Research On Signal Interference Suppression And Performance Of The Massive MIMO Communication Systems

The explosion of traffic data volume and the extreme rise in number of wireless links make the contradiction between the limited wireless spectrum resource and the demands of people for wireless communication service become more prominent.On this background,the massive MIMO technology proposed by Thomas.L.Marzetta etc in 2010 attracts extensive attention.Previous studies have confirmed that the massive MIMO can improve spectrum efficiency greatly by fully exploiting spatial degrees of freedom,at the same time the energy efficiency has several orders of magnitude of room for improvement compared with the current wireless communication systems.Therefore,the massive MIMO technology becomes one of the most promising technologies in the 5G(the Fifth Generation)mobile communication systems.However,the signal interference,especially the pilot interference among adjacent cells becomes the bottleneck of system performance improvement in massive MIMO multiple cells systems.Consequently,this thesis mainly researches the pilot signals interference of massive MIMO systems and proposes several pilot decontamination schemes in order to enhance the performance of the massive MIMO systems.The main work of this thesis is organized as follows:1.Scheme of smart pilots assignment for users based on water-filling algorithmThe number of orthogonal pilots,which are used to estimate channels,is limited by the coherence interval,so these orthogonal pilots are totally reused in different cells.The reused non-orthogonal pilots in adjacent cells cause severe signals interference-pilot contamination,which affects the accuracy of channel estimation and becomes a bottleneck to the improvement of system performance.However,the qualities of channels for different users are different and the levels of interference to different pilots are also variable.Therefore,we propose a smart pilots assignment scheme based on water-filling algorithm.In the proposed scheme,pilots that suffer from severe interference are assigned to the users with inferior channels and pilots that are slightly interfered are assigned to users with superior channels.Research and simulation verify that the proposed scheme is able to enhance the system performance significantly by sacrificing the performance of few users with inferior channels.2.Fast smart pilots assignment scheme based on particle swarm optimizationIn the massive MIMO multiple cells systems,smart pilot assignments play an important role in relieving the pilot contamination of the massive MIMO systems.However,the computational complexity of smart pilot assignment grows fast as the number of terminals increases.For example,when the number of users in one cell is 10 and the number of orthogonal pilots is also 10,there will be 10!=3628800 approaches of pilot assignment.In order to decrease the computational complexity of smart pilot assignment,we propose a smart pilot assignment scheme based on the particle swarm optimization(PSO)algorithm.In the proposed PSO scheme,the information among particles is shared timely,which helps to find out the optimal matches between users and pilots fast.Experiments demonstrate that the smart pilot assignment based on PSO can decrease the computational complexity significantly and reduce pilot contamination considerably.3.Scheme of terminals grouping and asynchronous pilots transmissionThe pilot contamination,which results from the non-orthogonal pilots reuse in different cells,leads to different side-effects to the users that locate in the cell-center area and in the cell-edge area.So we propose a pilot decontamination scheme based on terminals grouping and asynchronous pilot transmission to decrease pilot contamination.In the scheme,users are divided into cell-center users group and cell-edge users group according to the large-scale fading.And the cell-center users in all cells that suffer from slight interference send pilots simultaneously,while the more severely interfering cell-edge users in different cells send pilot in non-overlapped time slots.Simulation verifies that the scheme can decrease pilot contamination effectively for the massive MIMO multiple cells systems.However,in the scheme of terminals grouping and asynchronous pilot transmission,the cell-edge user in one cell may reuse non-orthogonal pilot with the cell-center user in other cells.Thus pilot contamination still exists between cell-edge user in one cell and cell-center user in other cells.In order to reduce such pilot contamination,this thesis proposed an improved users grouping and uplink pilots and downlink data aligned transmission for different groups of users.In the improved scheme,pilot transmission of one users group aligns to forward-link data transmission of another user group.Pilot contamination of the systems can be furthermore reduced by exploiting the irrelevance of reverse-link pilot signals and forward-link data signals between different users groups.The simulation results verify that the improved scheme performs better than the former scheme in pilot decontamination and system performance enhancement.4.Scheme of cells partition and pilots asynchronous transmission and pilot power controlIn massive MIMO multiple cells systems,cells are partitioned into several groups and users in different cell groups send pilots asynchronously and orderly.Therefore,pilot contamination is compressed into the cells in the same cell group.Apparently,the more the cell groups are,the less the pilot contamination is.But when the cells grouping factor is excessive,the reverse-link data transmission will be suppressed in a coherent interval leading to performance degeneration for the whole systems according to data transmission process in the proposed scheme.So the reasonable cells partition factor plays a key role in the scheme.We consider system performance from a comprehensive perspective that both the reverse-link and forward-link directions are taken into consideration to search for the optimal cell grouping factor.After exploration we get the reasonable cells partition factor and find out the balance point between the pilot decontamination and the whole system performance.Simulation results verify that the cell grouping factor that we obtained is effective in enhancing system performance and pilot decontamination.Additionally,pilot contamination still exists between the cells in the same cell group in the former scheme,so we introduce pilot power control to the scheme in order to reduce the effects of pilot contamination between cells in the same cell group.Simulation results verify the advantage of the proposed scheme in pilot decontamination for massive MIMO systems.