D2 D technology can't bring too much revenue,and centralized management will increase the system overhead.The use of distributed algorithms can reduce the unnecessary signaling costs of the base station,and retain the advantages of D2 D to improve the quality of communication within the network.The characteristics of the sparse D2 D hybrid network ensure that the D2 D link has enough solution space in the search of the optimal multiplexing channel,and the simulated annealing mechanism can avoid the local optimal solution.According to the simulation and analysis results,the above algorithm is used to solve the resource scheduling problem of three hybrid D2 D cellular network types,which can achieve the desired results,and has a good compromise in the minimum signal-to-noise ratio in single user demand,the system throughput and access rate.