Scalable Cell-Free Massive MIMO Systems: Impact of Hardware Impairments

Scalable cell-free CF (SCF) massive multiple-input-multiple-output (mMIMO) systems is a promising
technology to cover the demands for higher data rates and increasing number of users in fifth generation
(5G) networks and beyond. According to this concept, a large number of distributed access points (APs)
communicates with the users in the network by means of joint coherent transmission while facing the
main challenges against standard CF mMIMO systems being their high fronthaul load and computational
complexity. Given that the cost-efficient deployment of such large networks requires low-cost transceivers
being prone to unavoidable hardware imperfections, in this work, we focus on their impact on the
advantageous SCF mMIMO systems by means of a general model accounting for both additive and
multiplicative hardware impairments (HWIs). Notably, the scalability, depending on the time-variant
characteristics of the network, is clearly affected by means of HWIs being time-varying. There is no
other work in the literature studying the phase noise (PN) in CF mMIMO systems or in general any
HWIs in SCF mMIMO systems. Hence, we derive upper and lower bounds on the uplink capacity
accounting for HWIs. Especially, the lower bound is derived in closed-form by means of the theory of
deterministic equivalents (DEs) and after obtaining the optimal hardware-aware (HA) partial minimum
mean-squared error (PMMSE) combiner. Among the interesting findings, we observe that separate local
oscillators (SLOs) outperform a common LO (CLO) architecture, and the additive transmit distortion
degrades more the performance than the receive distortion.