Performance Analysis of Channel Mismatch in Cell-Free Massive MIMO IoT Networks With Ricean Fading

Channel mismatch, resulting from channel nonreciprocity (NRC) and channel aging, can make the estimated channel outdated when used for precoding and detection. In this article, we investigate the impact of channel mismatch for cell-free (CF) massive multiple-input-multiple-output (MIMO) systems in the context of Internet-of-Things (IoT) networks. Assuming that the wireless propagation channel among access points (APs) and IoT sensors (ISs) follows a Ricean fading model, we derive tight closed-form expressions for downlink achievable spectral efficiency (SE), accounting for both NRC and channel aging. Then, we analyze how channel NRC and channel aging affect the downlink SE. To ensure a uniform quality of service for each IS, we design a max-min fairness (MMF) power control scheme to maximize the minimum IS SE. Moreover, we propose a channel prediction algorithm based on the Wiener filter principle to alleviate the effect of channel aging by using the predicted channel to construct a downlink precoding vector. Simulation results show the joint impact of channel NRC and channel aging on the downlink achievable SE. Moreover, we also validate the effectiveness of the proposed MMF power control scheme and channel prediction algorithm.