Research On Channel Estimation In Massive MIMO System

Massive MIMO(Massive Multi-Input Multi-Output)is a hot topic in communication industry in recent years,it is a novel wireless communication technology based on the traditional MIMO technology.Massive MIMO can significantly improve the capacity and spectrum efficiency of communication system,it has become one of the key technologies of 5G(5th Generation)mobile communication system.These advantages of Massive MIMO are based on the accurate acquisition of channel state information(CSI),therefore,this work focuses on the factors affecting accurate CSI in Massive MIMO system.The factors that restrict system performance are different in millimeter-wave(mmWave)system and non-millimeter-wave system(cellular system),this work carries out the theoretical research and analysis respectively for the two communication systems.Firstly,multi-cell multi-user cellular Massive MIMO system is considered.In the TDD mode,the accurate CSI is not available to actual communication system due to pilot contamination,which has become one of the bottlenecks that restrict the system performance.Current research results are introduced from three directions:pilot allocation,channel estimation and precoding in this work.Many algorithms with complex operation or high time complexity are hard are difficult to apply in the complex multi-cell cellular Massive MIMO system,though they can theoretically eliminate pilot contamination completely.In this paper,a pilot allocation scheme based on Hungarian algorithm is proposed,which is simple and feasible.The simulation results show that compared with the traditional random pilot allocation strategy,the system can bring obvious performance gain.Next,single cell multi-user millimeter wave Massive MIMO system is consider.In the mm Wave massive MIMO system,the higher frequency of the mmWave band results in much smaller coherence time.Meanwhile,massive MIMO means a significant increase in the dimension of antennas processing,a much larger number of parameters will be needed to be tracked at the BSs for channel estimation,and the time required for pilot sequence scales up.Therefore,the heavy pilot overhead becomes the bottleneck for the mmWave massive MIMO system.A channel estimation algorithm based on alternating direction method of multipliers(ADMM)algorithm is proposed in this paper.The algorithm takes the low rank characteristics of the millimeter wave channel into account,and the channel estimation problem is modeled as the minimization of the kernel norm.The channel estimation matrix is obtained by optimizing the kernel norm.Finally,the performance of different algorithms is compared by simulation.The simulation results show that the ADMM algorithm requires less pilot overhead under the condition of certain channel estimation error,and the complexity of the algorithm is lower than the similar algorithm.