Spectral Efficiency Analysis for Cell-Free Massive MIMO Systems With Low-Resolution ADCs Under Imperfect CSI

The main challenge for the practical deployment of cell-free massive multi-input-multi-output (CF-mMIMO) systems under imperfect channel state information (CSI) is the high hardware cost caused by utilizing high-resolution analog-to-digital converters (ADCs). To tackle this challenge, an effective solution is to equip the systems with low-resolution ADCs at all access points (APs). In this article, the uplink pilot contamination mitigation method is proposed in a CF-mMIMO system with low-resolution ADCs under imperfect CSI. More specifically, the closed-form expressions for both uplink and downlink spectral efficiency (SE) are first derived from the additive quantization noise model (AQNM) and the maximum ratio combining (MRC) receiver filter under imperfect CSI. Then they are utilized to analyze the effects of the APs number, the user’s transmit power, and ADC resolution on the SE. Also, in order to maximize the SE while reducing the hardware cost, an efficient pilot assignment algorithm with graph coloring under low-resolution ADCs is presented that mitigates the pilot contamination. Finally, we present numerical results showing that the achievable uplink SE of the 5-bit low-resolution ADCs is close to that of the perfect-resolution ADCs, and the low-resolution quantization losses can be compensated by increasing the antenna number per AP. Thus, an optimal balance between the hardware cost and SE can be achieved by equipping the APs with 5-bit low-resolution ADCs. Furthermore, compared with the existing pilot assignment algorithms, the proposed algorithm mitigates better the pilot contamination and significantly improves the 95%-likely per-user achievable uplink SE.