摘要： The primary goal of this paper is the optimization of data transmissions and connections between 5G base stations (BSs) as well as the improvement of access technologies and transmission methods in consideration of massive multi-input multi-output, a key technology in 5G networks. In order to reach an access technology supported by multiple BSs and small cells, we use 5G millimeter wave (mmWave), due to its high directivity and sensitivity to blockage, to enhance the connection system. In the simulation, we will consider extremely high-frequency band and small angle of mmWave, and arrange obstructions in the environment in view of high attenuation characteristics in mmWave signals. After a wave beam penetrates through a wall, the power of the wave sharply decreases. For reduction of energy consumption, the wave therefore will select an mmWave BS with poor signal quality but without blockage to transmit data. Because the number of macro-cells will affect the communication quality and the computational complexity, this paper especially focuses on three factors of a network: delay, capacity, and path loss, and purposes to figure out the most energy-efficient BS densities for 5G-based green communications.