Analysis of One-Bit Quantized ZF Precoding for the Multiuser Massive MIMO Downlink

We present a mathematical analysis of linear precoders for downlink multiuser massive MIMO. In the scenario we consider, one-bit digital-to-analog converters are employed at the basestation antennas for mitigating power consumption and hardware complexity. Using the Bussgang theorem, a probability-of-error analysis for a general precoder is presented. The special case of zeroforcing (ZF) precoders is considered and simplified error expressions are derived for asymptotic scenarios. Our analysis illustrates that the probability of error for the quantized ZF precoder depends primarily on the ratio of the number of antennas to the number of users. Our simulations also show that the quantized ZF precoder outperforms the high complexity ML encoder for low-to medium-SNRs