Research On Interference Management In MIMO Systems

The wireless channel is an open and shared channel.Due to the broadcast nature of wireless media,all nodes in the coverage of transmitters can overhear their signals.Therefore,interference is inevitable in multi-user wireless networks which becomes the most important factor of limiting the channel capacity.Recently,as two kinds of novel approaches to handle interference,interference alignment and interference neutralization technologies attain extensive attentions and are applied in various interference networks.The two methods enable the capacity of wireless interference networks to obtain breakthrough growth.Therefore,it is very important to investigate interference management technologies for wireless communication systems.Focusing on multiple antennas technology,aiming at system degrees of freedom(DoF)improvement,this paper mainly studies interference management technologies in the multi-cell multi-user multiple input and multiple input(MIMO)cellular interference network and the multi-user multiway relay channel.The main contents of this thesis are summarized as the following several aspects:(1)For the multi-cell multi-user MIMO cellular interference network with perfect channel state information at the transmitter(CSIT),we study the interference alignment scheme which can achieve the optimal DoF.First,we analyze the feasible condition of interference alignment in the considered network and obtain the Do F upper bound.Secondly,in order to achieve the DoF upper bound,we propose a fixed beamforming cluster interference alignment(FBC-IA)scheme to divide the beam vectors of transmitters into multiple clusters.In addition,we utilize the one-to-many interference alignment to accomplish interference alignment and elimination.The DoF results indicate that the FBC-IA scheme can transmit the maximum number of data streams by using minimum number of antennas,thus achieving the DoF upper bound.However,the FBC-IA scheme cannot be applied in arbitrary system configurations.To tackle this problem,we propose a dynamic beamforming cluster interference alignment(DBC-IA)scheme to adjust the dimension of beam vectors according to different system configurations.In addition,we utilize one-to-many and many-to-one interference alignment to accomplish interference alignment and elimination in arbitrary system configurations.Finally,from simulation results,we verify that the numerical results of the proposed two schemes agree with the theoretical results.Furthmore,we show that the proposed two schemes can provide significant DoF gain over the time division multiple access(TDMA)scheme and grouping interference alignment schemes.(2)For the multi-cell multi-user MIMO cellular interference network with two kinds of feedback delays,we study the interference management scheme based on feedback delay.First,we propose a retrospective interference neutralization(RIN)scheme for the two-cell multi-user MIMO cellular interference network with hybrid CSIT.The proposed RIN scheme utilizes the received interference signal and abandons the idea of traditional interference neutralization,i.e.,eliminating interference in the process of transmission.First,each base station(BS)receives the interference signal.Secondly,according to the design of the user transmission mechanism and the transmit beamforming matrix at each user,the interference signal of multiple time slots can be added up to zero.Therefore,the RIN scheme can utilize the feedback delay to eliminate the inter-cell interference.Subsequently,for the two-cell multi-user MIMO cellular interference network with output feedback,we propose a retrospective interference reconstruction(RIN)scheme to eliminate inter-cell interference of arbitrary number of cells.The proposed RIN scheme utilize the output feedback feature of containing multiple CSIT information to effectively eliminate inter-cell interference of arbitrary number of cells according to the design of user transmission mechanism and the received beamforming matrix at each BS.The simulation results show that the RIN and RIR scheme can attain a higher DoF than the TDMA scheme and the retrospective interference alignment(RIA)scheme.(3)For the two-cluster multi-user MIMO multiway relay channel,we study interference neutralization schemes without CSIT feedback.First,according to cut-set theorem,we obtain the DoF upper bound of the two-cluster two-user MIMO multiway relay channel with arbitrary number of antennas at each user.To approach the Do F upper bound,we propose a blind interference neutralization scheme without using CSIT.In addition,we analyze the DoF change according to different number of antennas at each user.Subsequently,we study the DoF of two-cluster K-user multiway relay channel and obtain the DoF upper bound.According to the joint design of user retransmission mechanism and beamforming matrix at each relay,we propose a distributed interference neutralization and retransmission(DINR)scheme to achieve the DoF upper bound.The simulation results show that the proposed two schemes can achieve the DoF upper bound and attain a higher DoF than the TDMA scheme and the signal space alignment scheme.