Research On Key Technologies Of Channel Estimation For 5G Mobile Communication System

With the increasing demand for wireless communication services,the fifth generation of mobile com-munication system research is being launched in an all-round way.Since channel state information is both used in the forward precoding and the signal detection,the quality of channel estimation is critical for the goal to achieve high spectral efficiency in the the 5th Generation(5G)communication system.In this paper,the channel estimation technology in 5G communication system has been studied.Firstly,the principle of orthogonal frequency division multiplexing(OFDM)technology and multiple-input multiple-output(MIMO)technology are introduced.The basic channel estimation techniques based on training sequences in traditional MIMO OFDM systems are summarized.In addition,the problem of severe sub-carrier interference in OFDM systems under time-varying channel and the problem of reciprocity calibration in time division duplex(TDD)large-scale MIMO systems are analyzed.In actual communication system,these above problems will affect the quality of channel estimation.Secondly,the channel estimation method of OFDM system in high speed mobile scene is studied.In high-speed mobile environment,OFDM system faces the problem of destruction of the orthogonality among subcarriers,which leads to the increasement of the parameters of channel estimation.In order to solve this problems,the method of fast time-varying channel estimation based on base-expansion model is introduced,and the the influence of different channel characteristics,different base-expansion models and different model orders on channel estimation quality is analyzed.Moreover,for further improvement of the spectrum efficien-cy,a channel estimation technique combined with Kalman filter and decision-directed tracking is proposed,that is,the prediction result is corrected by decision-directed with the help of Kalman filter’s prediction.Sim-ulation results show that the proposed algorithm can effectively reduce the overhead of pilot.Finally,channel estimation technique in the 5G experimental platform is introduced,including the design of reference signal and the comparison of interpolation algorithm.The performance of different interpolation algorithms is evaluated by simulation.Based on block diagonalization(BD)precoding used in the experimen-tal platform,the performance under partial calibration and full calibration is analyzed.The simulation results show that the full calibration algorithm can effectively improve the system’s spectrum efficiency.