The “Data Challenge” for Fully Digital Phased-Array Radars: Potential of Nonuniform Quantization for Weather Applications

Abstract

Radars are essential for monitoring rapidly intensifying severe weather phenomena, enabling timely warnings and informed decision-making to protect lives and property. The WSR-88D radar network, consisting of more than 160 polarimetric radars across USA, has been recognized as one of the most reliable and highest quality weather radar networks in the world but is reaching end of life in the coming decades. To confront this challenge, phased-array radars (PARs) with their superior capability for rapid and flexible scanning are being considered as a replacement technology. Among PAR architectures, fully digital systems provide advanced capabilities and also are expected to reduce maintenance requirements and operational costs through software reconfigurability. Furthermore, fully digital PARs, which provide access to element-level in-phase (I) and quadrature-phase (Q) data, can perform scans with increased flexibility with the potential for adaptive beamforming. However, they can generate an enormous volume of data, presenting a significant challenge for operational use. To address this “data challenge,” this study examines the impacts of I/Q data quantization, both uniform and nonuniform, on spectral moments and polarimetric variables using data from “Horus,” the first fully digital phased-array weather radar developed at the Advanced Radar Research Center (ARRC) at the University of Oklahoma (OU). The findings demonstrate that nonuniform quantization has the potential to reduce data size while maintaining data quality and dynamic range.