Low-Cost Massive MIMO: Pilot Length and ADC Resolution

Abstract

When designing an energy efficient massive multiple-input multiple-output (MIMO) system where each receiver antenna is equipped with a low-resolution analog-to-digital converter (ADC), the number of base station (BS) antennas and quantization bits are generally two mutually conflicting system parameters. In this paper, we investigate the joint optimization of the number of BS antennas and ADC resolution in quantized massive MIMO systems, assuming imperfect channel state information (CSI). A tractable approximate expression for the uplink sum spectral efficiency (SE) using maximal ratio combining (MRC) receivers is derived, based on which the pilot length which maximizes the sum SE is put forward. Considering the effect of ADCs, a realistic model of total power consumption is given subsequently. Capitalizing on it, we formulate the optimization problem of selecting the number of BS antennas and ADC resolution to maximize the sum SE under a total power consumption constraint. Our results show that more pilot symbols should be assigned for massive MIMO systems with low-resolution ADCs, especially for the receivers with one-bit quantizers. Moreover, the results show the trade-off between the number of BS antennas and quantization bits. Numerical results suggest that there exists an optimal ADC resolution in massive MIMO systems, while lower quantization bits may cause a substantial degradation of the SE performance and higher one will consume more power.