Research And Implementation Channel Estimation Algorithm Of MIMO-OFDM System

This paper focuses on channel estimation for MIMO-OFDM system,and the full text is divided into the following three parts:In the first part,the MIMO-OFDM system's basic principle and the characteristics of wireless channel,which includes the large-scale fluctuation and the small-scale fluctuation,are studied,firstly.Then,the statistics characteristics of Zheng&Xiao channel model is studied,and according to the analysis of the model,it is selected as the basis of the research of this paper,lastly.In the second part,the channel estimation algorithms for MIMO-OFDM system are studied.Firstly,the principle of some pilot-assisted channel estimation algorithms is studied,including the LS,LMMSE,SVD-LMMSE and several interpolation algorithms,usually used for channel estimation.Then,the estimation algorithm based on EM(Expectation Maximum,EM)algorithm is studied.Finally,the MIMO-OFDM system simulation platform is built up,in which the antenna configuration type is 4×4.And the estimation algorithms above for the situation that the pilot interval is set to 6,are simulated.Meanwhile,simulating the EM channel estimation in the case,where the pilot interval is 4.Lastly,the simulation results are compared and analyzed.And the simulation results show that in the principle,the EM method yields the same MSE as the case that the pilot interval is 6,but the number of used pilots is smaller.In the third part,the channel estimation algorithm is implemented on the FPGA platform.Considering the complexity and spectrum utilization,this paper chooses the three iterative EM algorithm to implement,in which the pilot interval is 6.In the process of implementation,the diagram of channel estimation module is designed,and the functional sub-modules are designed,too.Then,every sub-module is implemented on the FPGA platform,so the whole channel estimation module is completely implemented.Finally,by comparing the FPGA simulation results with the MATLAB simulation,the error between them is given to verify that the logic function of the EM channel estimation module is right.And by the report of resource consumption,the logic feasibility of the module is verified.