Channel Estimation Performance Analysis of Massive MIMO IoT Systems With Ricean Fading

This article analyzes the channel estimation performance of massive multiple-input-multiple-output (MIMO) Internet-of-Things (IoT) systems with Ricean fading. First, by utilizing the least squares (LSs) and minimum mean squared error (MMSE) estimation methods, we consider the relative channel estimation error (RCEE) between the IoT device and base-station, and provide the approximations of the expectation of RCEE ( Exp
_rcee
). Then, it is found that when the number of antennas M becomes infinite, pilot contamination (PC) exists in both cases. However, for MMSE case, Exp
_rcee
scales down by the inverse of Ricean K-factor, and hence PC phenomenon disappears with a large Ricean K-factor. Moreover, as M→ ∞, the power scaling laws show that the pilot sequence power can be scaled down proportionally to 1/M
^α
( α > 0) with the MMSE case, where the performance is determined only by the Ricean K-factor. Next, the channel hardening and favorable propagation effects are examined via analyzing the approximations of the variance of RCEE ( Var
_rcee
). Analysis implies that Var
_rcee
decreases by 1/M when M→ ∞. For a large Ricean K-factor, Var
_rcee
approaches a nonzero constant for the LS case and scales down by the inverse of the square of Ricean K-factor for the MMSE case. Finally, all results are verified via Monte Carlo simulations.