Research On Key Technology Of Interference Alignment In 5G

In order to meet the increasing need in the capacity of mobile communication system, researchers have proposed the concept of ultra-dense network. The ultra-dense network can improve the spectrum efficiency and energy efficiency by reducing the cell area, so that the system capacity is improved. Although ultra-dense network is a promising technology, we have to solve the problem induced by inter-cell interference, which is considered to be a difficult problem in ultra-dense networks. If the system does not eliminate the strong inter-cell interference resulted from small distance between base stations, the system performance will be degraded. Recently, interference alignment, with its advantage in eliminating strong interference, becomes a popularized approach to interference cancellation in ultra-dense networks.There are still many problems to be solved prior to the implementation of interference alignment in ultra-dense networks, although interference alignment technology can achieve significant performance in interference cancellation. The number of antennas and data streams limits the performance of interference alignment. In addition, interference alignment requires accurate channel state information in order to ensure an excellent performance. In this thesis, we propose a generalized interference alignment scheme, which is more applicable to solve the aforementioned problem than the global interference alignment. The work of this thesis is summarized as follows.Firstly, this thesis analyzes the limitation of global interference alignment in ultra-dense networks, and introduces a generalized interference alignment scheme that is more suitable for ultra-dense networks by making use of the difference between interference links. We study the feasibility of implementing generalized interference alignment scheme in the ultra-dense networks. Subsequently, this thesis proposes a high-performance interference selection strategy based on MMSE and a low-complexity interference selection strategy based on greedy algorithm. In order to further improve the performance, we then devise an interference alignment algorithm by using the equivalent noise. Moreover, this thesis obtains closed-form solutions of generalized interference alignment in some special scenarios. The simulation demonstrates that the performance of our proposed scheme is better than that of signal-to-leakage-and-noise ratio maximization algorithm.Secondly, to reduce the dependence of the generalized interference alignment on perfect channel state information, this thesis proposes a robust generalized interference alignment scheme, including the interference selection strategy and algorithm amenable to random error channel model. This scheme takes advantage of the statistical information of channel errors to reduce the influence of imperfect channel state information and enhances the applicability.