Link Adaptive Method For Massive MIMO BDMA Systems

With the popularization of mobile communication equipments and the explosive growth of multi-media data, the fifth generation mobile communication network (5G) after 2020 must support the high-speed packet data transmission up to several hundreds of megabytes or even gigabytes per second. With the high-speed development of wireless communication technology and user’s expectations for the higher data rates and the better transmission quality, massive MIMO technology has drawn considerable attention, because of it-s substantial improvements in spectral efficiency and link reliability. Massive MIMO, which implements large-scale antenna array at the base station(BS), can provide more spatial resources to serve multiple user equipments(UEs) simultaneously. In this thesis, the channel spatial characteristics of massive MIMO system is considered, and the Beam Division Multiple Access(BDMA) transmission scheme is investigated on the basis of beam domain channel model. Specifically, we investigate the link adaptation method of the massive MIMO BDMA systems.Firstly, link adaptation performances under different receiving algorithms for massive MIMO BDMA system are investigated. Considering the spatial characteristics of massive MIMO beam domain channel, us-er scheduling, which bases on the maximum of system sum-rate, equivalently decomposes the MU-MIMO channel into multiple single-user MIMO channels. As a result, multiple users simultaneously communicate with the BS via non-overlapping beam set while reducing the inter-user interference. On this basis, an appro-priate channel quality indicator(CQI) is selected to achieve a higher spectral efficiency while guaranteeing a given Codeword Error Ratio(CWER) target. For the downlink single-cell multi-user scenario, we respectively derive the expression of Signal-to-Interference-plus-Noise-Ratio(SINR) under two different receiving algo-rithms. Based on the effective SNR mapping, each user obtains an equivalent SNR and determine the most effective CQI to match the current channel state. Furthermore, an improved and simple method is proposed to update CQI parameter. Consequently, the BS chooses suitable Modulation and Coding Scheme(MCS) for each codeword on each transmission layer according to the CQI received. Numerical evaluation results show that link adaptation method can obtain a high system spectral efficiency with a given target CWER con-straint in massive MIMO BDMA systems. Moreover, the improved method can further improve the spectral efficiency of the massive MIMO systems.Subsequently, we focus on the link adaptation method based on mutual information(MI) for massive MIMO BDMA systems. Under the premise of ensuring the CWER below 0.1, the proposed adaptive method chooses the CQI value according to the MI between the finite discrete constellation input and the correspond-ing output over complex-valued vector channels. Accordingly, the MCS is dynamically adjusted to take full advantage of the wireless channel. On the basis of beam domain channel model, we consider the downlink transmission of the massive MIMO system in the multi-user scenario, and derive the MI formula of the in-terference free MIMO link. Furthermore, we study the lower bound formula to reduce the computational complexity and extend the calculation of MI into MIMO link with interference. In the proposed method, each user is required to calculate the average MI on all resource blocks and select an appropriate CQI by comparison with different mutual information thresholds. Simulation results show that the proposed adap-tive method combined with iterative receiver can obtain a high spectral efficiency while guaranteeing a given CWER target.