Application of Clustering Technology and Gerchberg-Papoulis Algorithm for Spherical Near Field Antenna Measurements

This paper presents an optimized interpolation and extrapolation scheme for spherical near-field antenna testing. The method starts with a small amount of initial sampling data, and uses the optimization method combining partition cluster classification and Gerchberg-Papoulis algorithm to interpolate and extrapolate randomly sampled data. This optimization scheme effectively reduces the number of samples, and achieves near-far-field conversion results with the same accuracy as triple oversampling. The results demonstrate the feasibility of the proposed scheme.