| In recent years, with the development of multimedia service and the intelligent mobile terminals,wireless data traffic continue to grow. In order to meet the demand for higher data rates, all kinds of technology that improve the efficiency of spectrum have been great developed. MIMO(Multiple-Input, Multiple-Output) technology is being paid more and more attention because of its high spatial multiplexing gain. At the same time, reuse factor of 1 is used in the multi-cell network to improve the spectrum efficiency. In this way, the cell edge users inevitably receive the interference from the other base stations and the other antennas, after that the desired signal can not be obtained effectively. Because of these, interference has been the main factor to limit the system capacity. IA(Interference Alignment) as a special form of interference coordination has become one of the research hotpots.Interference Alignment subverts the traditional idea that the channel of MIMO system is limited by interference and a new method to get higher transmit rate is given in the situation of limited spectrum. Different from the traditional method, interference from different base stations are compressed into same space rather than different space which orthogonal with each other, so that each base station could get more Do F(degree of freedom) to transmit desired information. The research of the thesis focuses on how to achieve both the interference alignment and rate optimization.The existing interference alignment algorithms are analyzed firstly. Minimum interference leakage IA algorithm’s main idea is to make the power of interference signal least at each user without considering the channel status of useful signal has experienced. Rank constrained rank minimization interference alignment compress the interference into the smallest rank while each base station obtain the maximum transmission degrees of freedom, however, these algorithms can not achieve optimal sum rate. Optimization algorithm proposed in this paper can do all of these by searching pre-coding along the gradient ascent direction of throughput. In every step, The step length is decreased with multiplying an parameter smaller then one. In final, the proposed algorithm degrades to the rank constrained rank minimization IA algorithm. Thus, this proposed algorithm in this paper can achieve interference alignment at endand getting the throughput improvement. Compared with the existing algorithm of interference alignment, this optimization algorithm can improve the system sum rate obviously. |
