Probe-Corrected Phaseless Near-Field Antenna Measurements

Phaseless near-field antenna measurements consist of measuring only the magnitude of the nearfield of the antenna under test and obtaining the phase by mathematical means, so-called phase retrieval methods. Phaseless measurements are of interest when the phase cannot be reliably measured, and they are necessary when no phase reference of the signal source is available, or the system is entirely incapable of measuring the phase. Examples of the former are high-frequency measurements, where phase measurements may be possible but increasingly unreliable due to probe positioning inaccuracy, cable bending, thermal drift, etc, while the latter case includes e.g. power meter measurements, or drone measurements where a cabled reference signal cannot be passed from the signal source to the receiver.

Despite continued research interest over the years, phaseless near-field measurements remain in an immature state compared to the well-established complex (magnitude and phase) near-field measurement techniques. While several approaches have been suggested and validated to varying degrees, there is no single technique that is generally accepted and valid. This thesis reviews the basic theory and state of the art of the different phaseless techniques for spherical near-field antenna measurements, based on which a selection is made of the so-called two-scans phaseless technique for implementation and testing. Following this, the technique is discussed in depth, addressing several of the open questions for the validation of phaseless measurements, such as probe correction, phase retrieval of the cross-polar component, or uniqueness of solution.

This thesis project focuses on contributions towards a deeper practical understanding of the performance of the two-scans technique in real measurement scenarios. To this end, an extensive investigation of the technique is presented using both experimental and simulated data of different antennas, including a real and electrically large and challenging antenna such as the DTU-ESA VAST12 reflector. The investigation is in the form of a systematic study of the influence of measurement parameters on the performance of the phaseless algorithm, such as the radius of the measurement spheres, the sampling of the near-field magnitudes, the choice of probe antennas, and the initial phase guess. The novel contributions introduced in this thesis include the demonstration of techniques such as full probe correction and cross-polar component measurements, which are standard techniques in complex near-field measurements but remain very scarcely, if at all, studied in a phaseless context, and the demonstration of the two-scans technique in realistic measurement scenarios using experimentally acquired data when possible. In addition to this, a novel phaseless technique is presented and demonstrated for the first time. It is one of the three possible implementations of the two-scans technique and based on two differently dis placed positions of the antenna in the measurement coordinate system, rather than two measurement spheres.

The results presented show the potential of the spherical phaseless two-scans technique for the accurate characterization of challenging antenna patterns in realistic measurement scenarios based on the conditions of the DTU-ESA Spherical Near-Field Antenna Test Facility and the upcoming DTU Electromagnetic Test Centre.