User Power Control And Pilot Scheduling In Massive MIMO Systems

With the popularity of Smart phones and terminals, communication techniques gain iterative refinements. In the mature 4G age, the user data rates have been improved greatly. Nowadays people lay emphasis on not only the QoS but also the QoE. It can’t be denied that the earth environment is getting worse and the greenhouse effect is being more serious. The concept of green communication is then brought up and gains a widely acceptance. The above not only sets higher requirements on the next generation of communication but also points out the new research direction.The conventional MIMO technique is being utilized well in 4G age. However the LTE standards stipulate that the BS can be equipped with at most 8 antennas. Recently, massive MIMO becomes the research hotspot of wireless communication because of its advantages in system capacity, spectrum efficiency, reliability and energy efficiency. In massive MIMO systems, the number of antennas at the BS increases greatly and is much larger than 8. Research indicates that when the number of antennas at the BS approaches infinity, the effect of small-scale fading as well as the additive noise disappears. Adopting simple signal processing techniques can gain the quasi-optimal performances. When satisfying the QoS requirements, the user power can be scaled down according to a ratio, which is meaningful to the green communication.Researches on massive MIMO are mainly based on the TDD protocol. Firstly users send uplink pilot sequences and the BS estimates the user channels. Then the BS gets the downlink channels using the channel reciprocity. However, with the increasing of number of users and the time-variation of channels, the number of orthogonal pilots is limited and the reuse of pilots in different cells is inevitable. The interference caused by reusing pilots don’t decrease with the increasing of number of antennas at BS and this leads to the bottleneck of massive MIMO-pilot contamination.This thesis mainly researches on the pilot contamination problem in massive MIMO systems and the main contributions are listed as follows:We consider the uplink communication process of massive MIMO. Users firstly transmit the pilots and the BS estimates the channels. The BS then processes the data transmitted by users using the estimated channel. We aim to analyze the effect of power control on the user performances, so we assume that the power of users are different from each other. The SINR of users in the uplink is derived. Based on the SINR, we analyze the pilot contamination and the influence of user power. Then a user power control strategy is proposed and simulation results are presented.We study the downlink communication of massive MIMO. The BS of each cell forms the precoding matrix based on the estimated channels, precodes the data and then transmits to the aiming users. The users use the same transmit power and the antennas of BS also use the uniform transmit power. The MSE of channel estimation at the BS is derived and the effect of pilot contamination on the channel estimation is analyzed. Based on the MSE, a user pilot scheduling algorithm is proposed which preferentially decreases the effect of pilot contamination on the poorer users. Simulation results indicates that the algorithm improves the performances of users as well as the system.