| The fifth generation(5G)needs to satisfy the requirements of ultra-high capacity,massive connections,ultra-high energy efficiency,and ultra-low latency for 2020 and beyond.Ultra-dense network(UDN)is one of the key technologies of 5G.UDN can achieve several hundred fold of throughput improvement in the hot spot areas,because it can shorten the distance between the access points and the users by the dense deployment.However,the dense deployment of the access points,overlapping,and the heterogeneous characteristic of UDN,may cause more complexed interference in UDN,compared with traditional wireless networks.Therefore,the interference management is one of the key technologies to improve the spetral efficiency of the system and the user experience.The coordination in power domain can mitigate the interference and improve the system throughput.Furthermore,there may be many users who use the same time-frequency resource to access simultaneously,which may cause the severe interference.Superposition transmission can be used in order to mitigate interference,further improve the spetral efficiency,and support more users to transmit at the same time.Thus the superposition transmission is expected as one of the key technologies in the interference management of UDN.This thesis begins with the above two directions,and studies the power allocation scheme and the superposition transmission scheme in UDN.The main contributions of this thesis are as follows:This thesis made an assumption about the interference,and developed the power allocation schemes based on the water filling algorithm.Then the iterative water filling(IWF)based power allocation scheme and the low-complexity water filling(LCWF)based power allocation scheme were proposed.Simulation results showed that both IWF algorithm and LCWF algorithm can significantly improve the sum rate,compared with the average power allocation.Furthermore,LCWF algorithm can achieve the close performance as the IWF algorithm with lower complexity.Pattern division multiple access(PDMA),which is one of the superposition transmission schemes,was applied to UDN in this thesis.Multi-users’ signals were superposed by superposition coding at the transmitter side,and were separated from each other by the successive interference cancelation detector at the receiver side.Based on this framework,this thesis proposed a joint pattern and power allocation(JPPA)scheme to maximize the sum rate.The JPPA scheme was based on the water filling algorithm,and could converge to the optimum pattern and power allocation with several iterations.Simulation results show that the proposed JPPA scheme can improve the sum rate,so that can improve the user experience in the hot spot areas.In order to increase the number of access users in the hot spot areas and improve the system capacity,the enhanced superposition transmission scheme was studied in this thesis.Interleaver based pattern division multiple access(IPDMA)was proposed by bringing the random interleavers to PDMA.Also,the simulation platform was built,and simulation results showed that IPDMA can acquire higher spetral efficiency and support the superposition transmission of more users,compared to PDMA.Thus IPDMA can better satisfy the requirement of spetral efficiency in the hot spot areas. |
PDF Full Text Request