Receiver Design for OFDM Schemes With Low-Resolution ADCs

Abstract

Is is widely known that the power consumption of Analog-to-Digital Converters (ADCs) is strongly related with the number of bits of resolution. Using smaller resolutions can greatly reduce the power consumption of the RX frontend. However, the use of low-resolution ADCs introduces significant nonlinear distortion, especially when Orthogonal Frequency Division Multiplexing (OFDM) signals are employed, which might severely degrade performance. In this work, we consider nonlinear effects at the receiver side, associated with quantization and saturation of low-resolution ADCs, and we propose a Generalized Approximate Message Passing (GAMP) receiver that is specially designed to cope with nonlinear effects at the OFDM receiver. We show that our receiver can significantly mitigate the distortion that arises from low-resolution ADCs, allowing larger M-ary Quadrature Amplitude Modulation (M-QAM) constellations to be employed. The proposed receiver is shown to be robust to strong channel fading effects, as well as errors during the channel estimation process.