Objective To address the challenges of balancing accuracy and computational efficiency, excessive mesh density under high-frequency incidence conditions, and high computational resource consumption in the RCS calculation of large-scale array antennas, a subarray extrapolation method based on Gordon's integral method is proposed. By utilizing full-wave simulation results of small-scale subarrays, the RCS characteristics of large array antennas can be efficiently and accurately extrapolated.

Method  Considering the differences in the spatial characteristics of array elements, the entire array is divided into several typical subarray units, and the aperture fields of these units are extracted from different regions. Gordon's integral method is introduced to convert the two-dimensional surface integral into a line integral along polygon boundaries, thereby accelerating the near-field-to-far-field transformation, reducing mesh density requirements, and improving the computational efficiency of equivalent surface current calculations. A spatial extrapolation mapping of scattering characteristics is achieved through phase correction factors, and the overall scattering field of the target array is subsequently obtained.

Results  For Vivaldi array antennas and multi-layer stacked patch array antennas, the proposed method is validated through monostatic and bistatic RCS extrapolation experiments, respectively. Compared with full-wave simulation results, the proposed method achieves a threefold increase in computational speed while maintaining a relative root mean square error of less than 5%.

Conclusion  The proposed RCS extrapolation method achieves both high computational accuracy and excellent efficiency, providing an efficient and reliable approach for RCS calculation of large-scale array antennas.