SNR-Aided Accurate Phase Estimation in 1-Bit Software-Defined Receiver

Abstract

This work investigates the high-accuracy phase discriminator in 1-bit software-defined receiver (SDR) for three signal-to-noise ratio (SNR) ranges: low SNR, high SNR and moderate SNR. Unlike traditional approaches, our approach first distinguishes which SNR range an application falls into, and this SNR information is then utilized to select a proper phase discriminator to achieve high accuracy. For low-SNR applications, traditional arctangent phase discriminator (APD) is adopted and the analysis of asymptotic performance is derived. For high-SNR applications, the noise-balanced digital phase discriminator (NB-DPD) is adopted to improve accuracy [1]. Between them, for moderate-SNR applications, a novel SNR-aided phase discriminator (SNRaPD) is proposed. The analytical and simulation results verify the superiority of the proposed approach for applications in different SNR ranges. The determination of moderate SNR range is critical and discussed finally.