Research On Resource Management Of D2D Assisted Relay System

In recent years,with the continuous development of mobile communications and the growth of mobile services,the advent of the fifth-generation mobile communications(Fifth-generation,5G)has brought great opportunities.How to improve the user experience and reduce the system power consumption is important to the future mobile network.D2D(Device-to-device)communication technology makes it possible for users with close proximity to communicate directly without connecting to the base station,which has a broad prospect in improving the throughput of the system and reducing the user's power consumption.The D2D assisted relay system,which is proposed based on D2D communication technology,can greatly improve system performance with users acting as relays to assist the communication of others.And the key technology,which is UE-to-Network Relay,can effectively extend the system coverage and reduce the user's power consumption by enabling users with good channel quality to assist other users'communication with the base station.At present,the D2D assisted relay system has been widely discussed in industry and academia.And this thesis carries out researches on resource management technology in D2D assisted relay system,aiming at improving system throughput,reducing system power consumption and guaranteeing the service quality of users.The main works of this thesis are as follows:For the D2D assisted relay uplink transmission scenario,the research is performed considering that UEs acting as relays has their own traffic demand.The research focus on how to reduce system energy consumption while guaranteeing the basic service quality of users.Through the mathematical modeling and analysis of the problem,this thesis uses the matching algorithm and the concave-convex procedure to determine the user's transmission mode together with the relay selection and power allocation respectively.And a joined resource management strategy is proposed to reach the aim.Compared with the existing works,it is verified that the proposed algorithm can improve the system performance effectively.This thesis studies a discovery enhancement strategy for relay discovery under the architecture of UE-to-Network Relay.Because of the limited bandwidth capability of some users in the UE-to-Network Relay system,the relay discovery scheme under the traditional UE-to-Network relay architecture is no longer applicable.Through the research and analysis,two relay discovery enhancement strategies are put forward.One is to use the user's geographical location information to effectively coordinate the resource used for relay discovery,which can effectively reduce the user power consumption and enhance the discovery efficiency.And the other strategy utilizes the special relation between the users,which can improve the discovery efficiency through the coordination among users.By comparing with the existing schemes,the proposed scheme can improve the relay discovery efficiency remarkably.In this thesis,a path switch enhancement strategy is designed for UE-to-Network Relay path switching scenarios.Considering that the user's behavior is not defined when a user meets the trigger event of cell handover and relay re-selection at the same time,a scheme is proposed to effectively identify the switching scene and make a better decision.Through the path switch measurement configuration,the user can effectively reduce the energy consumption in the measurement.Under the defined path switch trigger event and the path switch signaling procedure,the user can make more reasonable choice under different path switching scene to efficiently enhance the user experience,reach the balance of system signaling overhead,latency and throughput.Through the comparation with the existing schemes,the effectiveness of the proposed scheme can be verified.To sum up,this thesis studies the resource management of D2D assisted relay system from both theoretical analysis and signaling process point of view.And this thesis also puts out simulation and evaluation results.