Multiple Channel Parameters Estimation And Compensation For Distributed MIMO System In Time-varying Channel

As the core technology of Long Term Revolution(LTE), Multiple Input Multiple Output(MIMO) can be effective in providing diversity gain and power gain. With further research in MIMO, the distributed MIMO system from the perspective of cell division has been proved to be the prominent direction of MIMO technology. The distributed system is becoming more and more attractive because of its advantages of higher capacity, wider coverage, lower power consumption and easier extension. And it has been one of the key technologies of future mobile communication.In distributed MIMO system, transmitting antennas in different positions use different oscillators, leading to multiple frequency offsets estimation which the centralized MIMO system never encounters. The rapid development of high-speed railway urgently needs new wireless and mobile communication technology under time-varying environments. Therefore, the research in multiply frequency offsets estimation of distributed MIMO systems under quasi-static and time-varying environment has important significance.Single Input Single Output(SISO) system and centralized MIMO system can easily compensate the frequency offset at the receive side. However, the multiple frequency offsets problem of distributed MIMO system makes it difficult to compensate the frequency offsets effectively at receive side. Therefore, it’s very necessary to research in multiple frequency offsets pre-correction based on precoding.Chapter 2 researches in multiple parameters estimation of distributed MIMO system under quasi-static environment. Firstly, we get the Maximum Likelihood(ML) model and Cramer-Rao Low Bound(CRLB) of distributed MIMO system through theoretical derivation. And then, we study the correlation-based and the Expectation Maximization(EM) multiple frequency offsets estimation algorithms. Finally, we compare the performance difference between Expectation Conditional Maximization(ECM) and Space-Alternating Generalized Expectation-maximization(SAGE) algorithm. Simulation results indicate that the EM can well overcome the multiple antenna interference(MAI) in higher Signal to Noise Ratio(SNR), achieving good estimation performance; SAGE algorithm can achieve convergence more quickly than ECM algorithm.In the third chapter, multiple parameters estimation of distributed MIMO system in time-varying channel is studied. Firstly, we research in the ML estimation model in time-varying channel. Secondly, we promote the ECM algorithm under quasi-static environment to the time-varying environment, completing the theoretical derivation. We also research in SAGE algorithm for improving the convergence rate of ECM. Simulation analysis proves that the estimation algorithm in the chapter can well compensate for the performance loss caused by the time variability and significantly improve the performance of multiple parameters estimation.In the fourth chapter, joint multiple frequency offsets and antenna gain pre-correction is studied in the distributed MIMO system under time-varying channel. Firstly, we research in multiple frequency offsets pre-correction based on precoding technology. And then, we research in antenna calibration of Single Input Single Output(SISO) system. Based on this, the antenna calibration model and calibration process are studied. Thridly, we take advantage of equivalent channel matrix constituted by channel and frequency information to achieve joint multiple frequency offsets and antenna gain pre-correction. Simulation results indicate that the antenna calibration based on channel prediction can well compensate the performance loss caused by time variability; joint frequency offsets and antenna gain pre-correction can complete frequency offsets pre-correction and antenna calibration simultaneously without increasing the amount of feedback.Summary, the multiple parameters estimation and compensation under time-varying channel is one of the key technologies of distributed MIMO systems. In this paper, we study multiple parameters estimation and compensation in depth. This paper has some theoretical study and application value.