Research On Limited Feedback Precoding Algorithm Of Multi-user Downlink MIMO System

As a new technology, multiple-input multiple-output (MIMO) wireless communicationtechnology can improve the data transfer rate and at the same time will not increasefrequency band and transmit power. It is an important part of the fourth generation of mobilecommunication technology. MIMO system basing on the original time dimension can obtainspatial multiplexing gain and spatial diversity gain by using multiple antennas. MIMO systemcan utilize multiple transmission paths caused by multiple antennas to obtain spatial diversitygain; on the other hand, MIMO technology provides high spectral efficiency. Therefore, thetechnology of precoding, user selection is very important in the design of technical MIMOsystem.In this paper, we analyze technology of limited feedback precoding and user selectionunder the downlink multi-user MIMO wireless communication system. Based on the coreidea of threshold selection, we propose a mechanism for optimizing the threshold value, bysetting a threshold value, only the channel conditions of the user which is better than thethreshold value will be considered. If the users’ channel conditions are not met the thresholdrequirements, the users remain silent without accessing bandwidth resources. Optimizationmethod can achieve a high threshold value and the feedback rate while increasing thefeedback rate.Meanwhile, the paper has researched the multi-user MIMO downlink feedback ofinformation to reduce the overhead, and propose a mechanism of suboptimal single thresholdand low feedback overhead transmission strategy for each user equipped with multi-antennacase. Using the proposed strategy, each mobile station performs channel estimationtechniques, by optimizing the number of optimal feedback users to determine the thresholdvalue, according to linear relationship to get average number of users. Applying simplemethod to get the final sub-optimal threshold. The sub-optimal and optimal thresholdalgorithm is compared. Simulation results show that sub-optimal threshold algorithmsimplifies the expression of the threshold value, barely increases the amount of feedback, andcan achieve high system capacity with fewer users. Simulation results show that thesub-optimal threshold performance is close to optimal threshold performance.