Research On Massive MIMO Transmission Technology For Wireless Communication Systems

With the exponential growth of wireless data traffic and user terminal, a new generation of mobile com-munication systems places greater demands on key indicators, such as data rates, spectrum efficiency, cov-erage, and so on. Although the massive multiple-input multiple-output (MIMO) transmission technology, as one of the key technologies of the next generation of mobile communication systems, has a lot of advantages, such as serving more users at the same time frequency resource, owning higher data rate and lower power consumption and so on, there are many problems to be solved, such as the optimization design of downlink pre-coding, high complexity, imperfect channel state information and other issues. In order to solve these problems, we research the massive MIMO transmission technology for wireless communication systems, and the main works are as follows:1、This paper first reviewed the multi-user MIMO system model, multi-user MIMO pre-coding, coor-dinated multi-point transmission technology and other issues. This chapter focuses on the zero-forcing (ZF) pre-coding, the maximum ratio transmit (MRT) pre-coding, the minimum mean square error (MMSE) pre-coding and analyzes the classification of coordinated multi-point transmission technology, the similarities and differences of coordinated multi-point transmission technology and MIMO, and the coordinated beamform-ing technology and so on. And the matlab simulation results show that at low signal to noise ratio (SNR), MMSE pre-coding outperforms ZF pre-coding; at high SNR, the performance of ZF pre-coding is close to the MMSE pre-coding; with the growth of the antenna number, the gap between ZF, MRT, MMSE pre-coding performance also decreases and so on.2、This paper studied the optimization design of multi-cell coordinated massive MIMO transmission scheme, and the optimization objective was maximizing the minimum signal to interference plus noise ratio (max-min SINR) in the constraint of the single base station power. According to the uplink-downlink duality theory, the original non-convex optimization problem can be changed into the equivalent uplink decoupling optimization problem, and then we proposed instantaneous updating uplink-downlink power algorithm which relies on instantaneous channel state information for the finite system, and the downlink power can be cal-culated directly by the extended coupling matrix without iteration. For the massive MIMO system, in order to solve the problem which power needs to instantaneous update, we proposed the updating uplink-downlink power algorithm which only requires statistical channel state information, and the beam form ing can be calcu-lated directly based on the obtained uplink power and local instantaneous channel state information available by applying the random matrix theory. The simulation results show the feasibility and effectiveness of the proposed algorithms.3、For the inverse problem of the large dimensional matrix existed in the aforementioned optimization design of multi-cell coordinated massive MIMO transmission scheme, according to the thought of the truncat-ed polynomial expansion, the MMSE beamforming consisted of a finite number of terms was given to reduce the complexity of the large dimensional matrix inversion. In order to avoid solving the downlink optimiza- tion problem with coupling variables, we proposed low complexity beamforming algorithm by the equivalent uplink SINR with the MMSE beamforming consisted of the truncated polynomial expansion. By introducing the random function with respect to the variable t, the algorithm can only need the statistical channel state information to solve the truncated polynomial coefficients. The simulation results show that the algorithm can achieve good performance even when the order of the polynomial is very low, and the order of truncated polynomial does not need to change with the increasing dimension. The order of the truncated polynomial should be increased with the growth of the SNR to maintain the gap between the truncated polynomial expan-sion beamforming and the MMSE beamforming, but the order of the truncated polynomial expansion with respect to the system dimension, it is still very small.4、On the basis of the above works, this paper finally studied the optimization design of multi-cell coor-dinated massive MIMO transmission scheme with the channel model which has the channel estimation error for the further consideration of the imperfect channel state information. First, the optimization objective was max-min SINR in the constraint of the single base station power. Then for the finite system, the algorithm which requires instantaneous channel state information to instantaneous update power was given. And for the massive MIMO system, we put forwarded the algorithm which only requires statistical channel state informa-tion to update the uplink-downlink power. To solve the high complexity of the large matrix inversion with the channel model which has the channel estimation error, we proposed the low complexity beamforming algo-rithm which has the controlled order of polynomial and only needs the statistical channel state information to solve truncated polynomial coefficients with the thought of the truncated polynomial expansion and introduc-ing random function. The matlab simulation results show the effectiveness of max-min SINR optimization and low-complexity beamforming algorithm when the channel model has the channel estimation error and even when the channel state information is poor, the algorithm is still able to achieve good performance.