Study On Pilot Contamination Reduction In Massive MIMO Systems

In 2010,Bell Labs proposed the base station equipped with a large number of antennas,which formed the Massive MIMO systems through sending and receiving data simultaneously.The Massive MIMO technology improves the system's performance in both energy efficiency and spectrum efficiency,and becomes a kind of candidate key technology for 5G.In the case that multi-cell shares the same pilot with users,pilot contamination becomes the main problem in the Massive MIMO systems that restricts the systems performance.Pilot contamination reduction technology is studied for the Massive MIMO systems in this thesis.The main research work includes the following three aspects:1.The thesis Introduces the development of the Massive MIMO systems and summary the status of pilot contamination.Then,describe the Massive MIMO systems model in TDD mode,and analyze the generation of pilot contamination in the uplink channel.At the same time,it is derived that pilot contamination seriously affects the performance of large-scale MIMO system in the downlink channel.At last,this thesis summaries and analyzes pilot contamination reduction techniques.2.The traditional channel estimation scheme does not consider pilot contamination.Based on this,an improved DFT channel estimation scheme is proposed.Considering that in the Massive MIMO-OFDM systems,the channel impulse response(CIR)of the effective signal is focused on the front point in the time-domain channel impulse response.Through the DFT and IDFT transformation,contamination signals that beyond the effective signals is taken as samples,which can be used to generate a threshold.This decision threshold filters the effective CIR,eliminates the interference caused by pilot contamination and improves the performance of channel estimation.3.5G network will encounter lots of connections.The previous scheme in which randomly assigned pilot is no longer applicable,new pilot allocation schemes need to be proposed to adapt to the different scenarios.In this thesis,two pilot assignment schemes are proposed through the mentioned anti-pilot contamination factor;the pilot allocation algorithm based on the maximum systems rate and the pilot allocation algorithm based on user fairness,the threshold is selected in the actual scene.Compared with the previous scheme,the simulation results show that the proposed method further improves the performance in the channel environment of the Massive MIMO systems.This thesis deeply studies the pilot contamination in the Massive MIMO systems,and obtains some research results.The study has theoretical value and practical meaning for pilot contamination in future.