LTE Downlink Precoding Technology Investigation And Realization

Mobile communication systems are aiming to achieve higher spectral efficiency in a certain bandwidth in order to support higher data rate. MIMO is one of effective methods to achieve this aim. But as a result, MIMO can introduce multi-antenna interference which is harmful for the system. Precoding can solve this problem. On the one hand, precoding can promote the capacity by eliminating interference between multi-antennas or multi-users. On the other hand, it can reduce the complexity of receivers by simplifying the signal process.The main work of this thesis is to analyze and simulate some precoding algorithms by comparing their performance and complexity. Then, an adaptive algorithm is given according to the characters of LTE system. First of all, some background knowledges of MIMO are introduced, including the system model and the capacity of MIMO. Also, some detection algorithms such as MLE, ZF and MMSE are introduced. Then, some linear and nolinear precoding methods are introduced. Then, the paper describes the work flow of LTE physical layer and limited feedback precoding, including codebook designation and selection methods of the limited feedback precoding. At last, an adaptive precoding method and an optimized codebook selection algorithm are explored. The results show that the new method can reduce the complexity fifty percent compared with traditional methods and the adaptive method can promote frequency efficiency dramatically.In this paper, the downlink precoding for LTE system is implemented on the multi-core DSP platform, including code selection module and the pre-coding module. By comparing different algorithms, an optimum method is selected. All the modules have been tested and can meet the requirements of LTE system, so they have a certain reference value for LTE system implementation.