EXPLORATION OF THE FEASIBILITY OF ADAPTIVE SPHERICAL NEAR-FIELD ANTENNA MEASUREMENTS – PART II Vincent Beaulé, Derek McNamara EECS, University of Ottawa, 800 King Edward Avenue Ottawa, Ontario K1N 6N5, Canada Daniël Janse van Rensburg Nearfield Systems Inc. 19730 Magellan Drive Torrance, CA 90502-1104, USA Leili Shafai DFL, Canadian Space Agency 3701 Carling Avenue Ottawa, Ontario K2H 8S2, Canada ABSTRACT The use of pre-conditioning interpolation schemes, as a possible means of enhancing the performance of previously introduced adaptive acquisition algorithms for spherical near-field (SNF) test time reduction, is evaluated. Investigations have been carried out to establish whether the adaptive SNF approach is suited to test engineering practice are reported. The pre-conditioning method involving the acquisition of two orthogonal polar cuts on the near- field sphere and the separate linear interpolation of two complex spherical components of the NF data is shown to be the preferred scheme. This method is evaluated for three different antennas using specific acquisition rules, and decision functions related to directivity, amplitude error, and side lobe level. 1. Introduction An adaptive acquisition algorithm for spherical near- field (SNF) test time reduction was proposed in [1]. It is based on the premise that near-field to far-field (NF-FF) transformation time is small compared to data acquisition time, so that such computations can be done repeatedly while data is being acquired. This allows the use of the transformed FF data to continuously compute and monitor pre-defined decision functions (formed from the antenna-under-test (AUT) specifications) while data is being acquired. The idea is to effectively allow the probe to follow a directed path under control of an acquisition rule, so that the sampled NF data points constitute an acquisition map on the measurement sphere. SNF data acquisition is then terminated based on decision function values, the intention being that the smallest amount of data needed to ensure accurate determination of the AUT performance measures will be acquired. The feasibility of the approach was demonstrated using actual NF data for several antennas, using various decision functions and acquisition rules. This paper is a continuation of the work reported in [1] and will report on tests that have been carried out to establish suitability of the approach to test engineering practice. In addition, a pre-conditioning concept is introduced that obviates the need to assume zero value NF start values. Different pre-conditioning interpolation schemes will be compared as to their influence on the adaptive SNF process. 2. Interpolation and Pre-Conditioning In [1] the start values of all NF data points were assumed to be zero, implying that fields on the measurement sphere were zero everywhere. These values were then replaced by actual measured data as the acquisition proceeded. At any stage only a portion of the fields on the measurement sphere would have been sampled. This introduced a significant and ever changing truncation effect. In order to soften this transition (that is, avoid the truncation effect) an interpolation method was implemented for adaptive planar near-field (PNF) systems in [2], as opposed to adaptive SNF, and referred to as pre-conditioning. Here, this pre-conditioning concept is applied to the SNF case. The idea is to use an approximation of the NF data as a starting point instead of an empty NF array, thus avoiding the use of zero field values entirely. Two or