Belief Propagation Detection For Large-Scale MIMO Systems With Low-Resolution ADCs

We derive the belief propagation (BP) detector for large-scale multiple-input multiple-out communication systems where the low-resolution analog-to-digital converters (ADCs) are used to save the power and hardware costs. By modeling the quantization noise as an additive noise element, we derive a new expression for the extrinsic information passed from the observation node to the symbol node on the Tanner graph. Furthermore, we study the performance of the BP-based MIMO detector under different low-resolution ADCs and multiple-input multiple-output (MIMO) configurations. The study results show that one can achieve almost the same performance of high-resolution MIMO systems with the 5-bit ADC in the worst scenario where the number of transmit antennas $M$ and the number of receive antennas $N$ are equal. In the favorable situation that N > M, for example $M$ = 10 and N = 30, the 3-bit ADC system has the performance that approaches the performance of the high-resolution ADC system.