Research On Channel Estimation For Millimeter Wave Massive MIMO System

Research on the next generation (5G) cellular communication systems is being carried out on a global scale. The data rate requirements of the next generation (5G) cellular communication systems will approach or surpass gigabits per second. Such a high data rate inevitably relies on an enormous available bandwidth. The huge bandwidth available at millimeter wave frequencies makes it one of the best candidates for future 5G cellular systems. However, extreme path fading restricts the propagation distance of millimeter wave and directional beam-forming with large antenna arrays appears to be inevitable to support longer outdoor links. Millimeter wave Massive MIMO technology has been considered one of the key technologies in the future 5G. Under this background, in this paper, we focus on the channel estimation of millimeter wave Massive MIMO systems.We first introduce the characteristics of Massive MIMO system, and summarize the basic knowledge of Massive MIMO system. Then the development of the millimeter wave Massive MIMO system model is introduced, which experience the pure digital structure, the pure analog structure and the present analog-digital hybrid structure. And then the characteristics and model of the millimeter wave channel are presented,which is the basis for the follow-up study of the channel estimation of the millimeter wave MIMO Massive system.In the third chapter, we research the channel estimation problem for single user millimeter wave Massive MIMO system. We propose an adaptive channel estimation strategy for the analog-digital hybrid structure of millimeter wave Massive- MIMO system. Firstly, the 2D-DFT based method is used to estimate the channel with a short training sequence. Then, we designs a feedback scheme to adjusts the training overhead adaptively with the change of channel quality for millimeter wave Massive MIMO systems. The key threshold in the feedback scheme is derived and its influence on the accuracy of the estimation results is analyzed. Simulation results confirm that the proposed algorithm can adjust the length of the training sequence adaptively according to the current channel condition maintaining the stability of estimation accuracy.In the fourth chapter, we proposed the NR-DFT algorithm based on the DFT-CEA method, and achieve a better estimation accuracy. Firstly,the DFT-CEA algorithm is used to estimate several paths, and the variance of corresponding submatrix and standard submatrix are calculated. And we use the threshold in the third chapter to judge all the paths and select the proper paths by the variance. Simulation results show that, NR-DFT algorithm has better estimation accuracy than the DFT-CEA algorithm under the same conditions.In the fifth chapter, we summarize our work in the channel estimation for the millimeter wave Massive MIMO systems, and analyze the shortcomings of the paper and the places that can be further studied.And lastly, the outlook of the millimeter wave Massive MIMO systems in the next generation (5G) cellular communication systems is presented.