System Level Simulation On Massive MIMO Multi-User Transmission

With the growing popularity of wireless communication and Internet technology, the data rates is in-creasing constantly in mobile communication. Multiple-input multiple-output(MIMO) technology is drawing extensive attention recently in both the industrial and academic communities, as it can exponentially increase system throughput without additional spectrum resources. Massive MIMO technology, which has become a key technology for the fifth generation(5G) wireless communication, makes a clean break on spectrum effi-ciency and the quality of signal transmission by employing a large number of antennas at the base station. By using the characteristics of MIMO channel, transmission scheme in the beam domain can achieve the performance of massive MIMO multi-user transmission system with lower system complexity. In this paper, massive MIMO multi-user transmission in the beam domain is studied by system level simulation, a high performance system level simulation platform is built based on the demands of massive MIMO simulation, and the performance of massive MIMO beam domain transmission scheme is evaluated.Firstly, massive MIMO beam domain multi-user downlink transmission scheme is investigated. To study the characteristics of massive MIMO channel, statistical characteristics transmission correlation matrices is analysised under physical channel model. When antennas at basestation goes to infinity, feature matrix of transmission correlation matrix is DFT matrix and statistical channel information is frequency independent. The beam domain transmission formula and statistical channel information calculation method are obtained by using spatial channel characteristics of massive MIMO. To study the optimal scheduling model, sum rate of multi-user transmission are optimized over beam domain channel. Under the optimal transmission scheme, each user are transmitted over different beams.Subsequently, system level simulation platform of massive MIMO system is investigated. The basic principle and implementation scheme is very similar to the LTE system level simulation platform, the model-ing of the elements in the communication process are discussed. A complete system level simulation platform includes cell topology, user mobility processing model, propagation and fading model, interface between link level and system level simulation, adaptive modulation and coding model and HARQ processing model. The massive MIMO multi-path channel model are deeply investigated. Optimization is applied by using corre-lation between different dimensions of the channel realization. Computing and storage resources are greatly reduced by pre-calculation and data merging on different path. The implementation of massive MIMO multi-path channel model is verified under different scenarios to be correct and efficient. The architecture and implementation method of system level simulation platform are discussed, and a massive MIMO system lev-el simulation platform is implemented with C++ language.Finally, the performance of massive MIMO beam domain transmission scheme is evaluated by system level simulation. Under massive MIMO beam domain transmission, the system level simulation platform is driven by calculation, centralized data essential to the optimization of each module. Calculation complexity of multiuser interference is discussed under both space channel and beam channel simulation. By using the characteristics of beam domain transmission, sparse matrix for precoding can efficiently reduce the complex-ity. Different beam domain scheduling schemes and their complexity are then discussed. Proportional fair factor can be added into the greedy and simplified greedy scheduling scheme. Interference noise level would impact the scheduling results and proper interference level should be set under multi-cell scenario. Simulation results show that when angle spread of channel is small over base station, the performance of massive MIMO transmission is very good as well as the massive MIMO beam domain transmission scheme. When dozens antennas are places at base station, an increase in the number of antennas can still lead to enhanced system performance.