The Research On The Precoding Technology Of Multi-user MIMO System

Using multiple antennas of the transmitter and the receiver, MIMO makes the communicationsystem capacity doubled, and improves the spectrum efficiency and system performance within alimited bandwidth. Therefore, MIMO has been the rapid development and wide applications,especially in the downlink of the wireless communication system, but there are also some problemsto be solved in MIMO systems, for example, the receiver is difficult to take effective interferencesuppression techniques to suppress the interference from the undesired transmitters in the downlinkof the multi-user MIMO system. However, by designing effective precoding and reasonable powerallocation schemes at the transmitter, this interference can be suppressed effectively and theperformance of systems can be enhanced. Thus, the precoding technology has become one of thekey technologies of interference suppression in MIMO system.Based on the above views, this thesis studies the precoding and power allocation algorithms inmulti-user and multi-cell MIMO system. Firstly, this thesis uses a maximizing SLNR precoding andoptimal power allocation algorithm to reduce co-channel interference in multi-user MIMO system.Simulation results show that the algorithm has a better performance than the traditional BD andMMSE algorithm, and even close to the SLNR iterative algorithm, what’s more, the proposedalgorithm has a significantly lower computational complexity. Secondly, an approach whicheliminates the interference generated by multiple data streams of users in the multi-user MIMOsystem by twice-precoding algorithm is presented. The first BD algorithm completely eliminateco-channel interference among users, and the second ZF or MMSE algorithm is used to eliminatethe data stream interference. Simulation results show that the algorithm has a significantly betterperformance than the method which uses BD algorithm only once. Finally, to suppress the inter-cellinterference in multi-cell MIMO system, this thesis proposes an iterative algorithm which is basedon maximizing SLNR and minimizing interference signals. Within BD algorithm, it eliminatesco-channel interference among users in each cell, and then the iterative optimization is used toreduce inter-cell interference. Simulation results show that the algorithm is superior in performanceto BD algorithm with optimal power allocation or SLNR algorithm based on the average powerallocation.