| The combination of millimeter wave transmission and device to device communication technology has become one of the key technologies of the fifth generation and future mobile communication networks.Because the millimeter wave signal is easily blocked by obstacles,when the network is sparse,the interference in the millimeter wave system is very small,and this is a noise limited system.As the network becomes denser and the intensity of inter-link interference increases,the millimeter wave system changes from a noise-limited system to an interference-limited system.At this time,interference problems and blocking problems also restrict the further improvement of system throughput.Therefore,how to effectively manage the interference and blocking issues has become one of the key challenges for the ultra-dense millimeter-wave D2 D communication system to further increase throughput.There are many anti-blocking transmission methods for millimeter-wave systems,such as anti-blocking methods based on reflection paths,anti-blocking methods based on multi-base station connections,and anti-blocking methods based on relay forwarding.Among them,the anti-blocking method based on the reflection path,for outdoor environments,lacks stable reflected waves,and the transmission attenuation of the non-line-of-sight path is large,so this method has a limited scope of application.The anti-blocking method based on multibase station connection is mainly aimed at cellular users and is not suitable for D2 D shortrange communication.The anti-blocking method based on relay forwarding realizes the antiblocking effect without completely changing the network topology or infrastructure.This paper studies the severe blocking and interference problems faced by dense millimeter-wave D2 D communication systems.The main research contents are as follows:(1)Establish a joint relay selection and resource optimization mathematical model: By analyzing the optimization relationship between "relay selection-resource allocation" and "interference intensity-available rate" of ultra-dense millimeter-wave D2 D communication networks,establish Rate-maximizing joint selection and resource optimization problems.It is expected that the problem of blocking the millimeter-wave signal of the system can be overcome by selecting an optimal relay,and at the same time,resource scheduling can reduce inter-link interference,increase the number of parallel transmission links,and ultimately improve the throughput performance of the system.In order to solve the established non-convex mixed integer programming problem with low complexity,we decompose it into two sub-problems of relay selection and resource allocation to solve.A bipartite graph-based relay selection method is proposed.We consider the D2 D user pair set and the relay set as two disjoint sets.The D2 D user pair and the relay are the vertices of two disjoint sets.The connection between them is called the edge of the bipartite graph.Each edge represents a set of matching relationships.We consider the relay-assisted D2 D link information rate under each set of matching relationships as the weight of the edges.The selection problem is transformed into an equivalent matching problem;the maximum matching method based on the Hungarian algorithm is used to obtain the optimal relay selection.(2)A bipartite graph-based relay selection method is proposed: by modeling D2 D user pair sets and relay sets as two disjoint sets,D2 D user pairs and relays are vertices of two disjoint sets.They The connecting lines are the edges of the bipartite graph,and each edge represents a set of matching relationships.The relay-assisted D2 D link information rate under each set of matching relationships is considered as the weight of the edges,and a weighted bipartite graph is constructed to transform the relay selection problem into an equivalent "one-to-one" weighted matching problem,thereby using The algorithm's maximum matching method gets the optimal relay selection.(3)A joint power optimization and time slot allocation method based on a constrained concave convex procedure(CCCP)is proposed: on the basis of obtaining the best relay,a joint power optimization and time slot scheduling algorithm based on CCCP is proposed to Increase the number of parallel transmission links,and further increase the speed.The original joint power optimization and slot allocation problem is still a non-convex problem,but it is observed that it has a special form of "log function-log function",which can be converted to a convex optimization problem using CCCP method to solve iteratively: The objective function is equivalent to the form of "convex function-convex function",and the convex function after the minus sign is linearly approximated to obtain the objective function after convex approximation,thereby obtaining the approximate convex optimization problem,and iteratively optimize it until convergence.Through a large number of simulations,the performance of the proposed algorithm under different network settings is evaluated,and it is compared with the anti-blocking algorithm based on relay selection and resource optimization under orthogonal channel allocation proposed in the literature,and based on randomization under non-orthogonal channel allocation.Following selection and resource optimization,the anti-blocking and antiinterference algorithms are compared.Simulation results show that,under the proposed nonorthogonal channel allocation,the combined anti-blocking and anti-interference algorithms of relay selection and resource optimization can effectively reduce inter-system interference,improve ultra-dense millimeter-wave D2 D communication systems and rates,and effectively reduce The total power of the ultra-dense millimeter wave D2 D communication system,and has good convergence.In addition,in the actual millimeter wave communication simulation environment with obstacles blocking,the proposed algorithm has better performance than the millimeter wave D2 D resource optimization algorithm without relay assistance,effectively reducing the impact of blocking on system throughput performance. |
PDF Full Text Request