Spectral Efficiency Maximization of Multiuser Massive MIMO With a Finite Dimensional Channel

In this paper, we develop a closed-form lower bound for the achievable uplink rate of users in a single-cell massive multiple-input multiple-output (MIMO) system using zero-forcing (ZF) receiver. The base station (BS) is equipped with a large number of uniformly and linearly spaced antennas and serves single-antenna users. It estimates channel state information (CSI) using uplink pilot symbols. The channel model is also supposed to have a finite dimension. We then define the spectral efficiency as the sum of derived uplink rates. Through power allocation between data and pilot symbols, we further optimize this approximate expression of spectral efficiency. Simulation results are presented to evaluate the approximation of spectral efficiency and to show the advantage of our power allocation scheme in comparison to equal power allocation. The results also demonstrate that in low signal-to-noise ratios, more power should be allocated to the pilot symbols.