Twice optimised near-field scanning system for antenna characterisation

The near-field (NF) acquisition time depends on the sampling strategy and on the measurement hardware. Optimised sampling based on singular value optimisation dramatically reduces the number of measurement points and the scanning path length compared to conventional samplings, without impairing the far-field reconstruction. Furthermore, an optimised control approach, tracking a sufficiently smooth trajectory allows better performance in terms of velocity and precision, significantly reducing the scanning time. We show that the combined use of an optimised sampling and an optimised controller drastically reduces the measurement time. The sampling strategy provides a raster grid of optimally selected rows hosting samples optimally arranged according to a row-wise non-uniform distribution. The optimised control drives the NF scanner and the vector network analyser to acquire the data along each row by a continuous probe movement with a variable velocity, accelerating and decelerating between two consecutive sampling points. The sparsity of the NF locations enables a significant increase in the average speed between two consecutive points. The controller is realised with a proportional-integral-derivative law with a compensation signal and a filter downstream. The controller does not demand complex hardware and can be realised in an extremely simple, cheap and off-the-shelf solution.