MIMO communications with low resolution ADCs

Abstract

The wide bandwidths and large antenna arrays in future communication systems impose big challenges for the hardware design of the receiver, which has to efficiently process multiple signals from antennas at a much faster pace. The analog-to-digital converter (ADC) is one of the bottlenecks. High speed, high resolution (e.g., 8-12 bits) ADCs are either not available, or are too costly and power-hungry for portable devices. To overcome these challenges, a possible solution is to simply employ ADCs with much lower resolution (e.g., 1-4 bits) which are much easier to implement and cheaper. The use of few- and especially 1-bit ADCs radically changes both the theory and practice of communication. The nonlinear and coarse quantization makes the channel capacity analysis difficult and has adverse effects on the channel estimation, symbol detection, etc.

In this dissertation, we analyzed the MIMO channel with low resolution quantization. We first derived upper bounds on quantized MIMO channel capacity. By assuming the channel is known at the transmitter, we found precoding methods approaching the upper bound. Second, we devised a channel estimation algorithm with low resolution ADCs. The sparsity of the millimeter wave channel is exploited in our algorithm. Third, by assuming the channel is perfectly estimated at the receiver, we designed new limited feedback methods in the quantized MIMO channel. At last, we proposed and analyzed an energy-efficient receiver architecture with hybrid precoding and low resolution ADCs.