Application Reasearch For Interference Alignment In Modern Communications

Recently, with the rapid development of wireless communication, interference management issue has become a major challenge in modern wireless communication systems. The scholars of information theory recently proposed a new technology, which is interference alignment (IA). The core idea of IA is providing more interference-free data transmission signal dimensions for the desired signal by compressing the signal dimensions of interference at the transmitters. The new technology enables the capacity of wireless interference network to obtain breakthrough growth. Hence, it is of great significance to research the IA algorithm for wireless communication systems.In all the nerwork scenarios that studied in IA, there are two interference network that closely related to the actual wireless communication system, which are K-user MIMO interference channel (IC) and multi-cell multi-user MIMO broadcast interference channel (IBC). Therefore, the two interferece models are main objects of this paper. The main contents of this paper are shown as follows.1、For the K-user M x N MIMO IC with time-varying channel, suppose M and N are the transmitting and the receiving antennas, ergodic interference alignment (ergodic IA), an achevable scheme is proposed and obtains KMN/M+N degrees of freedom (DoF). The ergodic IA scheme for K-user MIMO IC only adopts simple linear beamforming and finite symbols, and the achievable DoF is equal to the optimal DoF upper bound in theory. Then, since one of the biggest limiting factor of ergodic IA is that the system waitting time is too long, the mathematical expression of the system time latency is derived. The expression of the time latency will be useful to quantitatively analyze for the time latency of ergodic IA scheme. Finally, the expression of system achievable sum rate is derived. The simulation results show that the achievalbe sum rate increases with any one of the system parameters K、M and N, which is consistent with the conclusion of the DoF.2、For the G-cell K-user M × N MIMO IBC with time-varing channel, suppose M and N are the transmitting and the receiving antennas, respectively. Based on ergodic IA, an achevable scheme is proposed and obtains GKMN/M+KN DoF. The ergodic IA scheme for MIMO IBC also only adopts simple linear beamforming and finite symbols. The achievable DoF is equal to the optimal DoF upper bound in theory. In the processing of the scheme design, the scheme for the M × 1 MISO IBC is firstly designed and obtains GKMD/M+K oF. Then, the M × 1 MISO IBC scenario is extended to M × N MIMO IBC. By conforming the coperations of copying and extension on the super-symbol of MISO IBC ergodic IA scheme, the achievable scheme for MIMO IBC is designed and obtains GKMN/ DoF. M+KN3、For the ergodic IA scheme for the G-cell K-user M × N MIMO IBC, the scheme performances are analyzed in this paper. Firstly, based on the channel quantitative method, the mathematical expression of the system delay is derived. The expression of the time latency will be useful to quantitatively analyze for the time latency of ergodic IA scheme of MIMO IBC. Then, from two aspects of channel coefficient and super-symbol sequence, an optimization scheme is proposed and the exprssion of the optimized time delay is derived. The expression shows that the optimized time delay is greatly decreased. Finally, the achievable system sum rate of the ergodic IA scheme of MIMO IBC is derived and the simulation results are given. The simulations show that the achievable system sum rate increases with the any one of the system parameters G、 K、M and N. Specifically, for the influence of the system sum rate, the transmitting antennas M is bigger than the receiving antennas N. The analysis for the achievable sum rate is same with the one of DoF.4、For the MIMO interference channel, when the transmitters have no channel state information (CSI), the achievable DoF and the corresponding achievable scheme are studied in this paper. Firstly, for the K-user M × N MIMO IC, by using blind interference alignment (blind IA), the achievable DoF is KMN/M+N. The simulation shows that, when min(M,N)/max(M,N)≥K-1/K(K-2), the achievable DoF is same with the the optimal DoF upper bound. Again, on the condition of no CSI at transmiiters (CSIT), the achievable DoF in this paper is superior to the one that obtained by previous scholars. Then, for the G-cell K-user M × N MIMO IBC, by using blind IA, the achievable DoF of this paper is GKMN/M+KN The achievable DoF is consistent with the DoF upper bound of MIMO IBC. Nevertheless, on the condition of no CSIT, the achievable DoF of bind IA is worse than the one concluded by the previous researchers. There is a phenomenon called Loss DoF. The reason is that the conclusion of the previous researchers is obtained on the condition that the base stations can share information with each other. Once the condition is lost, wheather the conclusion of the previous scholars is effective is not known. Hence, the achievable DoF of this paper has certain reference significance for the MIMO IBC when there is no CSIT.