Decoding Method Based On ICA For Massive Array-antenna MIMO Systems

So far,the research on the fifth generation(5G)mobile communication is brewing.5G mobile communication technology has many breakthroughs and advantages relative to 4G.Meanwhile,the large-scale MIMO technology,which is called the core technology of 5G,is an important field of the research on the key technology of 5G.It occupies a key position in the 5G technology,which can significantly improve the capacity of the system,while reducing energy consumption.For a particular base station,the estimate of each channel in a large-scale MIMO communication system is a linear change in the result of the multiplication of the pilot training sequence transmitted by the user and the received signal in the corresponding base station.However,the base station can’t tell whether the received signal belongs to desired cell or neighboring cell.So the performance of the large-scale MIMO system will be restricted,this is what we called pilot contamination.In reality,all users in the same cell use orthogonal pilot training sequences,while the pilot training sequences used by users in neighboring cells are reused to the target cell.The channel information estimated between the local user and the base station and the channel information estimated at the base station side are different.That is because the pilot training sequence transmitted by the user in the neighboring cell will lead to the interference which is mixed in the desired estimation.In the system based on large-scale linear array antenna,a blind decoding algorithm based on DOA estimation and ICA is proposed.The proposed blind decoding algorithm uses the independent component analysis algorithm to separate and decode the received signal.Meanwhile estimating the channel by this algorithm could be obtained without the pilot training sequence.Compared with the large-scale independent antenna,in the large-scale linear array antenna system,the estimated channel information includes the direction of arrival.The direction of arrival information is used to identify each user without the pilot sequence for distinguishing the users.The blind decoding algorithm based on the linear array antenna utilizes the coherent spatial gain of the large-scale array antenna,and improves the decoding performance.In the system based on large-scale combined array antenna,a blind decoding algorithm is proposed to use channel energy to differentiate different cells and use DOA estimation to identify different users.It makes full use of the directional information contained in the combined array antenna system.The channel decoding algorithm based on combined array antenna model has better decoding performance under certain form of antennas.Because the user in the neighboring cell can be identified by channel energy,even if the total number of users is large and the angle interval between users is small,the proposed algorithm in this paper still has high channel estimation accuracy and excellent decoding performance.