Parallax Errors and Corrections for Dual-Antenna Millimeter-Wave Cloud Radars

Abstract

Dual-antenna radar designs avoid using a transmit/receive switch. In order to measure radar reflectivity accurately and to avoid a general decrease in system sensitivity, these systems require precise alignment of their high-gain/narrow-beamwidth antennas, which is difficult. Given precisely aligned antennas, a parallax correction to account for antenna beam overlap, which is range-dependent, must be used with the correct alignment information to produce accurate reflectivities. Calculations show that dual-antenna parallax errors are extremely sensitive to the alignment of the two antennas, especially for the current generation of W-band radars, which tend to use 0.91- and 1.21-m Cassegrain antennas with half-power beamwidths of typically ≤0.25°. For example, the minimum detectable reflectivity of a W-band radar system may be degraded by more than an order of magnitude for alignment errors on the order of the antenna half-power beamwidth. Moreover, parallax errors are essentially independent of range at cirrus altitudes, and it is not possible to separate parallax effects from offsets in calibration at these far ranges. Observations from a field experiment that include both single- and dual-antenna radar measurements are used to demonstrate these points. Alignment problems have led to the abandonment of dual-antenna pulsed W-band systems in the cloud remote sensing community, and the current generation of millimeter-wave frequency-modulated continuous wave systems must properly take these problems into consideration.