Research Of MIMO Detection Algorithm In LTE-A Uplink

This thesis first introduced the development process of mobile communication,and some important changes from Release 8 to Release 12 in LTE/LTE-A system.Then we analysed the characteristic of wireless channel,and presented the channel model in link simulation.Then we introduced some konwledge in physicical layer,such as subframe structure,physicial channel,physicial signal,the OFDM and MIMO techniques.The key point of this thesis were the analysis and inprovement about noise power estimation algorithm and detection algorithm.Noise power was an important part of SNR calculating,which would be used in detection algorithm,link adaption and Turbo decoding.Traditional noise power estimation algorithm would let out singal power and would make the estimated result too large.A new method was proposed in this thesis,which used second-order difference algorithm to restrain the leaked singal power and made the result more accurate.For detecting algorithm,we analyzed the principles of classical algorithms and the applicabilities for uplink.To solve the problem of error propagation cased by serial interference cancellation algorithm,we used the detected information to replace hard decision information.This would restrain error propagation to a certain extent,and make better performance.At last,we built an uplink simulation platform with Matlab.Then we simulated noise power estimation algorithm and detection algorithm in various conditions.It was proved that the new noise power estimation algorithm we proposed got smaller minimum squared error,and more close to the ideal value.The improved serial interference cancellation algorithm also showed better performance than the original.